Re: [Ubuntu-phone] Landing process, restoring "trunk" for development

[Michi Henning]

On 9 Apr 2014, at 15:08 , James Henstridge  
wrote:

> 
> Things would be a lot simpler if we could use the trunk branch as we
> did before and have the CI Train process manage a separate production
> branch.

I strongly agree with this. In addition, it is awfully easy to submit an MR to 
the wrong branch by mistake because, in Launchpad, trunk is the default merge 
target. If I forget to click the radio button for the devel branch, I've 
accidentally pushed to trunk. Fixing that mistake isn't the end of the world, 
but it takes time to notice the mistake, more time to rectify it, and it 
generates annoying email noise for everyone.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] Nexus 4 root partition

[Michi Henning]

Hi guys,

I was trying to build on the phone today (unity-scopes-api) and found that the 
build dependencies for it won't fit on the root partition.

Is there anything I can do (easily) to get more space on the root partition so 
I can install all the packages I need?

Thanks,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Nexus 4 root partition

[Michi Henning]

Thanks for that Saviq, will try this!

Michi.


On 8 May 2014, at 18:46 , Michał Sawicz  wrote:

> On 08.05.2014 10:12, Michi Henning wrote:
>> I was trying to build on the phone today (unity-scopes-api) and found that 
>> the build dependencies for it won't fit on the root partition.
>> 
>> Is there anything I can do (easily) to get more space on the root partition 
>> so I can install all the packages I need?
> 
> Cross-build, really :)
> 
> There's a few tips'n'tricks on the wiki below to speed things up even
> more (like prepopulated chroots and dirty builds).
> 
> Hit me up if you have issues with any of those.
> 
> https://wiki.ubuntu.com/SimpleSbuild
> https://wiki.ubuntu.com/CrossBuilding
> -- 
> Michał (Saviq) Sawicz 
> Canonical Services Ltd.
> 


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Nexus 4 root partition

[Michi Henning]

Thanks heaps for that, will give that a shot tomorrow!

Michi.


On 8 May 2014, at 19:47 , Michał Sawicz  wrote:

> On 08.05.2014 11:08, Michał Sawicz wrote:
>> There's issues with your build-dependencies that break cross-building,
>> I'll try and cook up an MP to fix those.
> 
> Here's what you need:
> 
> https://code.launchpad.net/~saviq/unity-scopes-api/fix-cross-build/+merge/218768
> 
> Unfortunately ltt packages need Multi-Arch fixing, so we'll have to wait
> for that - xnox promised to have it done tomorrow at the latest :)
> 
> -- 
> Michał (Saviq) Sawicz 
> Canonical Services Ltd.
> 


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Autopilot tutorial videos!

[Michi Henning]

> 
> I'm keen to hear any feedback you might have - what areas would you like to 
> see covered? Was there something that you didn't understand in this video? 
> any other comments? Feel free to reply to me either on, or off-list.

That's really, really nice, congratulations! I love the easy-going 
conversational style.

My main wish would be to have more, with actual code examples and the like.

Cheers,

Michi.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Unity API status update

[Michi Henning]

On 16/06/14 15:29, Olivier Tilloy wrote:


Do I understand it correctly that it’s a generic and transparent image 
cache?
If so, please consider including it in the UITK, to benefit all 
applications that load images/icons from the network.


The cache implementation is independent of the network. It's just a 
simple class that provides a persistent map of  pairs.
It maintains the cache in LRU order and, once the cache is full, 
automatically trims the cache down to make room for the newest entry. It 
also supports optional explicit expiry times and dynamic resizing.


It's not particularly fast because it uses sqlite under the covers. 
That's because we need guaranteed recovery if something goes wrong, such 
as a partial write if the battery dying at the wrong moment (which could 
potentially brick the phone). But the cache is still plenty fast enough 
compared to waiting for an HTTP request to complete for each image.


The cache is currently MR'd here:

https://code.launchpad.net/~michihenning/unity-api/PersistentCache

The plan is to have the shell hook the cache up with the network. (I 
wanted a stand-alone cache class so it can be used in other contexts.)


Michi.

--
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] vivid+stable-phone-overlay on desktops

[Michi Henning]

For what it's worth, in case someone else stumbles into this:

apt-get install --reinstall xserver-xorg

This fixed it. (Thanks dobey!)

Michi.


On 14 Aug 2015, at 6:53 , Michi Henning  wrote:

> Thanks for your help guys!
> 
> There are no obvious errors in Xorg.0.log, but in lightdm.log, I find:
> 
> DisplayServer x-0: Can't launch Xserver X -core, not found in path
> DisplayServer x-0: Xserver stopped
> 
> Any ideas?
> 
> Thanks,
> 
> Michi.
> 
> 
> On 14 Aug 2015, at 0:05 , Łukasz 'sil2100' Zemczak 
>  wrote:
> 
>> Hey,
>> 
>> I uploaded the intel, ati and nvidia drivers to the overlay PPA when
>> working on getting this resolved. Those were the three drivers that
>> seemed to be affected by the original bug.
>> 
>> Cheers,
>> 
>> W dniu 13.08.2015 o 15:15, Albert Astals Cid pisze:
>>> Ah then probably you're not using the intel driver, i'd say have a
>>> look at the /var/log/Xorg* and see which driver is failing, maybe we
>>> need rebuilds of more drivers, myself i'm using
>>> 
>>> ii  xserver-xorg-core
>>> 2:1.17.2-1ubuntu0.1~overlay1  amd64
>>> Xorg X server - core server
>>> ii  xserver-xorg-dev
>>> 2:1.17.2-1ubuntu0.1~overlay1  amd64
>>> Xorg X server - development files
>>> ii  xserver-xorg-video-intel
>>> 2:2.99.917-1~exp1ubuntu2.3~overlay1   amd64
>>> X.Org X server -- Intel i8xx, i9xx display driver
>>> 
>>> 
>>> On Thu, Aug 13, 2015 at 2:19 PM, Michi Henning
>>>  wrote:
>>>> 
>>>> On 13 Aug 2015, at 21:05 , Albert Astals Cid  
>>>> wrote:
>>>> 
>>>>> If you update && dist-upgrade you should get the new intel driver from
>>>>> the stable-phone-overlay that fixes it.
>>>> 
>>>> Thanks! I tried that, but no joy. I still get a black screen with a 
>>>> blinking cursor at the top.
>>>> 
>>>> I've tried both with both versions mentioned earlier, same result.
>>>> 
>>>> What version should I be seeing for things to work?
>>>> 
>>>> All this is in a VMWare Fusion VM. I've rebuilt Vmware Tools with both 
>>>> driver versions installed, but that didn't make any difference.
>>>> 
>>>> Is anyone else around who has Vivid with the overlay working in VMWare 
>>>> Fusion? I'm running the latest version (7.1.2).
>>>> 
>>>> Thanks,
>>>> 
>>>> Michi.
>>>> 
>>>> 
>>>>> 
>>>>> 
>>>>> On Thu, Aug 13, 2015 at 11:35 AM, Michi Henning
>>>>>  wrote:
>>>>>> After doing an upgrade and installing xserver-xorg-core=2:1.17.1-0ubuntu3
>>>>>> again, I still don't get a desktop.
>>>>>> The same thing happens with 
>>>>>> xserver-xort-core=2:1.17.2-1ubuntu0.1~overlay1.
>>>>>> 
>>>>>> Does anyone have any suggestions for how I can my desktop back?
>>>>>> 
>>>>>> Thanks,
>>>>>> 
>>>>>> Michi.
>>>>>> 
>>>>>> 
>>>>>> On 6 Aug 2015, at 8:25 , Kevin Gunn  wrote:
>>>>>> 
>>>>>> Hi all -
>>>>>> So I felt like this needed to be shared just in case.
>>>>>> Some folks have added the ppa for stable-phone-overlay to their vivid
>>>>>> desktop to assist their efforts with development activities for the 
>>>>>> phone.
>>>>>> 
>>>>>> This has worked with no issues in recent history, however today it was
>>>>>> discovered by one of our team mates that if you're using
>>>>>> vivid+stable-phone-overlay on your desktop and you update(today) on intel
>>>>>> you may not get graphics but...
>>>>>> 
>>>>>> "Failed to load /usr/lib/xorg/modules/drivers/intel_drv.so:
>>>>>> /usr/lib/xorg/modules/drivers/intel_drv.so:  undefined symbol: xorgMir
>>>>>> in /var/log/Xorg.*"
>>>>>> 
>>>>>> To recover your graphics
>>>>>> sudo apt-get install xserver-xorg-core=2:1.17.1-0ubuntu3
>>>>>> 
>>>>>> I'm on wily so I can't speak to the experience or specifics, but I 
>>>>>> wanted to
>>>>>> share just in case this might be effecting so silent but frustrated 
>>>>>> folks.
>>>>>> br,kg
>>>>>> --
>>>>>> Mailing list: https://launchpad.net/~ubuntu-phone
>>>>>> Post to : ubuntu-phone@lists.launchpad.net
>>>>>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>>>>>> More help   : https://help.launchpad.net/ListHelp
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> --
>>>>>> Mailing list: https://launchpad.net/~ubuntu-phone
>>>>>> Post to : ubuntu-phone@lists.launchpad.net
>>>>>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>>>>>> More help   : https://help.launchpad.net/ListHelp
>>>>>> 
>>>> 
>>> 
>> 
>> 
>> -- 
>> Łukasz 'sil2100' Zemczak
>> Foundations Team
>> lukasz.zemc...@canonical.com
>> www.canonical.com
>> 
>> -- 
>> Mailing list: https://launchpad.net/~ubuntu-phone
>> Post to : ubuntu-phone@lists.launchpad.net
>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>> More help   : https://help.launchpad.net/ListHelp
> 


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Future of Scopes

[Michi Henning]

Scopes, by design, are information providers, and interaction with them is 
deliberately kept to a minimum. In return, we get scopes that are stateless, 
take up minimal resources, don't need access to a display surface, and start up 
in a fraction of a second.

The way to make scopes interactive to bundle them with an app. The scope 
surfaces content and, when more sophisticated interactions are required, the 
scope can notify the app, which then takes over for the fancy stuff.

Cheers,

Michi.


On 31 Aug 2015, at 0:28 , Krzysztof Tataradziński  wrote:

> Hello,
> I'm thinking does Canonical has got any plan for make scope more 
> 'interactive'?
> Lets see NearBy - we can look at places from Yelp's database and eventually 
> see it on HERE Maps.
> First - is there any possibility to change default map app? I mean I would 
> love if we click on map button in Nearby scope -> Yelp we can select to show 
> that place in GPS Navigation or OSMScout. Is that the Scope problem of maps 
> developers?
> Second - why NearBy shows places from Yelp only? Why not to use Google 
> Places? Great solution would be to add that source and allow users to turn it 
> on/off.
> Third - for me, most of scopes look nice, present data clearly, but still 
> nothing more and only similar to RSS feeds, not very interactive at all. In 
> example again Nearby -> Yelp - why scope don't allow to rate place, add new 
> one, etc.? Only opens website. Is there limitation of Yelp itself or 
> technology used in Scope or it's just a matter of time and resource and 
> someone will make that possible?
> 
> Best regards,
> Krzysztof Tataradziński
> https://launchpad.net/~ktatar156
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Future of Scopes

[Michi Henning]

> I'm not sure if what you described is like that ^, but the idea is that the
> aggregator seems to know the order it wants to put things, and it knows what 
> is
> up to date and out of date. Keep the order and replace out of date with up to
> date in place when ready.

This doesn't deal with the "jumping icons" problem though. The general issue 
with in-place updates is that the cardinality of a query result set can change 
as well as individual items returned by that query.

The accidental activation issues this causes are real. I've been annoyed in the 
past by other applications that fill a list with actions/items and me 
activating the wrong thing by mistake because the list updates just before I 
click.

Caching results from queries can be problematic too because, in general, 
queries and results do not have any reliable identity. Just because the query 
is the same does not necessarily mean that, when the query is run a second 
time, it will produce the same results as it did the first time. And, for 
results, the issue generally is that two different URLs can identify the same 
resource, and that the same URL can identify different resources over time, or 
lose its resource altogether.

Usually, the scope implementation knows a little more about the semantics of 
the results it is dealing with, so that is typically the right place to do 
caching. But, even a scope may not understand much about the results it 
produces, if the scope is a simple front end that scrapes information from some 
back-end source. And, for aggregators, the situation is worse, unless the 
aggregator always aggregates from the same fixed set of well-understood child 
scopes.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] vivid+stable-phone-overlay on desktops

[Michi Henning]

After the latest apt-get upgrade, I again have a VM that won't boot. 
Unfortunately, it seems to fail very early during the boot. I can't get a VT, 
and there are no error messages, just an underscore cursor at the top left.

At any rate, if the vmware driver is missing in the PPA, it would be really 
great to add it!

Cheers,

Michi.


On 4 Sep 2015, at 20:23 , Pete Woods  wrote:

> You guys have uploaded the other video drivers to the PPA, but I suspect we 
> also need to upload:
> 
> xserver-xorg-video-vmware
> 
> in addition. I am encountering the same problems as michi.
> 
> On Fri, Aug 21, 2015 at 6:07 AM, Michi Henning  
> wrote:
> For what it's worth, in case someone else stumbles into this:
> 
>  apt-get install --reinstall xserver-xorg
> 
> This fixed it. (Thanks dobey!)
> 
> Michi.
> 
> 
> On 14 Aug 2015, at 6:53 , Michi Henning  wrote:
> 
> > Thanks for your help guys!
> >
> > There are no obvious errors in Xorg.0.log, but in lightdm.log, I find:
> >
> > DisplayServer x-0: Can't launch Xserver X -core, not found in path
> > DisplayServer x-0: Xserver stopped
> >
> > Any ideas?
> >
> > Thanks,
> >
> > Michi.
> >
> >
> > On 14 Aug 2015, at 0:05 , Łukasz 'sil2100' Zemczak 
> >  wrote:
> >
> >> Hey,
> >>
> >> I uploaded the intel, ati and nvidia drivers to the overlay PPA when
> >> working on getting this resolved. Those were the three drivers that
> >> seemed to be affected by the original bug.
> >>
> >> Cheers,
> >>
> >> W dniu 13.08.2015 o 15:15, Albert Astals Cid pisze:
> >>> Ah then probably you're not using the intel driver, i'd say have a
> >>> look at the /var/log/Xorg* and see which driver is failing, maybe we
> >>> need rebuilds of more drivers, myself i'm using
> >>>
> >>> ii  xserver-xorg-core
> >>> 2:1.17.2-1ubuntu0.1~overlay1  amd64
> >>> Xorg X server - core server
> >>> ii  xserver-xorg-dev
> >>> 2:1.17.2-1ubuntu0.1~overlay1  amd64
> >>> Xorg X server - development files
> >>> ii  xserver-xorg-video-intel
> >>> 2:2.99.917-1~exp1ubuntu2.3~overlay1   amd64
> >>> X.Org X server -- Intel i8xx, i9xx display driver
> >>>
> >>>
> >>> On Thu, Aug 13, 2015 at 2:19 PM, Michi Henning
> >>>  wrote:
> >>>>
> >>>> On 13 Aug 2015, at 21:05 , Albert Astals Cid 
> >>>>  wrote:
> >>>>
> >>>>> If you update && dist-upgrade you should get the new intel driver from
> >>>>> the stable-phone-overlay that fixes it.
> >>>>
> >>>> Thanks! I tried that, but no joy. I still get a black screen with a 
> >>>> blinking cursor at the top.
> >>>>
> >>>> I've tried both with both versions mentioned earlier, same result.
> >>>>
> >>>> What version should I be seeing for things to work?
> >>>>
> >>>> All this is in a VMWare Fusion VM. I've rebuilt Vmware Tools with both 
> >>>> driver versions installed, but that didn't make any difference.
> >>>>
> >>>> Is anyone else around who has Vivid with the overlay working in VMWare 
> >>>> Fusion? I'm running the latest version (7.1.2).
> >>>>
> >>>> Thanks,
> >>>>
> >>>> Michi.
> >>>>
> >>>>
> >>>>>
> >>>>>
> >>>>> On Thu, Aug 13, 2015 at 11:35 AM, Michi Henning
> >>>>>  wrote:
> >>>>>> After doing an upgrade and installing 
> >>>>>> xserver-xorg-core=2:1.17.1-0ubuntu3
> >>>>>> again, I still don't get a desktop.
> >>>>>> The same thing happens with 
> >>>>>> xserver-xort-core=2:1.17.2-1ubuntu0.1~overlay1.
> >>>>>>
> >>>>>> Does anyone have any suggestions for how I can my desktop back?
> >>>>>>
> >>>>>> Thanks,
> >>>>>>
> >>>>>> Michi.
> >>>>>>
> >>>>>>
> >>>>>> On 6 Aug 2015, at 8:25 , Kevin Gunn  wrote:
> >>>>>>
> >>>>>> Hi all -
> >>>>>> So I felt like this needed to be shared just in ca

Re: [Ubuntu-phone] Can't flash my Arale (tried a bunch of different suggestions)

[Michi Henning]

I've been experiencing similar things. Trying to flash from my Vivid desktop is 
useless, I never get anywhere, with things failing randomly (and almost always).
Flashing from a VM on my Macbook Pro works better, but also fails at random 
every now and then. That's when trying to flash a Nexus 4.

Cheers,

Michi.


On 8 Sep 2015, at 1:23 , Pete Woods  wrote:

> On Mon, Sep 7, 2015 at 4:17 PM, Pete Woods  wrote:
> Hi folks,
> 
> I have an early 2011 17" MacBook Pro and one of the earlier Arale devices.
> 
> I've tried a bunch of suggestions to get this Arale flashed.
> 
> I have tried:
> Flashing normally from Ubuntu
> Flashing using udbflash from OSX
> Flashing from a VMware Ubuntu install
> In terms of hardware I have tried:
> A whole bunch of different USB cables. The device is now only visible to my 
> Mac when I use a rather short Anker branded 1ft USB cable. Any longer than 
> that and it's a no-show.
> USB (2.0) HUB, powered and unpowered.
> All the ports on my MacBook.
> Waiting until the devices are cool (in-case it was some quirky thermal 
> management thing).
> This also happens with some regularity on my other MediaTek device, the 
> krillin. However flashing my Mako is totally reliable.
> 
> Some of the highlights of my logs: http://paste.ubuntu.com/12307654/
> 
> I had to modify udbflash to get useful errors out. Originally it was just 
> swallowing the standard IO and not reporting it back. But what this reveals 
> is that:
> Sometimes "fastboot flash recovery XXX" fails.
> Sometimes "fastboot format cache" fails.
> Sometimes when booting into recovery ADB never connects. However I can work 
> around this by unplugging and quickly replugging the device when it gets into 
> recovery.
> The larger UDB file pushes always fail before getting to the end. The amount 
> of progress they make varies each time.
> I also tried modifying the flash tool to "adb push" the files serially 
> instead of in parallel, and this didn't help either.
> 
> I have a co-worker who also cannot flash his Krillin in the same manner as I 
> cannot flash my Arale. This is with an Ubuntu supported Dell laptop.
> 
> As I work on the network indicator, at the moment I cannot debug any issues 
> present on the Arale, so this is a pretty big deal to me.
> 
> Please help!
> 
> Thanks, Pete
> 
> I should add that this previously (a couple of months ago?) was working 
> extremely reliably.
> 
> I have also had success in the past using the MediaTek flashing tool, 
> although I haven't tried this recently.
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Can't flash my Arale (tried a bunch of different suggestions)

[Michi Henning]

This is not what I'm seeing.

I'm getting random failures while the images are downloading. All of this 
worked fine previously, but has become very flaky in the past two months or so.

I've made sure that it's not a bad USB cable, etc.

Michi.


On 8 Sep 2015, at 7:43 , Alfie John 
 wrote:

> On Tue, Sep 08, 2015 at 07:12:55AM +1000, Michi Henning wrote:
>> I've been experiencing similar things. Trying to flash from my Vivid desktop
>> is useless, I never get anywhere, with things failing randomly (and almost
>> always). Flashing from a VM on my Macbook Pro works better, but also fails at
>> random every now and then. That's when trying to flash a Nexus 4.
> 
> I'm on Debian Jessie, and am running Ubuntu Trusty under VirtualBox. At first,
> flashing would also fail at random times for me too. After some debugging, I
> realised that sometime after the inital fastboot flash, VirtualBox drops the
> USB connection.
> 
> So if flashing under a VM, select the USB device to connect, but then keep an
> eye out for any connection drop because you'll have to reselect the device
> (usually happens when the black fastboot screen changes to the Ubuntu logo).
> 
> As a note, the USB device name changes here too (I can't recall, and am away
> from my machine). During fastboot, it seems like it's the name of the USB
> chipset manufacturer, then once the Ubuntu logo is displayed, the device name
> changes to BQ E5.
> 
> Hope this helps.
> 
> Alfie
> 
> -- 
> Alfie John
> https://www.alfie.wtf


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Scopes not working without WiFi

[Michi Henning]

> Scopes like the Ubuntu Store, Reddit, Ebay, 7Digital or the Weather Channel 
> don't work if I'm not connected to a WiFi network. It loads and stops after a 
> while, completely blank. My 3G (well, H) connection seems fine and I can use 
> the browser and webapps with reasonable speed. Other scopes like Wikipedia, 
> Yelp or Youtube are working OK. All scopes work as expected over WiFi.

Hmmm... Are you saying that the remote scopes (such as Weather Channel) work if 
you are connected via WiFi, but don't work if you are connected via the 
cellular network?

If so, that suggests that the scopes runtime is almost certainly innocent. It 
has no idea how the phone is connected and just makes HTTP requests to the 
remote server. Failing only when connected via the cellular network would 
suggest some problem in the lower-level networking stuff, such as 
QNetworkAccessManager or below that.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] vivid+stable-phone-overlay on desktops

[Michi Henning]

On 18 Sep 2015, at 2:35 , Pete Woods  wrote:

> The stable overlay has now had the rebuilt vmware driver added to it. This 
> has enabled me to upgrade my vmware machines without X breaking any more.

Yea, thank you!

Michi.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Fixing Volume Controls

[Michi Henning]

This is not a totally trivial problem. Just think of calendar alerts which 
should probably have their own volume level that is *not* under application 
control, but user control. Moreover, they should probably always play over the 
speaker, unless the phone is silenced.

I wrote an article that touches on some of these topics a few years ago:

http://www.triodia.com/staff/michi/blog/Apple.pdf

But, in general, I tend to agree. Fewer roles is probably better, in the 
interest of user sanity.

Cheers,

Michi.


On 21 Sep 2015, at 18:30 , Michael Zanetti  
wrote:

> Hi list,
> 
> I'd like to discuss a topic which, as a daily user of the phone drives
> me a bit bad and makes me lose countless phone calls: Volume controls.
> 
> If you go through launchpad bugs, you'll find lots of bugs related to
> volume control [1][2][3][4]. Let me summarize it up here:
> 
> The biggest issue is that the volume buttons are completely
> unpredictable. They control a different volume level each time you touch
> them. For instance, play some game with a background music, press the
> volume buttons. It will adjust the game's sound level (maybe, with a bit
> of luck). Then start the game "Dinosaur" from the store and try again,
> you'll notice that while playing this game the volume buttons don't
> affect the sound made by the game any more but instead they'll leave the
> game at 100% volume, while silencing your phone call ringtones. This is
> the thing that makes me lose phone calls all the time. I play some game
> with the phone, and when I'm done my ringtones are silenced without me
> even knowing.
> Another issue is that currently there's not really way to figure what
> volume level your device is set to. The slider in the indicator is quite
> meaningless, given it will fool you by displaying some value, but next
> time a sound is produced it will jump to some different value before
> playing the sound. This also creates the issue that there's no way to
> control your sound level *before* you start a game, means you can't
> really start a game on Ubuntu Touch in a place where you can't make
> noise, because you have no chance to set the volume before it starts
> playing.
> 
> I've spent some time to figure where the issue is coming from and what a
> solution could be. The main reason where this issue is coming from is
> that we have a number of different audio roles that applications are
> using. The volume buttons, as well as the slider in the indicators try
> to be clever in knowing which role they should address whenever you
> touch them. This, however, seems to reliably fail in 99% of the cases
> you want to use it.
> 
> So here's my proposal:
> 
> * Let's drop all the audio roles we have except for ringtone (alert) and
> multimedia. We don't need more of them, it'll make things unnecessarily
> complex.
> 
> * Make the hardware buttons always control the multimedia volume, that
> is, game sounds, music player etc but don't ever touch ringtone volume
> with them.
> 
> * Make the volume slider in the indicator always control the multimedia
> volume, never touch the ringtone volume.
> 
> * Apps should always use the multimedia role, regardless of what they do
> (unless thy deal with actual calls or incoming messages).
> 
> * Add a second "slider" to the indicators that controls ringtone volume.
> That slider could be a bit special tho, i.e. that it works more like a
> profile selector. 0 is "vibrate only", 1 is "beep" and all the greater
> values adjust volume of the ringtones.
> 
> One thing I've not mentioned here yet are alarms. Alarms, again
> could/should have their own role. But one never controls that role with
> either the buttons or a slider. Instead, each alarm has his own volume
> value assigned when the alarm is created, and an alarm is always played
> at the volume its config says (except maybe only vibrate when the phone
> profile is set to vibrate only - but those are small details we could
> work out on the road).
> 
> IMO this behaviour would make the volume way more predictable,
> especially since one needs to use a slider in the indicators to change
> the ringtone volume but at the same time giving the user visual feedback
> on what the current volume is.
> 
> What do you think? If you're working on anything related to this, please
> help to get this issue fixed and improve the Ubuntu Phone experience.
> 
> Br,
> Michael
> 
> 
> [1] https://bugs.launchpad.net/canonical-devices-system-image/+bug/1484463
> [2] https://bugs.launchpad.net/canonical-devices-system-image/+bug/1478506
> [3] https://bugs.launchpad.net/ubuntu/+source/indicator-sound/+bug/1478075
> [4] https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1291458
> 
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu

Re: [Ubuntu-phone] Fixing Volume Controls

[Michi Henning]

I would look to the iPhone for inspiration. These days, it works pretty well 
most of the time.

As far as I know, there is no way to adjust playback level in advance for 
something I haven't started yet on the iPhone either. If I want to start a game 
silently, I can switch the phone into mute mode though before starting the 
game. Once the game has started, the volume control changes the game volume 
(not the ringer), and I can take phone out of mute mode after adjusting the 
volume.

Michi.


On 21 Sep 2015, at 19:00 , Michael Zanetti  
wrote:

> Hi Alan,
> 
> I see your concerns, valid points there.
> 
> Probably the hardware buttons should then control ringtone volume while
> no app is focused (or a call indication is on top) and control the
> multimedia volume whenever an app is focused or media-hub is playing in
> the background.
> 
> Switching the functionality solely depending if audio plays or not (as
> we currently do) will always have the problem that it switches too
> often/quickly so you'd be never able to adjust a volume in games like
> Dinosaur with them.
> 
> Would this be something you think it would work?
> 
> Br,
> Michael
> 
> On 21.09.2015 10:52, Alan Pope wrote:
>> Hi Michael,
>> 
>> On 21 September 2015 at 09:30, Michael Zanetti
>>  wrote:
>>> * Make the hardware buttons always control the multimedia volume, that
>>> is, game sounds, music player etc but don't ever touch ringtone volume
>>> with them.
>>> 
>> 
>> While this works for your specific use case of "Game is making noise
>> in a library", what if I'm in the library and my phone rings? I want
>> to reduce the volume quickly so the call goes to voicemail. If the
>> volume keys only work for multimedia then I can't quiet a ringing
>> phone.
>> 
>>> * Make the volume slider in the indicator always control the multimedia
>>> volume, never touch the ringtone volume.
>>> 
>> 
>> Again, how do I reduce the volume of an incoming call?
>> 
>>> * Apps should always use the multimedia role, regardless of what they do
>>> (unless thy deal with actual calls or incoming messages).
>>> 
>>> * Add a second "slider" to the indicators that controls ringtone volume.
>>> That slider could be a bit special tho, i.e. that it works more like a
>>> profile selector. 0 is "vibrate only", 1 is "beep" and all the greater
>>> values adjust volume of the ringtones.
>>> 
>> 
>> This won't help if your phone is locked (and indicators aren't
>> accessible in one of our security settings).
>> 
>>> One thing I've not mentioned here yet are alarms. Alarms, again
>>> could/should have their own role. But one never controls that role with
>>> either the buttons or a slider. Instead, each alarm has his own volume
>>> value assigned when the alarm is created, and an alarm is always played
>>> at the volume its config says (except maybe only vibrate when the phone
>>> profile is set to vibrate only - but those are small details we could
>>> work out on the road).
>>> 
>>> IMO this behaviour would make the volume way more predictable,
>>> especially since one needs to use a slider in the indicators to change
>>> the ringtone volume but at the same time giving the user visual feedback
>>> on what the current volume is.
>>> 
>> 
>> I'd rather we just did what Android does and have an icon in the
>> volume popup so you know what you're adjusting.
>> 
>> e.g. http://imgur.com/ggQdZBS (adjusting media volume) vs
>> http://imgur.com/3Y1vapF (adjusting alert volume).
>> 
>> Cheers,
>> 
> 
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] Stuck during apt-get upgrade

[Michi Henning]

I'm seeing this on Vivid plus overlay on my desktop as well as on my Nexus 4 
when using citrain device-upgrade.

During package installation, the bluez install hangs. Looking at the processes, 
I see

 28628 pts/1S+ 0:00 sudo apt-get upgrade
 28629 pts/1S+29:25 apt-get upgrade
 36342 pts/20   Ss+0:00 /usr/bin/dpkg --status-fd 66 --unpack 
--auto-deconfigure /var/cache/apt/archives/bluez_4.101
 36348 pts/20   S+ 0:00 /bin/sh /var/lib/dpkg/info/bluez.prerm upgrade 
4.101+15.04.20150914.2-0ubuntu1
 36351 pts/20   S+ 0:00 /bin/sh /usr/sbin/invoke-rc.d bluetooth stop
 36392 pts/20   S+ 0:00 systemctl stop bluetooth.service
 36393 pts/20   S+ 0:00 /bin/systemd-tty-ask-password-agent --watch
 36394 pts/20   Sl+0:00 /usr/bin/pkttyagent --notify-fd 5 --fallback

Can anyone suggest a fix/work-around?

Thanks,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Fixing Volume Controls

[Michi Henning]

On 30 Sep 2015, at 20:21 , Michael Zanetti  
wrote:
> 
> IMHO, apps should not be able to change the system volume... They might
> have a slider inside which then adjusts their own sounds relative to the
> current system value, but in no way an app should be able to change
> another app's volume.
> 
> Example: If I set the system's multimedia volume to 50%, those 50%
> should the maximum for all apps. Inside apps, they can have sliders if
> they want to, but setting them to 100% should give the current 50% of
> the system volume, setting them to 50% would then end up playing at 25%
> of what the system supports.

I strongly agree with that. Just think about the potential for annoying people 
with loud nagging and the like if applications can change system volume.

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Request for Comments: Potential New CI Train design

[Michi Henning]

On 30 Sep 2015, at 21:33 , Jean-Baptiste Lallement 
 wrote:

> Actually it looks like poeple want the old dashboard back :)

The week of living dangerously…  :-)

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Why js not c++?

[Michi Henning]

> you can indeed build apps in C++ if you want, but C++ is more complex to
> understand and write (if you ever did a dynamic web page in your life
> you likely know the basics of js). C++ needs a (cross) compile
> environment set up while js means you can just dump a txt (well, .qml)
> file in place and it just works. Doing C++ is just a lot more work and
> while you can use C++ I think people find it easier to simply use js (I
> surely do, i can write a ready made QML app including rolling the click
> and uploading it to the store with a plain text editor within 20min, I
> (personally) cant do that in C++)) ... 

Writing a dynamic web page in C++ is a bit like writing a neural network in 
COBOL. Don't.

Would I write a device driver in JS? Probably not.

Would I try to tighten a Phillips head screw with a nail file? Probably not 
either.

I think you get the drift… :-)

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Landing team 03.11.15

[Michi Henning]

Thanks Łukasz!

Just a question: how long does it typically take for the new stuff to make it 
into rc-proposed?
I just downloaded the image for Mako and got build 278, which doesn't include 
the new thumbnailer yet, as far as I can tell.

Thanks,

Michi.

On 4 Nov 2015, at 10:01 , Łukasz 'sil2100' Zemczak 
 wrote:

> Hello everyone,
> 
> A bit more exciting today landing-wise, it seems that bug-fixes started
> flowing more steadily recently. First, we had a new bugfix UITK landing
> plus a new thumbnailer. Those were followed by signon (adding new error
> codes in the UI), a new upstream release of libaccounts-qt and finally
> finished off with a buteo-syncfw landing.
> 
> With the help of Colin and Martin we also switched the times of the
> auto-generation of langpacks to Wednesdays as mentioned earlier. This
> way it fits a bit more to our release model nowadays.
> 
> We also slightly modified the release schedule [1] - starting with OTA-8
> we will try not to freeze the overlay PPA during the release
> preparations. We'll do this by snapshotting during the candidate
> creation and building all our concurrent release candidates in the rc
> channels instead of rc-proposed. This way rc-proposed can stay open,
> making things easier for developers.
> 
> Cheers!
> 
> 
> * What landed in our images:
> http://people.canonical.com/~lzemczak/landing-team/ubuntu-touch
> 
> #163 rc-proposed/krillin
> -
> http://people.canonical.com/~lzemczak/landing-team/ubuntu-touch/rc-proposed/163.commitlog
> 
> 
> * Smoke-testing results:
> devel : http://ci.ubuntu.com/smokeng/wily/touch/
> rc : http://ci.ubuntu.com/smokeng/vivid/touch_stable/
> 
> 
> * Some useful documentation:
> https://wiki.ubuntu.com/LandingTeam
> 
> https://wiki.ubuntu.com/citrain/RTMLandingApproaches
> https://wiki.ubuntu.com/citrain/LandingProcess
> https://wiki.ubuntu.com/LandingTeam/Smoketesting
> https://wiki.ubuntu.com/LandingTeam/ReleaseSchedule [1]
> http://developer.ubuntu.com/start/ubuntu-for-devices/image-channels/
> 
> https://requests.ci-train.ubuntu.com/
> https://trello.com/b/AE3swczu/silo-testing
> http://people.canonical.com/~lzemczak/issues/
> 
> 
> * Blocking issues (stable):
> 
> None.
> 
> 
> * Blocking issues (devel):
> 
> None.
> 
> 
> 
> Best regards,
> 
> -- 
> Łukasz 'sil2100' Zemczak
> Foundations Team
> lukasz.zemc...@canonical.com
> www.canonical.com
> 
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] Find out if running on a phone

[Michi Henning]

I’d like to find out (at run time, from C++), whether I’m running on a 
phone/tablet as opposed to a desktop.

Is there any way to do that?

Checking for Arm isn’t good enough, because a Chromebook runs on Arm, but isn’t 
a phone or tablet.

Or, to put it differently, I’d like to find out whether I’m running in a touch 
environment as opposed to a desktop environment.

Does anyone have any ideas?

Thanks,

Michi.


signature.asc
Description: Message signed with OpenPGP using GPGMail
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Find out if running on a phone

[Michi Henning]

> That said, there might be some rare cases where you need info on the
> currently connected input devices. We're working on an API for that [1]
> which I think Timo is already looking to add to the images, but it's not
> landed yet.

Thanks Michael! Is there any doc or header files I can look at? The silo 
doesn’t include an MR link.

> In any case, if you want to know if you are "on a phone or on a
> desktop", you are probably trying to walk the wrong direction.

This is for the thumbnailer. Depending on whether hardware codecs (phone) or 
software codecs (desktop) are installed, I need to change performance tuning 
parameters. The hardware and software codecs have very different timings, and, 
unfortunately, different semantics. For example, the hardware codecs cannot be 
run concurrently on the device because of a kernel/android bug:

https://bugs.launchpad.net/ubuntu/+source/thumbnailer/+bug/1466273

Michi.

> 
> [1] https://requests.ci-train.ubuntu.com/#/ticket/763
> 
> Hope this helps,
> Michael
> 
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Find out if running on a phone

[Michi Henning]

> 
> Right... forgot that it's a gerrit MR, not Launchpad:
> 
> https://codereview.qt-project.org/#/c/101049/ 
> 

Thanks for that!

I just had a look at 
https://codereview.qt-project.org/#/c/101049/15/src/systeminfo/qinputinfo.h 


But this requires an input to then find out what the source was (mouse, touch 
screen, etc.).

I need to know whether I’m running on touch or on the desktop, without any 
input events.

Any other ideas? Is there some place in the filesystem I could look at, or test 
some env var or whatever?
It has to work under confinement though.

Thanks,

Michi.-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Find out if running on a phone

[Michi Henning]

> On 11 Jan 2016, at 14:12 , Robert Park  wrote:
> 
> Pardon my ignorance, buy can you not just check which codecs are installed/in 
> use and then make the right decisions based on that? It is a really bad 
> practice to write code that says "desktops be like this while phones be like 
> that" because you never know when one day a new device might come along that 
> shatters your expectations.
> 
> Feature detection > device detection
> 

I agree with the sentiment. But how would I, from a confined process, work out 
what’s installed? I won’t even be allowed to look there.

I guess I can conclude that, if I’m not allowed to look, I’m confined. And, if 
I’m allowed to look, use what I can see. But it also seems a very round-about 
way of figuring this out.

The real question actually is “If I thumbnail from an audio or video file, will 
I be using the android/hybris hardware codecs or a software codec?”

But there is no API for that either, as far as I know.

Michi.-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Find out if running on a phone

[Michi Henning]

> I wouldn't make the assumption that confined processes are necessarily 
> running in the phone as i believe the convergence plan is to have confined 
> apps on the desktop.
> 
Right.
> Aside from that, i would just try to do whatever it is you're trying to do 
> and then catch the failure as gracefully as possible when it fails on the 
> phone, if you can. Better to ask forgiveness than permission ;-)
> 
No can do, I’m afraid. The decision as to how many concurrent extractions to 
allow fundamentally depends on knowing whether or not we are using the Android 
hardware codecs. There is no way to figure this out that I know of, other than 
to ask “am I running on touch?” because, by definition, we *know* (at least for 
the time being) that, on touch, we are using hardware codecs.

I cannot catch the failure. For one, it is non-deterministic. Secondly, when 
the failure does happen, it manifests itself as several unkillable processes 
that are stuck in the kernel, each spinning at 100% CPU for 30 seconds. That 
failure mode is not only rather hard to detect, it’s also not a good way to 
find out that I shouldn’t have done what I did…

Right now, we are working around the problem by disabling concurrent extraction 
whenever we are running on Arm. But that disables concurrency even when 
software codecs are in use, such as on a Chromebook. In turn, that means we are 
not using three perfectly good CPUs that we could use, if only we had an answer 
to the question :-(

Michi.-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Find out if running on a phone

[Michi Henning]

>> 
>> But this requires an input to then find out what the source was (mouse,
>> touch screen, etc.).
> 
> It does not need an event first. QInputInfoManager::deviceMap() will give a 
> list of devices. You can filter these by setting a filter first.

Ah, yes, right you are, thanks!

Hmmm, but what to look for to decide whether I’m running on touch or desktop? A 
lot of laptops have touch screens these days.

I suspect that the QInput API is simply the wrong API for my particular problem.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Find out if running on a phone

[Michi Henning]

> So, from a confined process, can I find out whether Hybris is installed?

OK, so I can successfully stat() /usr/lib//libhybris from a confined 
process.

Looks like that’s the ticket I need to get me off the hook, thanks!

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Find out if running on a phone

[Michi Henning]

> it's an API that is being developed. :)

Right. And I’m not trying to blame the API here, seeing that it really tries to 
solve a different problem. I’m not asking for the API to do what I need because 
I suspect that my question should be answered by a different API entirely. (For 
one, an API that doesn’t have a Qt dependency would seem more appropriate.)

> If it has a touchscreen and the screen is size > xx might be guess. or, if it 
> has hybris on it. :)

Hmmm…  The hybris angle sounds promising, thanks!

So, from a confined process, can I find out whether Hybris is installed?

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Find out if running on a phone

[Michi Henning]

> Yes, as I said in the other mail, this approach sounds definitely wrong
> to me. What you really want is to find out whether the android
> hardware/drivers are around or not.

Right. I agree with the concerns about it being hacky, I don’t like it either.

It turns out that testing for the presence/absence of /usr/lib//libhybris 
works (at least at the moment) to distinguish touch from the desktop. I tested 
on the phone and on my Chromebook, and get the right answers.

But it turns out that the point may well be moot. I re-enabled concurrent 
extraction for audio and video files on my Chromebook, just to see how it’ll 
go. Turns out that codecs get stuck and just hang on my Chromebook when more 
than one runs at a time, so we can’t enable concurrent extraction on Arm anyway 
for the time being.

Thanks for the suggestions everyone!

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Find out if running on a phone

[Michi Henning]

> 
> Would this get your hardware stuff enabled here then? If so, the check
> is not good enough.

Right. So much for that.

>> But it turns out that the point may well be moot. I re-enabled concurrent 
>> extraction for audio and video files on my Chromebook, just to see how it’ll 
>> go. Turns out that codecs get stuck and just hang on my Chromebook when more 
>> than one runs at a time, so we can’t enable concurrent extraction on Arm 
>> anyway for the time being.
> 
> Wasn't the idea to enable concurrent extraction for when we are using
> the software codecs?

What I was trying to do was turn on concurrent extraction on Arm, but *only* if 
we are not on a phone. Turns out that concurrent extraction is broken on Arm in 
general, not just on the phone, so the point is moot now. I’ll leave it as for 
the time the being, which is to run only one gstreamer pipeline at a time on 
Arm, which is the only thing that works.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Upgrading OTA-9 manually via local server

[Michi Henning]

I’ve had similar problems for many weeks with Mako. Something about the way USB 
works isn’t right. I’m using the Mako with three machines:

Core i7 desktop running xenial. When I connect the USB cable, the phone 
immediately shows that it has power, and ubuntu notices the new device and 
mounts it. Thereafter, it’s up and down like a yoyo every few seconds: 
disconnect, connect, disconnect, connect, …, indefinitely.

On my Macbook Pro running vivid, wily, and xenial in Fusion VMs, I can connect, 
but adb no longer transfers files larger than about 1.5 MB. Smaller files than 
this work fine with adb push. Anything larger hangs the USB connection 
completely. For example, if I adb push of a large file in one terminal and have 
a phablet-shell running in another terminal, both terminals hang and no longer 
respond to SIGINT. I have to unplug the USB cable to get things back to life.

On my Chromebook, things work better, but are unreliable. It often takes three 
or four tries to flash the phone. When it fails, it fails because adb hangs.

It’s definitely not the cable, I’ve tried with several. I guess it could be my 
phone’s USB port. Anyone know how I could find out whether that’s the case?

Thanks,

Michi.


> On 30 Jan 2016, at 8:49 , Camilo Ernesto Blanco Peña  wrote:
> 
> Hi Pat:
> 
> Thanks for your support! 
> 
> > Do you see the files download from your mirror?
> 

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] OTA 9.1 Bug

[Michi Henning]

Sounds like something is crashing. Could you have a look at /var/crash and see 
whether there are any reports there?

Cheers,

Michi.


> On 25 Feb 2016, at 8:40 , Louis  wrote:
> 
> Hello
> 
> I have a BQ Aquaris E5 Ubuntu Edition.
> I installed the OTA-9.1
> Now Scopes have a problem.
> Scopes can't start successfully. They always shutdown and start again.
> What should I do ?
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Landing team 04.03.16

[Michi Henning]

> A lot of exciting changes were happening this week. I won't be
> mentioning all of them as it would be a rather big list (and most of you
> probably know the changes now anyway) - but a lot of them were good
> fixes (mir, urfkill, scopes).

Also try out media scanner and thumbnailer.

James improved the media scanner. It's now 4.5 times faster for indexing, and 
it greatly reduces the I/O load on the device.

The thumbnailer is 12 times faster now when extracting cover art from local 
audio files. That makes a big difference to the UI when you have a cold cache 
and a large music collection.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] E5 OOB exerience

[Michi Henning]

I just got a new Aquaris E5.

The OOB experience was somewhat lacking.

On my first attempt to set up Wifi access, I selected my network, tapped “Show 
password”. Then, every time I hit a key on the OSK, I ended up with three or 
four random characters. It was impossible to enter the password. I put the 
phone down for a while. Returned to it (screen had locked itself in the mean 
time), and tried again, this time without “Show password”. On this second 
attempt, the keyboard worked as it should.

After setup (I allowed location data), the phone correctly got my location 
(well, within 10 km or so—there is no SIM in the phone). However, it still set 
my time zone to UTC+0, even though it should be UTC+10.

The today scope showed me the weather in Brisbane (well, near there), using 
Fahrenheit units. Here in Australia, we use Celsius but neither Today scope nor 
Weather scope allow me to change units. After changing the language to English 
(Australia), I get Celsius. However, I prefer to use US English for spelling 
and the like. Mashing language and units together in this way is fine for 
setting defaults, but not if there is no way to change one without also 
changing the other.

I then copied a bunch of music files onto the phone, using drag and drop from 
Nautilus. The copy started out quite well but, after a few seconds, transfer 
speed got less and less. After a few dozen files, it was down to 20 KB/sec. 
(Each file is a little over 8 KB in size.) Now, after 300 files, it’s still 
copying, but transfer speed has dropped to 6.5 KB per second, and is still 
dropping steadily (5.5 KB per second after 400 files). Now, after 15 minutes of 
waiting, it has transferred 4.2 MB (500 files) out of 8.7 MB total, and it 
tells me that I need to wait another 15 minutes for the remaining 4.5 MB, with 
the reported transfer speed at 4.5 KB/sec.

The summary is that, so far, I tried to do four things: enable Wifi access, get 
the current time, look at the weather, and copy some files into the Music 
folder. Not one of these worked as expected. Not a good first impression.

Anyone know what the appropriate projects in launchpad are to report these 
issues?

Thanks,

Michi.


signature.asc
Description: Message signed with OpenPGP using GPGMail
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> It is now half an hour later, and the copy is still in progress. It now has 
> copied 935 of the 1001 files. Each 8 KB file takes around 6 seconds to copy 
> at this point.

In the end, I gave up and stopped the copy at around 950 files because I got 
sick of waiting. In the mean time, the software update in the Settings app had 
downloaded a new image, so I installed that. After flashing, the phone rebooted 
as expected. It has now been sitting on the while start-up screen saying “bq” 
and “powered by unbuntu” for a good ten minutes.

The base image I received the phone with was:

$ system-image-cli -i
current build number: 2
device name: vegetahd
channel: ubuntu-touch/stable/bq-aquaris.en
alias: ubuntu-touch/ubuntu-rtm/14.09
last update: 2015-06-15 08:55:13
version version: 2
version ubuntu: 20150508
version device: 20150527-dd39a0f
version custom: 20150507-685-29-216

I don’t know what the current image is because I can’t get the phone to 
respond. I’ll wait a while longer and then try a forced reboot if it doesn’t 
come up.

Michi.


signature.asc
Description: Message signed with OpenPGP using GPGMail
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> I then copied a bunch of music files onto the phone, using drag and drop from 
> Nautilus. The copy started out quite well but, after a few seconds, transfer 
> speed got less and less. After a few dozen files, it was down to 20 KB/sec. 
> (Each file is a little over 8 KB in size.) Now, after 300 files, it’s still 
> copying, but transfer speed has dropped to 6.5 KB per second, and is still 
> dropping steadily (5.5 KB per second after 400 files). Now, after 15 minutes 
> of waiting, it has transferred 4.2 MB (500 files) out of 8.7 MB total, and it 
> tells me that I need to wait another 15 minutes for the remaining 4.5 MB, 
> with the reported transfer speed at 4.5 KB/sec.

It is now half an hour later, and the copy is still in progress. It now has 
copied 935 of the 1001 files. Each 8 KB file takes around 6 seconds to copy at 
this point.

Michi.



signature.asc
Description: Message signed with OpenPGP using GPGMail
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> I don't have those issues either with copying music or wifi - I'd be 
> interested to know which update the phone came with, and what the current one 
> is now. Still very early days for the system, but I much prefer it to 
> android. Good luck, and send me an email with anymore ideas for Australian 
> content.

The originally installed image was build 2. The new one is OTA-9.1 (build 10).

I’m doing this on a MacBook Pro, running Xenial in a Fusion VM. I’ll try the 
USB copy on my desktop and my Chrombook when I get back from holidays. It’s 
possible that the USB problems relate to VMWare.

Still, if if the USB issue turns out to be VMWare related, the other points I 
reported still don’t make for a good user experience.

Michi.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> On 21 Mar 2016, at 17:24 , Gareth France  wrote:
>> 
> It is well established that Ubuntu is not yet ready for prime time on phones. 
> My handset has not lived up to the bar you set since they day I got it yet I 
> wouldn't trade it for the world. Issues are being resolved every month and in 
> my opinion the fact a smaller company like Canonical has achieved what the 
> likes of Google and Apple have is a major achievement. The team working on 
> this is tiny by comparison and you wouldn't expect them to be able to 
> complete this journey in anything near the time the big boys can.

I don’t dispute that Canonical has achieved a lot, and has done it with very 
few people and in record time. But that doesn’t change the fact that, had I 
paid money for this phone, after the experience I just had, I would have 
returned it to the store the same day, demanding a refund.

To the customer, it doesn’t matter how hard people are working, or how many 
people Canonical employs for this. To the customer, only one thing matters: 
whether the thing works adequately or not.

What I just experienced is a long way from “working adequately”.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> On 21 Mar 2016, at 17:27 , Mitchell Reese  wrote:
> 
> I'd suggest contacting BQ - assuming that's where you bought the phone from. 

Mitchell, I’m a Canonical developer, and I didn’t pay for the phone. It was 
sent to me by Canonical for my development work.

Not that it makes any difference. What I just went through totally sucks. 
Literally everything I did with the phone since I unpacked it, no matter how 
trivial, has failed. Followed by the “hmmm… I wonder whether the upgrade it 
just flashed has bricked my phone” experience (after the phone didn’t reboot 
following the OTA-9.1 update).

It’s hard to think of a more disappointing experience.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> On 21 Mar 2016, at 17:30 , Mitchell Reese  wrote:
> 
> I get your point, but I disagree - I think touch is ready, and getting better 
> every day.

Given what just happened to me, I’d say this is quite a long way from ready.

If we want to be taken seriously, we had better lift our game.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> The big question for me is this. You were obviously running the system with 
> quite old software, the OTA update has resolved that now. So, you are 
> assessing the system as it is today based on an experience with out of date 
> code. Is it still the same after the update?

Well, let’s see…

USB transfer is still unusable. It might be a VMWare problem, which I’ll verify 
when I get back from holidays.

The phone still is set to the wrong time zone.

The language and units are still tied together, so I can have either US English 
and Fahrenheit, or Australian English and Celsius. But US English and Celsius 
is impossible.

It tried to reproduce the issue with entering the Wifi password.

So, I went to the Settings up and said “Forget this network” to get rid of the 
current connection.

After doing that, I figured I can re-connect and re-enter the password. No joy.

Now the settings up shows me “Previous Networks” and “Connect to hidden 
network…”. (My network is not hidden.) When I tap on “Previous networks”, I get 
a completely blank page. (Why does it invite my to look at previous networks 
when there aren’t any?) No wifi networks are shown at all now, even though 
there are several around.

Pulling down on the network indicator has the same problems. It doesn’t offer 
me any networks I could connect to.

I suppose the next step will be to reboot to see whether that changes anything.

See what I mean when I say that this far from ready?

I haven’t even tried to use the phone in earnest yet, and I’ve run into six 
separate problem by now, all within an hour or so.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> Pulling down on the network indicator has the same problems. It doesn’t offer 
> me any networks I could connect to.
> 
> I suppose the next step will be to reboot to see whether that changes 
> anything.

Sure enough, after a reboot, I can see the available networks again. So, I made 
the mistake of disconnecting from my network and telling the phone to forget 
that network. Thereafter, I ended up in a state where the phone didn’t show me 
any wifi networks. (It did invite me to look at the previous networks though, 
even though there weren’t any.)

So, I figured, now that it’s shown, I’ll connect to my network again. So, I tap 
the name of my network in the list and get the little dialog to enter the 
network password. Because I’m trying to reproduce the earlier error with the 
on-screen keyboard, I tap “Show password”. Nothing happens. I tap “Show 
password” again. Still nothing happens. I tap “Show password” on the text 
instead of the tick box. Nothing. Then I madly stab at the tick box several 
times. Low and behold, it turns into a green tick, after considerable delay.

At this point, there is a blinking cursor in the text entry field for the 
password, but no OSK. I tap on the text box so I get the OSK. No OSK appears. I 
tap again. No OSK. Just brilliant.

I tap cancel to get back to the network list, and tap my network again. I tap 
the “Show password” tick box. No reaction. I tap it a second time. This time it 
turns green. I tap on the text entry field. No keyboard.

Rinse and repeat: This time, I tap on the network name, the password dialog 
appears, and I manage to turn on “Show password” on the very first tap. But I 
still can’t enter the password because it appears impossible to get the OSK to 
appear.

So, I return to the network list. In an attempt to reset things so I can try 
again, I tap the toggle to turn Wifi off. The list of networks goes empty, as 
expected. Then I tap the toggle again to turn wifi back on. The toggle turns 
green, as expected. But now, the network list is empty again, and there is no 
way to bring it back.

Time for another reboot.

Oh, wait, no. About two minutes later, just as I’m about to reboot, suddenly 
the network list appears. So, I again try to enter my network password. I 
can’t, because there is no keyboard.

So, I guess it’s a reflash now, with —bootstrap, in the hope that I’ll 
eventually get a keyboard.

This is ridiculous.

If I can produce this many bugs without even seriously trying in the first two 
hours after receiving the phone, I really wonder what’s going on during testing.

Let me put this very clearly: with this level of UI design and software 
reliability, we will lose 90% of all customers in the first day after purchase. 
This is a lot worse than just “not yet ready”. This is unusable.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> So, I guess it’s a reflash now, with —bootstrap, in the hope that I’ll 
> eventually get a keyboard.

No need to bootstrap. I just did a reboot. Now I’m getting the OSK again. 
Previously, the OSK went missing not just for the wifi password, but for 
absolutely everything.

So, I can finally try to enter my password so I can connect to my wifi network. 
I tap “Show password”. It takes two taps to get the green tick box.

I start entering my password. The first letter works. Then I hit the second 
letter. It’s an ‘a’. I tap the ‘a’ and the text widget shows me “fga”.

I randomly tap a few other letters. Sometimes I get a single letter echoed, 
sometimes two, three or four. Even the backspace key produces two or three 
letters.

This is starting to look like there is something seriously wrong with input in 
general, for any text input field. For example, when I simply press and hold 
down the backspace key, the phone madly types random letters, and I can here 
the vibrator humming like crazy. So, is this a software issue, or do I simply 
have a phone with a bad touch screen?

Michi.



-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> On 21 Mar 2016, at 18:28 , Gareth France  wrote:
> 
> STOP! Don't reflash just yet. The OSK issue I experienced a little while back 
> and a reboot solved this. You will just have to trust me when I say this 
> experience is the exception and not the rule. The delays you encountered when 
> trying to select show password haunt me, especially when trying to hit the 
> make call button or switching from a text preview to the text messaging app. 
> This is something that it being caused by several issues and is being solved 
> in stages.

Well, the reboot doesn’t make any difference. The OSK is misbehaving, that much 
is clear. I don’t know whether that’s a hardware defect or not though.

> It sounds to me like you are wrapping yourself up in issues and getting too 
> worked up about it. Perhaps time to give it a break for a bit and return to 
> it refreshed?

I don’t think the problem is me being tired. The problem is too many things 
being unreliable all at once.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

>   My guess would be that this is a touch input problem, rather than a
> keyboard specific problem but that it's just more visible in the
> keyboard because that's one of the only things that supports
> multitouch. To verify that hypothesis could you try installing the
> Finger Painting app from the store and see if multiple lines appear
> when you try drawing with a single finger?

The finger painting app works fine.

>  If so, that'd definitely suggest some sort of (presumably hardware?)
> issue in the touch detection itself.

I don’t think so because everything else seems to be working just fine. I can 
tap on icons, swipe, etc. The right thing happens all the time. Except for the 
OSK.

Cheers,

Michi.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> Sounds like either your phone or install is seriously borked - I'd reflash 
> with the bootstrap option and see if that helps. Also, check that your 
> software system actually updated. Are you still on OTA 2, or is it 9.5? It 
> may be that such a big update (2 to 9.5) didn't work - the software has been 
> released incrementally. Don't give up yet - the BQ 4.5 is essentially the 
> same hardware (sans the screen), and I have none of those issues.

Well, the update to OTA-9.1 worked just fine, except for the first reboot 
problem. The phone tells me it has 9.1 installed (image 10).

I left the phone lying on the table and went out for a few hours. I didn’t 
change anything, didn’t reboot, didn’t install anything, etc. Three hours 
later, the OSK suddenly worked.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> On 22 Mar 2016, at 1:31 , Andrew Hayzen  wrote:
> 
> Hi Michi,
> 
> The weather-app [0] allows you to adjust your Units :-) Maybe you can use 
> that as a stop-gap for now.

Andrew, the point is not whether there is a work-around. It don’t need a 
stop-gap, in the sense that I use the phone only for testing. The point is that 
an ordinary user gets to see Fahrenheit in the Today scope when they want 
Celsius. There is no obvious way to address the problem because the places the 
user would look (the settings in the Today scope, settings in the weather 
scope, and the settings app) don’t allow the units to be changed.

The ordinary average user will walk away and go “this phone is no good, it can 
only show me temperatures in the wrong units”.

Cheers,

Michi.


signature.asc
Description: Message signed with OpenPGP using GPGMail
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] E5 OOB exerience

[Michi Henning]

> I believe it is the expected behaviour for mobile phone OEMs to make phones 
> in batches, and for those batches to be sold with the software that was 
> current at the time of production. It's particularly expensive to re-visit a 
> phone already made, so I believe no-one routinely updates the software 
> between manufacture and the customer getting it.
> 
> We certainly need to consider how long the OOBE lives in such a batch 
> manufacturing world, and I believe you will see bugs which impact the 
> experience between first boot and first OTA are given release-blocking status 
> as a consequence.

The first few minutes after the phone comes out of the box the most important 
minutes in the life of that phone (at least for Canonical).

I realize that it’s impossible to ensure that phones have a recent image. But 
the problems I experienced apparently are unrelated to the age of the image. 
After I updated to 9.1, all the issues I saw persisted. (The one issue that 
might related to the old image was the stuck first reboot after the upgrade.)

All this is about looking at the world from the perspective of a non-technical 
person who buys the phone and living through their experience. If a customer 
goes through the same experience as me, that’s a lost customer for sure. We 
need to evaluate and test our software from the perspective of a non-technical 
user. If we don’t, we completely miss our audience.

If I can’t reliably connect to my wifi network OOB, that’s a killer bug. If I 
can’t transfer my music collection, that’s another killer bug.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Saving pictures from the phone

[Michi Henning]

> On 4 Apr 2016, at 18:51 , Michael Zanetti  
> wrote:
> 
> Hey all,
> 
> so my wife finally got fed up with Mac OS changing iTune's complete
> visuals twice per month and deleting Applications like iPhoto without
> asking her and asked me to install Ubuntu

It’s amazing how quickly Apple have managed to go from the hallmark of 
excellence in UI design to something that’s just “meh”. iTunes is probably the 
prime example of this. It once was a really efficient and usable interface that 
allowed me to slice and dice the 25000+ titles in my media collection with 
ease. These days, I get continuously annoyed at how I can’t find the stuff I’m 
interested in for all the visual garbage it insists on throwing at me. Where I 
used to be able to see lots of detail about 50 or so things at once and find 
things quickly by scrolling a little bit, I now get maybe 15 things with no 
detail whatsoever, but lots of large pictures I couldn’t care less about.

My web banking interface has gone the same way. Until recently, I had a nice 
table on screen that allowed me to see lots of transactions at once, with lots 
of detail, lots of accounts at once, lots of payment receivers at once, etc. 
Now, I get something that looks like a phone interface, and only eight or so 
things fit on the screen at a time. Just so it looks more “modern”. Eight 
things on a 30” screen is pretty good, wouldn’t you say? Especially if it is 
eight pictures that look like credit cards, five of them identical except for a 
few digits… :(

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Saving pictures from the phone

[Michi Henning]

> 
> cd ~/Pictures
> adb pull /home/phablet/Pictures
> 
> too geeky?

I’d say so, for the average consumer.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] MX4 + adb

[Michi Henning]

> On 12 Apr 2016, at 11:29 , ubuntu.me...@spamgourmet.com wrote:
> 
> I have progress to report!  Both ADB and MTP work somewhat better on my iMac 
> at work (21.5-inch, Late 2013.  My Macbook is Mid-2010.  Both running El 
> Capitan.)

I have never been able to use Android File Transfer on my Macbook Pro (early 
2013). It doesn’t work with a Nexus 4, and it doesn’t work with an Aquaris E5. 
Cables are not to blame–this is with cables that, otherwise, work fine with the 
same devices (e.g. for flashing the phone). Every time I plug in the device, 
Android File Transfer, after about 10 seconds, comes up with a message “Could 
not connect to device. Try reconnecting or restarting your device."

Is there some alternative to Android File Transfer for El Capitan?

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Scopes questions

[Michi Henning]

> 
>> It is basically up to the developer of the scope to decide how to do
>> that. You will need to block in the Query::run, hold the reference to
>> the Reply pointer, and manage asynchronous calls in the background, via
>> additional threads or event loops.
> 
> Am I correct in understanding that the results show up in the scope each time 
> reply->push is called?  If so, I could probably get nearly the same result by 
> making a series of synchronous requests with pushes in between.  Presumably 
> we'd also need to be checking if cancel had been called anywhere in there.

As Rodney said, you can do anything you like in terms of doing asynchronous 
work. If you override ScopeBase::run(), you can set up whatever async machinery 
you want when the scope is first initialized. The run time has no further 
interest in this thread, that is, it just calls ScopeBase::run() on a separate 
thread, and you can do with that thread whatever you like. Your only 
responsibility is to make sure that ScopeBase::run() returns in response to a 
call to ScopeBase::stop() in a timely manner. That's because the run time will 
try to join with the run() thread. If your run() implementation does not return 
within around 4 seconds, the registry will kill the scope process.

The other option is to implement SearchQueryBase::run() such that it keeps a 
copy of the reply proxy and returns immediately. The query is kept alive by the 
run time as long as there is at least one copy of the reply proxy still around. 
Once you drop the last reply proxy, the finished() message goes to the dash.

And, yes, results are pushed immediately and rendered immediately every time 
you call push(). Note that, because cancellation is frequent, you need to 
respond promptly to a cancel() message. This is particularly important if you 
implement run() synchronously, because there is only a single thread in the 
dispatch pool that calls run(). In other words, the next call to run() cannot 
start until after the preceding call to run() has completed. If you implement 
run() asynchronously, you can effectively have multiple queries in flight, so 
responding to cancel() quickly is less important (but still consumes 
resources): once the cancel() for a query is received by the scope, the 
scope-side run time drops all calls to push() for the cancelled query on the 
floor.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Scopes questions

[Michi Henning]

On 4 Nov 2014, at 13:19 , Robert Schroll  wrote:

> Thanks, all, for the helpful explanations.  I think I'm finally starting to 
> grok what's going on with scopes.  I need some time to digest everything 
> you've told me, but I'll be back with more questions soon.
> 
> One comment: There seems to be several reasons to prefer asynchronous 
> behavior, but the sample projects shipped with QtCreator and the tutorials on 
> developer.ubuntu.com all use synchronous requests.  A simple example using 
> asynchronous requests would be very helpful.  Obviously such an example will 
> be a bit more complicated.  But we're already dealing with a system in which 
> various things are happening in different threads.  Async requests ought to 
> fit into this existing mindset pretty easily.

There is a demo in the unity-scopes-api tree, under demo/scope-D that uses an 
async implementation. It just queues the query and returns from run(), and the 
query is processed by a background thread.

It's difficult to provide concrete examples, since the possible ways to 
implement async queries is near enough infinite, and depends heavily on how the 
backend data source API works. But I believe there is an example using Qt in 
the works for the QtCreator template.

> On Mon, Nov 3, 2014 at 4:50 PM, Rodney Dawes  
> wrote:
>> I do think 100ms is probably to fast, personally. 300ms would be much
>> better, and on par with average users.
> 
> I'm happy to file a bug about this, if you can tell me which project to file 
> it against.

That would be unity-scopes-shell.

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Scopes] Responding to preview actions

[Michi Henning]

On 7 Nov 2014, at 5:52 , Robert Schroll  wrote:

> Hi all,
> 
> I'm trying to understand how to have a scope respond to the user's activation 
> of a preview action, by running a new query for example.  This only 
> documentation I've seem on it is in the API docs, and that's not entirely 
> clear.  I don't understand the difference between ScopeBase::perform_action 
> and ScopeBase::activate, for instance.

ScopeBase::perform_action is what is called by the run time if you have 
registered one or more buttons, that is, if your preview includes a 
PreviewWidget of type "actions". You return an instance derived from 
ActivationQueryBase from perform_action. The ActivationQueryBase has an 
activate() method that you implement. The job of activate() is to tell the dash 
what to do in response to the button press. If you want to run a new query, you 
return an ActivationResponse from activate() with a status of PerformQuery, by 
constructing the ActivationResponse from a CannedQuery. The shell runs that 
query in response to the button press.

Normally, when a result is activated, the shell will use the URI of the result 
to trigger the correct action. This work if the URI schema has a handler in the 
system. If you want to override this default behavior, or if the result 
contains a schema that doesn't have a handler, you can override 
ScopeBase::activate(). If you do and, in addition, have called 
set_intercept_activation() on the result when you pushed it, the run time calls 
ScopeBase activate() instead of triggering the default action. From there, 
things are much like for previews, that is, your activate() method returns an 
ActivationQueryBase that then has its activate() method called to tell the 
shell what to do.

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Proposal for a "devel" click store

[Michi Henning]

On 13 Nov 2014, at 13:43 , Victor Thompson  wrote:

> Since the platform will support different store fronts for various providers, 
> could we create a "devel" click store so we can have a community of people 
> using and testing bleeding edge content? In the very least, it might be nice 
> to have a separate CI Dashboard that runs against such a store or image, I 
> guess, to pinpoint regressions before they are released.

Sounds like a good idea to me. That would also be useful for end-to-end testing 
of packaged scopes, so we won't end up with an incorrectly packaged scope in 
the store, only to have it fail once the user has installed it.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Scopes] Easy (?) questions

[Michi Henning]

Hi Robert,

thanks for the feedback, that is really useful!

I can only provide partial answers…

> 1) Can a setting be disabled based on the behavior of another?  I have two 
> boolean settings, but the second only has an effect when the first is 
> enabled.  Can I have the second be greyed out until the first is turned on?

No, there is no provision for this at the moment. I doubt that we'll add this 
because expressing the relevant constraints in the settings.ini file would 
rapidly get very complex.

> 2) Is there a way to mark strings in the settings.ini file as translatable, 
> and then to put those strings in the installed settings.ini file?

There is a mechanism with some tools built around gettext. I believe Pete may 
be able to comment on this. But, in general, you can localize DisplayName and 
DisplayValues in the settings.ini file directly, without any additional tool 
support. This is described here:

http://developer.ubuntu.com/scopes/tutorials/adding-settings-to-your-scope/

> 3) Is there a way to put text in the emblem, rather than an image.  Right 
> now, I'm using SVG images to fake it, but the alignment and font size is 
> noticeably off.

I don't think this is possible at the moment.

> 4) Is there any way to run an additional query to load more results when then 
> user scrolls to the bottom of the query page?  I'm purposefully not trying to 
> get all the results at first, but the pull-for-more behavior seems very 
> natural and a compliment to the pull-for-refresh at the top.

There is no support for incremental result delivery such as this. If you have 
lots of results, the shell will automatically add a "Shore more" button or some 
such, but there is no way for explicit paging that would fetch more results 
from the backend in response to a user action right now.

> 5) Is it possible to change the properties of existing results?  I have 
> preview actions that cause a change in the remote resource, but when the user 
> returns to the query page, the result corresponding to that resource has not 
> been updated.

No. The only way to get results updated is to rerun the query, possibly by 
attaching a CannedQuery to an action. This is deliberate because the aim is for 
the shell to render things as soon as possible, and any API to update a result 
"after the fact" would probably end up being rather complex.

> 6) If (5) is not possible, is it possible to mark the query results as 
> invalid, so that the remote server is hit for all the results again?

I believe you can attach a canned query URI to a preview action, in which case 
the query will be run by the shell.

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Scopes] Easy (?) questions

[Michi Henning]

On 20 Nov 2014, at 16:43 , Robert Schroll  wrote:

> On Thu, Nov 20, 2014 at 12:44 AM, Michi Henning  
> wrote:
>>> 2) Is there a way to mark strings in the settings.ini file as translatable, 
>>> and then to put those strings in the installed settings.ini file?
>> There is a mechanism with some tools built around gettext. I believe Pete 
>> may be able to comment on this. But, in general, you can localize 
>> DisplayName and DisplayValues in the settings.ini file directly, without any 
>> additional tool support. This is described here:
>> http://developer.ubuntu.com/scopes/tutorials/adding-settings-to-your-scope/
> 
> But I can't localize these things myself -- I need the translators to do 
> that.  Thus, I need to get these strings into the POT file.  And ideally, 
> dropping the resultant PO files into my repository would automatically add 
> the localized strings to to the .ini file, without me doing anything by hand.
> 
> But before I work myself into a lather about this, I'll see what Pete has to 
> say.

There is a tool to do this. You mark the entries that need localizing with an 
underscore prefix, so the tool can pick them up. The translations then come 
from catalogs. Pete should be able to give you more detail--I've never used 
this myself.

>>> 3) Is there a way to put text in the emblem, rather than an image.  Right 
>>> now, I'm using SVG images to fake it, but the alignment and font size is 
>>> noticeably off.
>> I don't think this is possible at the moment.
> 
> Is it worth submitting a bug about this?

By all means!

>>> 4) Is there any way to run an additional query to load more results when 
>>> then user scrolls to the bottom of the query page?  I'm purposefully not 
>>> trying to get all the results at first, but the pull-for-more behavior 
>>> seems very natural and a compliment to the pull-for-refresh at the top.
>> There is no support for incremental result delivery such as this. If you 
>> have lots of results, the shell will automatically add a "Shore more" button 
>> or some such, but there is no way for explicit paging that would fetch more 
>> results from the backend in response to a user action right now.
> 
> In my case, Google with throttle me if I try to request too many results at 
> once.  Having the ability to query the server only when the user needs more 
> results would be very useful.  Shall I create a bug here?

Yes, go ahead. It'll most likely end up on a wishlist though. I don't think 
we'll be able to do this quickly.

> 
>>> 6) If (5) is not possible, is it possible to mark the query results as 
>>> invalid, so that the remote server is hit for all the results again?
>> I believe you can attach a canned query URI to a preview action, in which 
>> case the query will be run by the shell.
> 
> I've done that for other things, but that will show the query right away.  
> (Or, it should.  See bug #1335761.)  I don't want to run the query now, or 
> possibly ever.  I only want it to be rerun if the user returns to the 
> previous view.
> 
> FWIW, the OnlineAccountClient::PostLoginAction::InvalidateResults seems to do 
> this, though I've never tested it from a preview page.

Pawel may be able to help here. He knows much more about those interactions 
than I do.

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Scopes] Easy (?) questions

[Michi Henning]

> 
> 2) You can't have spaces around the equal signs in translatable entries your 
> INI files.  (Because ' ?= ?' is too complicated a regex for intltool?)  The 
> default projects and all the examples use spaces.  To save other people the 
> hour I wasted tracking that down, we should remove those spaces.

Better still, intltool should accept valid .ini syntax. I've raised a bug here:

https://bugs.launchpad.net/intltool/+bug/1395561

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Scopes] Easy (?) questions

[Michi Henning]

On 24 Nov 2014, at 12:37 , Robert Schroll  wrote:

> On Sun, Nov 23, 2014 at 8:29 PM, Michi Henning  
> wrote:
>> Better still, intltool should accept valid .ini syntax.
> 
> We can always dream.

I've just pushed an MR:

https://code.launchpad.net/~michihenning/intltool/fix-ini-spaces/+merge/242608

> However, even if intltool fixes this promptly, it won't make it into existing 
> releases without an SRU.  In contrast, the project templates can be updated 
> in the PPA and make it out to most people with a minimum of bureaucracy.  The 
> online examples can be updated right away.  Let's just fix it everywhere we 
> can.

Pete, I'm not sure where all the places are where this will need fixing. Can 
you help?

Thanks,

Michi.-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Bundling scopes with webapps

[Michi Henning]

Have a look at ScopeBase::app_directory().

This returns a directory that is writable to the app and readable to the scope, 
provided they are in the same click package.

Michi.


On 20 Mar 2015, at 23:42 , nick luigi eusebio  wrote:

> Can I also ask a question? :)
> How about if a scope is bundled with a qml app and the scope gets its data 
> from the local data of the app? 
> This is like how the Tasks app/scope works.
> 
> From: Mitchell Reese 
> To: ubuntu-phone@lists.launchpad.net 
> Sent: Friday, March 20, 2015 2:27 PM
> Subject: Re: [Ubuntu-phone] Bundling scopes with webapps
> 
> Hi Vincent, it's very possible. Look at the documentation for 
> url-dispatcher, that's your link. You basically combine your scope and 
> webapp code into the same click package. 
> 
> Have a go at making a seperate webapp and scope first, then put them 
> together after. The dev community for Ubuntu Touch on google+ is a great 
> resource - I'm also happy to help if you get stuck. Best way is to tag me 
> on google+. 
> 
> Good luck!
> 
> Mitchell
> 
> On Friday, 20 March 2015 10:31:28 PM AEDT, Vincent JOBARD wrote:
> > Hi All
> > 
> > I'm very interested to do the same, but I don't really know how to generate
> > those two.
> > 
> > I want to create a webapp through the SDK and a scope through scopecreator
> > and link them together.
> > 
> > Is it possible, and how can I do it ?
> > 
> > Cheers
> >
> 
> 
> -- 
> Sent using Dekko from my Ubuntu device
> 
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to: ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help  : https://help.launchpad.net/ListHelp
> 
> 
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] BQ4.5 damage, first report.

[Michi Henning]

> > Sorry, but I keep thinking that if anyone who found a device which
> > is locked by a passcode, but SIM signed in, can accept any incoming
> > call, that this is a security issue.

Yeah, happens all the time, dunnit? Bad guy walks past just as the phone rings 
unattended, picks up the call, says "hello, who's there? Could you please tell 
me some compromising information about the owner of this phone? Name and bank 
details would be good."

That really is a serious security issue. We should immediately require 6-digit 
codes (at the very least) whenever the phone rings. Or, preferably, a 
pass-phrase. Because I'm totally happy to enter a lengthy password every time 
my phone rings. After all, I so enjoy the feeling of security that gives me. 
Never mind the occasional missed call because I couldn't get the password in 
quickly enough…

Michi.-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] Getting cpuinfo hardware field via API

[Michi Henning]

Hi folks, I'm looking for some advice for how to figure out the phone model we 
are running on.

Specifically, we are having major grief with gstreamer. On a Nexus 4, we can 
run two processes side by side that each use a single gstreamer pipeline (at 
least most of the time). On a BQ, we can run only one, otherwise, things crash, 
hang, silently deliver the wrong information without an error, etc.

So, by looking at /proc/cpuinfo, I can tell a BQ from, say, a Nexus 4. On a BQ, 
I get:

Hardware: MT6582

On a Nexus 4, I get:

Hardware: QCT APQ8064 MAKO

Now, this is the information I need, and I can set things up such that they 
work based on that.

But having to parse the output of /proc/cpuinfo sucks majorly. It's brittle and 
just awful.

I tried using QDeviceInfo (even though it's unofficial), but there don't appear 
to be any methods in there that deliver the hardware string I need.

Any suggestions as to how I can get at this information without having to parse 
/proc/cpuinfo and mucking around with regex and the like?

Thanks,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Getting cpuinfo hardware field via API

[Michi Henning]

On 24 Jun 2015, at 18:45 , Marco A. Harrendorf  wrote:

> Hello Michi,
> 
> I am not sure if it helps. However, since it seems to be a threading
> problem you could try the C++11 hardware_concurrency to find out how
> many threads are allowed. Have a look here:
> http://en.cppreference.com/w/cpp/thread/thread/hardware_concurrency

Thanks, but that doesn't fix it for me. The problem isn't how many cores there 
are (I know that anyway).
It's figuring out whether it's a BQ or not.

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Getting cpuinfo hardware field via API

[Michi Henning]

I forgot to mention.

When gstreamer does not fail, it takes 3-4 seconds to pull a single frame out 
of a 5-second video. That's on a Nexus 4.

It would be almost (but not quite) as fast to just play the video and take a 
screenshot while the video is running...

Michi.


On 24 Jun 2015, at 21:29 , Michi Henning  wrote:

> HI Simon,
> 
>> This sounds very bad and should be investigated, we can't rely on device
>> detection code in every app.
> 
> It's not an app. This is for the system-wide thumbnailer. But I agree. This 
> just has "hack" written all over it and is bad.
> 
>> CPU detection will also already fail right
>> now because
>> 
>> - the Aquaris E4.5 (krillin) and the Aquaris E5 (vegetahd) share the
>> same SoC
>> 
>> - devices with the same SoC use different device enablement packages
>> 
>> - gstreamer relies on hardware acceleration that is provided by Android
>> blobs in the device enablement package
>> 
>> So you could easily end up with having to work around an issue on
>> krillin, which is at the same time not present on vegetahd, and two
>> months later you have to update all your workarounds because all phones
>> got an update in the meantime.
> 
> Yes, the whole thing is brittle. What's at the core of it all is that 
> gstreamer is hopelessly broken on Arm, and is broken (albeit less severely) 
> on the desktop for H264 decoding. (We haven't seen issues with decoding mp3 
> files so far.)
> 
> What we are seeing is that gstreamer successfully pulls a thumbnail out of a 
> file 80 times in a row and then, on the 81st run, suddenly delivers an error 
> "negotiation failed". That's for the same physical untouched file. Other 
> times, we are seeing it hang, crash, try to allocate 24 GB of memory, 
> returning a black thumbnail at random, or hanging (at uninterruptible 
> priority, so a SIGKILL followed by waitpid() hangs). The fewer cores and the 
> less powerful a CPU we have, the worse it gets if we run more than one 
> process that uses gstreamer at a time.
> 
> But, even with only a single process using gstreamer (system-wide), we are 
> still seeing random failures.
> 
>> (Not to mention that you will not be able to ship all the correct
>> workarounds for all devices, at the end of this year there will likely
>> already be five different devices on the market.)
>> 
>> 
>> So if you want to ship your app right now, you obviously have to add a
>> workaround for krillin, but in the long term there will be no way around
>> filing all those issues against ubuntu-system-image and/or gstreamer.
> 
> The simple matter of the fact is that, if we are running on the BQ, we cannot 
> have more than one process trying to pull a thumbnail out of a video file at 
> a time. If we do, we get tons of gstreamer failures. On the Nexus 4, we can 
> run two processes at a time that each use gstreamer, and get away with it 
> most of the time. If we try to run 3 or 4, we are seeing the same avalanche 
> of failures.
> 
> So, we either come up with a reliable way to limit the number of concurrent 
> gstreamer processes on the BQ to 1, and allowing 2 concurrent gstreamer 
> processes on other Arm devices, or we have to pick the lowest common 
> denominator for all Arm devices and just say "if on Arm, run only one 
> thumbnail extractor at a time". The latter option is unpalatable because it 
> means that we are only half as fast as we could be on the Nexus and other Arm 
> devices. That's a heavy price to pay...
> 
> So, back to my question: is there a way that I can tell a BQ from everything 
> else on Arm that doesn't require parsing /proc/cpuinfo? If so, I'd love to 
> learn about it. If not, I'm afraid it'll be down to parsing the output from 
> cpuinfo until gstreamer is fixed (if ever).
> 
> Cheers,
> 
> Michi.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Getting cpuinfo hardware field via API

[Michi Henning]

HI Simon,

> This sounds very bad and should be investigated, we can't rely on device
> detection code in every app.

It's not an app. This is for the system-wide thumbnailer. But I agree. This 
just has "hack" written all over it and is bad.

> CPU detection will also already fail right
> now because
> 
> - the Aquaris E4.5 (krillin) and the Aquaris E5 (vegetahd) share the
> same SoC
> 
> - devices with the same SoC use different device enablement packages
> 
> - gstreamer relies on hardware acceleration that is provided by Android
> blobs in the device enablement package
> 
> So you could easily end up with having to work around an issue on
> krillin, which is at the same time not present on vegetahd, and two
> months later you have to update all your workarounds because all phones
> got an update in the meantime.

Yes, the whole thing is brittle. What's at the core of it all is that gstreamer 
is hopelessly broken on Arm, and is broken (albeit less severely) on the 
desktop for H264 decoding. (We haven't seen issues with decoding mp3 files so 
far.)

What we are seeing is that gstreamer successfully pulls a thumbnail out of a 
file 80 times in a row and then, on the 81st run, suddenly delivers an error 
"negotiation failed". That's for the same physical untouched file. Other times, 
we are seeing it hang, crash, try to allocate 24 GB of memory, returning a 
black thumbnail at random, or hanging (at uninterruptible priority, so a 
SIGKILL followed by waitpid() hangs). The fewer cores and the less powerful a 
CPU we have, the worse it gets if we run more than one process that uses 
gstreamer at a time.

But, even with only a single process using gstreamer (system-wide), we are 
still seeing random failures.

> (Not to mention that you will not be able to ship all the correct
> workarounds for all devices, at the end of this year there will likely
> already be five different devices on the market.)
> 
> 
> So if you want to ship your app right now, you obviously have to add a
> workaround for krillin, but in the long term there will be no way around
> filing all those issues against ubuntu-system-image and/or gstreamer.

The simple matter of the fact is that, if we are running on the BQ, we cannot 
have more than one process trying to pull a thumbnail out of a video file at a 
time. If we do, we get tons of gstreamer failures. On the Nexus 4, we can run 
two processes at a time that each use gstreamer, and get away with it most of 
the time. If we try to run 3 or 4, we are seeing the same avalanche of failures.

So, we either come up with a reliable way to limit the number of concurrent 
gstreamer processes on the BQ to 1, and allowing 2 concurrent gstreamer 
processes on other Arm devices, or we have to pick the lowest common 
denominator for all Arm devices and just say "if on Arm, run only one thumbnail 
extractor at a time". The latter option is unpalatable because it means that we 
are only half as fast as we could be on the Nexus and other Arm devices. That's 
a heavy price to pay...

So, back to my question: is there a way that I can tell a BQ from everything 
else on Arm that doesn't require parsing /proc/cpuinfo? If so, I'd love to 
learn about it. If not, I'm afraid it'll be down to parsing the output from 
cpuinfo until gstreamer is fixed (if ever).

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] Aquaris E5 owners, can you help please?

[Michi Henning]

Hi,

I'm looking for someone with an Aquaris E5. It would be really useful if you 
could help us out by logging into the phone and posting the output of

cat /proc/cpuinfo

Any takers? A free beer as payment for the effort at the next sprint! First in, 
best dressed :-)

Thanks,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Aquaris E5 owners, can you help please?

[Michi Henning]

:-)

42 isn't what I was looking for :-)

Is that really all you get from "cat /proc/cpuinfo"?

Michi.


On 25 Jun 2015, at 13:10 , Victor Thompson  wrote:

> 42.
> 
> Victor
> 
> On Wed, Jun 24, 2015 at 10:03 PM, Michi Henning  
> wrote:
> Hi,
> 
> I'm looking for someone with an Aquaris E5. It would be really useful if you 
> could help us out by logging into the phone and posting the output of
> 
> cat /proc/cpuinfo
> 
> Any takers? A free beer as payment for the effort at the next sprint! First 
> in, best dressed :-)
> 
> Thanks,
> 
> Michi.
> --
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp
> 

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Aquaris E5 owners, can you help please?

[Michi Henning]

On 25 Jun 2015, at 13:24 , Victor Thompson  wrote:

> Sorry, I don't have an E5. I just wanted to be "first in" and thought I'd 
> have some fun. :)

No, sorry. If you want a beer, you have to come up with the cpuinfo. Fair's 
fair... :-)

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Aquaris E5 owners, can you help please?

[Michi Henning]

On 25 Jun 2015, at 16:35 , Dario Cavedon  wrote:

> Hi Michi,
> here it is:

Bloody awesome, thank you so much!

Hit me up for that beer at the next sprint please!

Cheers,

Michi.-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Aquaris E5 owners, can you help please?

[Michi Henning]

On 25 Jun 2015, at 22:29 , sturmflut  wrote:

> Hi Michi,
> 
> just out of curiosity: the content of /proc/cpuinfo should be identical
> on the E4.5 and E5. Is that so?

Yes, indeed. I needed to be certain though, which is why I asked.

Cheers,

Michi.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Call for testing] New Thumbnailer

[Michi Henning]

> 
> If you are a devel, please also make sure to check the logs for your
> app in ~/.cache/upstart/

Clarification: the thumbnailer messages go into ~/.cache/upstart/dbus.log

You can run tail -f on that log to see the requests that are made to the 
thumbnailer.

> A warning will be placed there if your app is asking for an invalid
> thumbnail size (usually -1), and a full size image will be returned
> instead, which in many cases may affect the performance of your app or
> scope. If you get this warning, please adjust your code to provide the
> real image size needed instead of -1.

The warnings are written on the qml side and so go into whatever log is used by 
your app.

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Call for testing] New Thumbnailer

[Michi Henning]

On 27 Jun 2015, at 13:36 , Victor Thompson  wrote:

> Have there been any recent changes to the silo? Testing looked good
> the last time I personally checked, if there are deltas let me know
> and I'll test as well.

The main delta is that, on Arm, we are now limiting the number of concurrent 
gstreamer pipelines to 1 because anything else is too risky.
On other architectures, we use as many gstreamer pipelines as there are CPU 
cores. You should see around 780% CPU utilization on an 8-core machine with 
heavy workloads, such as thumbnailing a whole bunch of videos at once.

Once the cache is hot, it delivers the thumbnails *very* fast. Please test your 
applications not only with a cold cache, but also a hot one; the way memory 
allocations happen is markedly different that way.

If you want to clean the cache and start afresh, just run

thumbnailer-admin clear

That wipes the cache completely.

You can see cache statistics by running

thumbnailer-admin stats -v

Please keep an eye out for requests with an invalid QSize (width or height == 
-1). Get rid of all of these. Instead, ask for the thumbnail in the size you 
need it. This drives the cache in the most efficient way possible.

Your application must be prepared to handle a smaller thumbnail than what it 
asked for because we never up-scale. If original artwork is available only as 
512 and you ask for 640, the actual thumbnail you get will be 512. You are 
guaranteed that you will never get something larger than what you ask for, 
unless you ask for (0,0), in which case we deliver full size. (Don't do this 
unless there is a good chance that you will need the same image again shortly; 
otherwise, it just eats space in the cache and makes it less effective.)

Cheers,

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Call for testing] New Thumbnailer

[Michi Henning]

One more request:

please test with *large* numbers of images/photos/videos/mp3. 2000 or more is 
good start.

To avoid having to copy tons of videos, mp3s or jpg images, you can just make 
hard links to a single file. The name of the link is part of the cache key, so 
the effect is the same as having lots of separate files, and you don't have to 
copy hundreds of megabytes around that way.

Cheers,

Michi.


On 27 Jun 2015, at 13:48 , Michi Henning  wrote:

> 
> On 27 Jun 2015, at 13:36 , Victor Thompson  wrote:
> 
>> Have there been any recent changes to the silo? Testing looked good
>> the last time I personally checked, if there are deltas let me know
>> and I'll test as well.
> 
> The main delta is that, on Arm, we are now limiting the number of concurrent 
> gstreamer pipelines to 1 because anything else is too risky.
> On other architectures, we use as many gstreamer pipelines as there are CPU 
> cores. You should see around 780% CPU utilization on an 8-core machine with 
> heavy workloads, such as thumbnailing a whole bunch of videos at once.
> 
> Once the cache is hot, it delivers the thumbnails *very* fast. Please test 
> your applications not only with a cold cache, but also a hot one; the way 
> memory allocations happen is markedly different that way.
> 
> If you want to clean the cache and start afresh, just run
> 
> thumbnailer-admin clear
> 
> That wipes the cache completely.
> 
> You can see cache statistics by running
> 
> thumbnailer-admin stats -v
> 
> Please keep an eye out for requests with an invalid QSize (width or height == 
> -1). Get rid of all of these. Instead, ask for the thumbnail in the size you 
> need it. This drives the cache in the most efficient way possible.
> 
> Your application must be prepared to handle a smaller thumbnail than what it 
> asked for because we never up-scale. If original artwork is available only as 
> 512 and you ask for 640, the actual thumbnail you get will be 512. You are 
> guaranteed that you will never get something larger than what you ask for, 
> unless you ask for (0,0), in which case we deliver full size. (Don't do this 
> unless there is a good chance that you will need the same image again 
> shortly; otherwise, it just eats space in the cache and makes it less 
> effective.)
> 
> Cheers,
> 
> Michi.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Design] removing scopes

[Michi Henning]

On 16 Jul 2015, at 7:02 , Renato Filho  wrote:

> Why we have so specific scopes as part of the image? Why we can not
> make all scopes as click?

Some of the locally installed scopes have privileges that scopes installed from 
click packages cannot obtain. For example, the photo scope can look at your 
Pictures folder, which a click-installed scope cannot do.

If you don't use some local scopes, they don't do any harm, other than using a 
very small amount of disk space. Just disable the ones you don't like by 
unfavoriting them.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [Design] removing scopes

[Michi Henning]

> IMHO it would be good to know which scopes are installed locally and which 
> are remotely implemented.

Hmmm... What would you learn by knowing that? Whether a scope executes remotely 
or locally seems irrelevant. You'll get the exact same result regardless. So, 
where the CPU lives that produces the results shouldn't matter, I would think?

Cheers,

Michi.-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Application startup

[Michi Henning]

On 4 Aug 2015, at 21:17 , Arash  wrote:

> Hi.
> 
> A phone is a phone and if you don't intend to use it like a phone, you have 
> to deal with the consequences and not the phone makers.
> 
> But to me, this is a text book example of "I need a faster horse" kind of 
> problem. Making application X load faster will not solve anything. It is only 
> a symptom to another problem. I believe Improving application loading time 
> will improve the experience.

While I agree with that as a real concern, it also sounds a bit like "if I had 
all the money in the world, everything would be cheap".

Working in a constrained environment under pressure of many conflicting 
requirements is hard. Finding the most palatable trade-off through the maze is 
what makes for great design. There is no one answer to solving all the problems 
(only clever answers to particular problems). Keeping the dialer running may 
well be the right answer, as might be to not keep it running but making it 
start up really fast.

Making the right trade-off in the face of such conflicting demands is the 
hallmark of good design. Think about how little the iPhone 3G could do, and how 
much it changed the world. It did that because designers picked the compromise 
that pissed off he least number of people while providing the most value to the 
largest number of people.

If running the dialer in the background solves the problem, and if doing that 
doesn't steal too much CPU and memory, that's probably a worthwhile trade-off. 
Meanwhile, improving start-up performance across the board is a worthwhile 
thing to do too. If start-up performance improves to the point where it's no 
longer necessary to run the dialer in the background, everyone is better off 
for it, and we stop running it in the background. In the mean time, if we can 
afford it, let's run it in the background.

We need both fundamentalism and pragmatism.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] vivid+stable-phone-overlay on desktops

[Michi Henning]

On 6 Aug 2015, at 8:48 , Oliver Grawert  wrote:

> i told tsdgeos to contact sil2100 on IRC today to get the fixed xorg
> package synced, seems that didnt happen so i'm CCing Łukasz in ths mail
> additionally to get it done tomorrow ...

I just did an update, and it pulled amd64 2:1.17.2-1ubuntu0.1~overlay1, which 
seems to be the right version? (Kevin mentioned 2:1.17.1.)

Michi.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] vivid+stable-phone-overlay on desktops

[Michi Henning]

On 6 Aug 2015, at 9:19 , Oliver Grawert  wrote:

> Am Donnerstag, den 06.08.2015, 08:51 +1000 schrieb Michi Henning:
>> 
>> I just did an update, and it pulled amd64 2:1.17.2-1ubuntu0.1~overlay1, 
>> which seems to be the right version? (Kevin mentioned 2:1.17.1.)
>> 
> 
> 2:1.17.2-1ubuntu0.1~overlay1 seems to be in the PPA since last week
> (29th) ... this sounds to me like it might be the broken version kevin
> has been referring to initially since his command actually rolls you
> back to the former version. 

Ah, OK, so it needs a downgrade. Cool, thanks!

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Sources in Scopes

[Michi Henning]

To add a new data source to a scope, someone has to write code because the 
backend data sources can be anything.

There are no plans to allow what you suggest. It would, in effect mean writing 
code, no matter what sort of description language we were to choose.

Michi.


On 9 Aug 2015, at 20:54 , Krzysztof Tataradziński  wrote:

> Hello,
> I'm happy user of Ubuntu phone for a while. I see there is many problems, 
> most of them with scopes and store. Please clarify me one thing: who is 
> responsible to add new data source to scope?
> In example I want to add Endomondo, Runkeeper, etc. account to my activity 
> scope? Can I do that or I need to ask a developer of that scope to do that? 
> On which side is the problem?
>  My question is more general: are 'main' scopes (like Today, NearBy, etc.) 
> developed in a way that some third party developer can add data source by 
> itself?
> If we want to get really powerful scopes, not just rss feeds with some chosen 
> by someone data sources, we need to give third party developers some easy 
> tools for adding their data/accounts/apps into existing 'main' scopes - that 
> is my opinion.
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Ability to remove scope is an annoying behaviour

[Michi Henning]

> 
> I also did hit the star quite often by accident, over time my finger
> memory seems to have adapted to it and it doesn't happen any more but it
> is really annoying to have to re-add the scope.

The same has happened to me more than once too. Accidental activation is a real 
problem here.
Also, the current UI is inconsistent because, for some scopes, we have a menu 
that then contains the star whereas, for others, we only have the star.

Michi.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] vivid+stable-phone-overlay on desktops

[Michi Henning]

After doing an upgrade and installing xserver-xorg-core=2:1.17.1-0ubuntu3 
again, I still don't get a desktop.
The same thing happens with xserver-xort-core=2:1.17.2-1ubuntu0.1~overlay1.

Does anyone have any suggestions for how I can my desktop back?

Thanks,

Michi.


On 6 Aug 2015, at 8:25 , Kevin Gunn  wrote:

> Hi all -
> So I felt like this needed to be shared just in case.
> Some folks have added the ppa for stable-phone-overlay to their vivid desktop 
> to assist their efforts with development activities for the phone.
> 
> This has worked with no issues in recent history, however today it was 
> discovered by one of our team mates that if you're using 
> vivid+stable-phone-overlay on your desktop and you update(today) on intel you 
> may not get graphics but...
> 
> "Failed to load /usr/lib/xorg/modules/drivers/intel_drv.so:
> /usr/lib/xorg/modules/drivers/intel_drv.so:  undefined symbol: xorgMir
> in /var/log/Xorg.*"
> 
> To recover your graphics
> sudo apt-get install xserver-xorg-core=2:1.17.1-0ubuntu3
> 
> I'm on wily so I can't speak to the experience or specifics, but I wanted to 
> share just in case this might be effecting so silent but frustrated folks.
> br,kg
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] vivid+stable-phone-overlay on desktops

[Michi Henning]

On 13 Aug 2015, at 21:05 , Albert Astals Cid  
wrote:

> If you update && dist-upgrade you should get the new intel driver from
> the stable-phone-overlay that fixes it.

Thanks! I tried that, but no joy. I still get a black screen with a blinking 
cursor at the top.

I've tried both with both versions mentioned earlier, same result.

What version should I be seeing for things to work?

All this is in a VMWare Fusion VM. I've rebuilt Vmware Tools with both driver 
versions installed, but that didn't make any difference.

Is anyone else around who has Vivid with the overlay working in VMWare Fusion? 
I'm running the latest version (7.1.2).

Thanks,

Michi.


> 
> 
> On Thu, Aug 13, 2015 at 11:35 AM, Michi Henning
>  wrote:
>> After doing an upgrade and installing xserver-xorg-core=2:1.17.1-0ubuntu3
>> again, I still don't get a desktop.
>> The same thing happens with xserver-xort-core=2:1.17.2-1ubuntu0.1~overlay1.
>> 
>> Does anyone have any suggestions for how I can my desktop back?
>> 
>> Thanks,
>> 
>> Michi.
>> 
>> 
>> On 6 Aug 2015, at 8:25 , Kevin Gunn  wrote:
>> 
>> Hi all -
>> So I felt like this needed to be shared just in case.
>> Some folks have added the ppa for stable-phone-overlay to their vivid
>> desktop to assist their efforts with development activities for the phone.
>> 
>> This has worked with no issues in recent history, however today it was
>> discovered by one of our team mates that if you're using
>> vivid+stable-phone-overlay on your desktop and you update(today) on intel
>> you may not get graphics but...
>> 
>> "Failed to load /usr/lib/xorg/modules/drivers/intel_drv.so:
>> /usr/lib/xorg/modules/drivers/intel_drv.so:  undefined symbol: xorgMir
>> in /var/log/Xorg.*"
>> 
>> To recover your graphics
>> sudo apt-get install xserver-xorg-core=2:1.17.1-0ubuntu3
>> 
>> I'm on wily so I can't speak to the experience or specifics, but I wanted to
>> share just in case this might be effecting so silent but frustrated folks.
>> br,kg
>> --
>> Mailing list: https://launchpad.net/~ubuntu-phone
>> Post to : ubuntu-phone@lists.launchpad.net
>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>> More help   : https://help.launchpad.net/ListHelp
>> 
>> 
>> 
>> --
>> Mailing list: https://launchpad.net/~ubuntu-phone
>> Post to : ubuntu-phone@lists.launchpad.net
>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>> More help   : https://help.launchpad.net/ListHelp
>> 


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] vivid+stable-phone-overlay on desktops

[Michi Henning]

Thanks for your help guys!

There are no obvious errors in Xorg.0.log, but in lightdm.log, I find:

DisplayServer x-0: Can't launch Xserver X -core, not found in path
DisplayServer x-0: Xserver stopped

Any ideas?

Thanks,

Michi.


On 14 Aug 2015, at 0:05 , Łukasz 'sil2100' Zemczak 
 wrote:

> Hey,
> 
> I uploaded the intel, ati and nvidia drivers to the overlay PPA when
> working on getting this resolved. Those were the three drivers that
> seemed to be affected by the original bug.
> 
> Cheers,
> 
> W dniu 13.08.2015 o 15:15, Albert Astals Cid pisze:
>> Ah then probably you're not using the intel driver, i'd say have a
>> look at the /var/log/Xorg* and see which driver is failing, maybe we
>> need rebuilds of more drivers, myself i'm using
>> 
>> ii  xserver-xorg-core
>> 2:1.17.2-1ubuntu0.1~overlay1  amd64
>> Xorg X server - core server
>> ii  xserver-xorg-dev
>> 2:1.17.2-1ubuntu0.1~overlay1  amd64
>> Xorg X server - development files
>> ii  xserver-xorg-video-intel
>> 2:2.99.917-1~exp1ubuntu2.3~overlay1   amd64
>> X.Org X server -- Intel i8xx, i9xx display driver
>> 
>> 
>> On Thu, Aug 13, 2015 at 2:19 PM, Michi Henning
>>  wrote:
>>> 
>>> On 13 Aug 2015, at 21:05 , Albert Astals Cid  
>>> wrote:
>>> 
>>>> If you update && dist-upgrade you should get the new intel driver from
>>>> the stable-phone-overlay that fixes it.
>>> 
>>> Thanks! I tried that, but no joy. I still get a black screen with a 
>>> blinking cursor at the top.
>>> 
>>> I've tried both with both versions mentioned earlier, same result.
>>> 
>>> What version should I be seeing for things to work?
>>> 
>>> All this is in a VMWare Fusion VM. I've rebuilt Vmware Tools with both 
>>> driver versions installed, but that didn't make any difference.
>>> 
>>> Is anyone else around who has Vivid with the overlay working in VMWare 
>>> Fusion? I'm running the latest version (7.1.2).
>>> 
>>> Thanks,
>>> 
>>> Michi.
>>> 
>>> 
>>>> 
>>>> 
>>>> On Thu, Aug 13, 2015 at 11:35 AM, Michi Henning
>>>>  wrote:
>>>>> After doing an upgrade and installing xserver-xorg-core=2:1.17.1-0ubuntu3
>>>>> again, I still don't get a desktop.
>>>>> The same thing happens with 
>>>>> xserver-xort-core=2:1.17.2-1ubuntu0.1~overlay1.
>>>>> 
>>>>> Does anyone have any suggestions for how I can my desktop back?
>>>>> 
>>>>> Thanks,
>>>>> 
>>>>> Michi.
>>>>> 
>>>>> 
>>>>> On 6 Aug 2015, at 8:25 , Kevin Gunn  wrote:
>>>>> 
>>>>> Hi all -
>>>>> So I felt like this needed to be shared just in case.
>>>>> Some folks have added the ppa for stable-phone-overlay to their vivid
>>>>> desktop to assist their efforts with development activities for the phone.
>>>>> 
>>>>> This has worked with no issues in recent history, however today it was
>>>>> discovered by one of our team mates that if you're using
>>>>> vivid+stable-phone-overlay on your desktop and you update(today) on intel
>>>>> you may not get graphics but...
>>>>> 
>>>>> "Failed to load /usr/lib/xorg/modules/drivers/intel_drv.so:
>>>>> /usr/lib/xorg/modules/drivers/intel_drv.so:  undefined symbol: xorgMir
>>>>> in /var/log/Xorg.*"
>>>>> 
>>>>> To recover your graphics
>>>>> sudo apt-get install xserver-xorg-core=2:1.17.1-0ubuntu3
>>>>> 
>>>>> I'm on wily so I can't speak to the experience or specifics, but I wanted 
>>>>> to
>>>>> share just in case this might be effecting so silent but frustrated folks.
>>>>> br,kg
>>>>> --
>>>>> Mailing list: https://launchpad.net/~ubuntu-phone
>>>>> Post to : ubuntu-phone@lists.launchpad.net
>>>>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>>>>> More help   : https://help.launchpad.net/ListHelp
>>>>> 
>>>>> 
>>>>> 
>>>>> --
>>>>> Mailing list: https://launchpad.net/~ubuntu-phone
>>>>> Post to : ubuntu-phone@lists.launchpad.net
>>>>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>>>>> More help   : https://help.launchpad.net/ListHelp
>>>>> 
>>> 
>> 
> 
> 
> -- 
> Łukasz 'sil2100' Zemczak
> Foundations Team
> lukasz.zemc...@canonical.com
> www.canonical.com
> 
> -- 
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Best means for getting photos OFF Meizu MX4 Ubuntu Edition?

[Michi Henning]

I’ve tried this with the Nexus 4 and never got it to work. Seems like it 
depends on what device you have.

Cheers,

Michi.


> On 10 Jun 2016, at 13:33 , Vic G  wrote:
> 
> Hi Bob!
> 
> If you want browse and transfer files from any Ubuntu/Android device(both 
> internal memory and SD card) to your Mac OS try this:
> 
> https://www.android.com/filetransfer/  
> 
> I've tested it with bq E4.5/E5 and M10 and works like a charm.
> 
> Regards,
> 
> V
> 
> El 09/06/16 a las 08:01, Bob Summerwill escribió:
>> 
>> This is probably of interest only to a very small number of people on this 
>> list, but this evening I have accidentally discovered that the Meizu MX4 
>> Ubuntu Edition's USB seems to work really well with Windows 10 on a Surface 
>> Pro 3.
>> 
>> https://bobsummerwill.com/2016/06/08/windows-10-nuke-and-pave/ 
>> 
>> 
>> This was a completely accidental discovery, as I plugged my MX4 into this 
>> laptop to charge its battery.This isn't my primary laptop, just one 
>> which I am using purely for Windows builds which I need to do as part of my 
>> work. But the media drive just popped up and worked a treat.
>> 
>> Just another interesting data-point!This same device has never really 
>> worked at all when connecting to my MacBook Air.
>> 
>> I will be retiring the MX4 imminently anyway, because I have just taken 
>> delivery of an MX5 Pro :-)
>> 
>> On Mon, May 9, 2016 at 9:40 AM, Michael Zanetti 
>> mailto:michael.zane...@canonical.com>> wrote:
>> 
>> 
>> On 09.05.2016 17:36, Tony Espy wrote:
>> > On 05/09/2016 10:27 AM, Michael Zanetti wrote:
>> >>
>> >>
>> >> On 09.05.2016 15:14, Oliver Grawert wrote:
>> >>> hi,
>> >>>
>> >>> Am Montag, 9. Mai 2016 02:00:20 PDT schrieb Michael Zanetti
>> >>> mailto:michael.zane...@canonical.com>>:
>>  Hmm... The fact that people seem to recommend ssh and even adb over
>>  MTP,
>>  indicates that something seems wrong with our MTP support.
>> 
>> >>>
>> >>> nah, it is rather the fact that USB is unreliable on the MX4 ...
>> >>
>> >> It's not working properly for me nor my wife, both on krillin.
>> >
>> > There may indeed be issues with our MTP support.  Have you filed a bug
>> > for the issues you and your wife see with krillin?
>> 
>> Hey Tony, yes I did:
>> 
>> https://bugs.launchpad.net/canonical-devices-system-image/+bug/1565675 
>> 
>> 
>> Please let me know if you need additional logs. Happy to help with testing.
>> 
>> >
>> > That said, the MX4 USB hardware is known to be flaky.  It's hard to have
>> > solid MTP support with an unreliable transport.
>> 
>> 
>> Yes, my MX4's USB port died completely. But no, all the issues I am
>> seeing with MTP are happening with krillin devices.
>> 
>> Br,
>> Michael
>> 
>> 
>> --
>> Mailing list: https://launchpad.net/~ubuntu-phone 
>> 
>> Post to : ubuntu-phone@lists.launchpad.net 
>> 
>> Unsubscribe : https://launchpad.net/~ubuntu-phone 
>> 
>> More help   : https://help.launchpad.net/ListHelp 
>> 
>> 
>> 
>> 
>> 
>> -- 
>>  b...@summerwill.net 
>> 
>> 
>> 
> 
> -- 
> 
> Víctor González Troncoso
> Área de Postventa
> Soporte SAUX
> +34 91 142 06 82 <> Ext: 2003
> Marie Curie, 19 Bloque 2 Planta 6 Oficina 6
> 28521 Rivas Vaciamadrid (Madrid)
> bq.com
> 
>  
>    
>   
>   
>   
>    
>  Nos encantan los árboles. No me imprimas si no es necesario.
> 
> Protección de Datos: Mundo Reader S.L. le informa de que los datos personales 
> facilitados por Ud. y utilizados para el envío de esta comunicación serán 
> objeto de tratamiento automatizado o no en nuestros ficheros, con la 
> finalidad de gestionar la agenda de contactos de nuestra empresa y para el 
> envío de comunicaciones profesionales por cualquier medio electrónico o no. 
> Puede consultar en www.bq.com  los detalles de nuestra 
> Política de Privacidad y dónde ejercer el derecho de acceso, rectificación, 
> cancelación y oposición.
> Confidencialidad: Este mensaje contiene material confidencial y está dirigido 
> exclusivamente a su destinatario. Cualquier revisión, modificación o 
> distribución por otras personas, así como su reenvío sin el consentimiento 
> expreso está estrictamente prohibido. Si usted no es el destinatario del 
> mensaje, por favor, comuníqueselo al emisor y borre todas las copias de forma 
> inmediata. Confidentiality: This e-mail contains material that is 
> confidential for de sole use of de intend

Re: [Ubuntu-phone] Ubuntu Touch OTA-13 released

[Michi Henning]

> 
> - I would appreciate maybe a short overview of what storage framework is all 
> about. My initial idea is for cloud syncing.

The storage framework is basically a way to provide access to arbitrary cloud 
storage services, such as Owncloud or Dropbox, with a single API. It’s built as 
a client-server library that provides a storage access API that clients can use 
to upload and download files, navigate the contents of the cloud store, etc. 
There is also a server-side API that allow new providers to be added to the 
machinery. Basically, a provider implementation is a piece of adapter code that 
receives requests for file access from the storage framework runtime and then 
makes corresponding requests using the cloud service’s REST API. We also plan 
to add a local filesystem provider.

With the storage framework, higher-level services, such as backup or syncing, 
can be layered on top.

It’s early days yet. The client-side API is still going through changes, and 
the server-side API may also still end up with some tweaks. Consider it alpha 
code for now.

Cheers,

Michi.-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] free and open source OS and personal computing

[Michi Henning]

> After the news that Canonical is abandoning unity the following became clear 
> to me. Free and open source OS on personal computing devices will never 
> became a success.

Depends on how you define “success”. If “success” is synonymous with “making a 
profit”, you are probably mostly right. It’s hard to create a business model 
around open source. There are a few companies who’ve managed to do it but, by 
and large, something that is available free of charge, pretty much be 
definition, has no monetary value. So, if I want to make a business out of open 
source, I have to find some way to add value, otherwise no-one will buy what 
I’m selling.

But there are other measures of “success”. For example, if we were to count all 
the installed systems world-wide that intrinsically rely on some open source 
components to function, I suspect we’d find that the open source “participation 
rate" is close to 100%. Open source has indeed changed the computing landscape 
in a fundamental and dramatic way, because it has enabled closed source 
software to be developed so much more cheaply and efficiently.

> Only technical users and nerds benefit from free and open source OS.

I strongly disagree with that. There are tons of companies that make a very 
good living out of software that is largely built on top of LGPL open source. 
It’s not that long ago that companies had to pay per developer seat for a 
compiler, for example. But, what’s behind that isn’t so much that the source is 
open, but that it doesn’t cost anything. The argument that open source is a 
safer choice because you get the source code is mostly a red herring. If I’m a 
vendor of accounting software, the source code for gcc does me no good 
whatsoever because, if there is a bug in gcc, I’ll have neither the skills nor 
the inclination to learn enough about how gcc works to fix a bug or add a 
feature.

Ordinary people benefit from open source every day. The router in my home runs 
some open source. My TV does. My solar system does. Virtually ever web server 
on the planet does. Etc, etc. And one of the reasons why these devices are as 
cheap as they are is that the vendors didn’t have to spend years of research 
and development to create all that code themselves.

Michi.
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp