State of Bitbucket Repsitories

[Gregory Nutt]

I am continuing to accept changes to the (deprecated) NuttX repositories 
on Bitbucket.org.  Can someone please inform me when INFRA is ready to 
instantiate the Apache Github repositories?  I will then make the the 
Bitbucket repositories read-only and prohibit forks and PRs.  Let's try 
to make that transition as "knife-edge" as possible.


Greg

Re: State of Bitbucket Repsitories

[Flavio Junqueira]

I can see this INFRA issue for NuttX:

https://issues.apache.org/jira/browse/INFRA-19542 


But that does not talk about creating a gitbox repo.

Justin, do we need to create a new INFRA ticket?

-Flavio

> On 10 Dec 2019, at 15:25, Gregory Nutt  wrote:
> 
> I am continuing to accept changes to the (deprecated) NuttX repositories on 
> Bitbucket.org.  Can someone please inform me when INFRA is ready to 
> instantiate the Apache Github repositories?  I will then make the the 
> Bitbucket repositories read-only and prohibit forks and PRs.  Let's try to 
> make that transition as "knife-edge" as possible.
> 
> Greg
> 
> 

Re: State of Bitbucket Repsitories

[Nathan Hartman]

On Tue, Dec 10, 2019 at 9:34 AM Gregory Nutt  wrote:

> I am continuing to accept changes to the (deprecated) NuttX repositories
> on Bitbucket.org.  Can someone please inform me when INFRA is ready to
> instantiate the Apache Github repositories?  I will then make the the
> Bitbucket repositories read-only and prohibit forks and PRs.  Let's try
> to make that transition as "knife-edge" as possible.


After the bit bucket repos are made read only, I suggest to keep them
available indefinitely, both for historical reference and for anyone who
wishes to create something based on the original BSD-licensed code.

Cheers,
Nathan

Re: State of Bitbucket Repsitories

[Gregory Nutt]

I am continuing to accept changes to the (deprecated) NuttX
repositories
on Bitbucket.org.  Can someone please inform me when INFRA is
ready to
instantiate the Apache Github repositories?  I will then make the the
Bitbucket repositories read-only and prohibit forks and PRs. 
Let's try
to make that transition as "knife-edge" as possible.


After the bit bucket repos are made read only, I suggest to keep them 
available indefinitely, both for historical reference and for anyone 
who wishes to create something based on the original BSD-licensed code.


I will follow the recommendation of vote of the PPMC.  My plan was to 
remove them after a grace period to:  make it more difficult for people 
to create something based on the original BSD-licensed code and instead 
encourage people to use the Apache licensed code, and , more 
importantly, to make it more difficult to fork the BSD licensed code 
under a different, license that is incompatible with Apache.  Yours and 
mine are two differing positions.  I will follow the will of the PPMC in 
all things.

Re: State of Bitbucket Repsitories

[Nathan Hartman]

On Tue, Dec 10, 2019 at 10:16 AM Gregory Nutt  wrote:
>
>
>> I am continuing to accept changes to the (deprecated) NuttX repositories
>> on Bitbucket.org.  Can someone please inform me when INFRA is ready to
>> instantiate the Apache Github repositories?  I will then make the the
>> Bitbucket repositories read-only and prohibit forks and PRs.  Let's try
>> to make that transition as "knife-edge" as possible.
>
>
> After the bit bucket repos are made read only, I suggest to keep them 
> available indefinitely, both for historical reference and for anyone who 
> wishes to create something based on the original BSD-licensed code.
>
> I will follow the recommendation of vote of the PPMC.  My plan was to remove 
> them after a grace period to:  make it more difficult for people to create 
> something based on the original BSD-licensed code and instead encourage 
> people to use the Apache licensed code, and , more importantly, to make it 
> more difficult to fork the BSD licensed code under a different, license that 
> is incompatible with Apache.  Yours and mine are two differing positions.  I 
> will follow the will of the PPMC in all things.

Fair enough; it's just a suggestion. :-)

Nathan

Re: State of Bitbucket Repsitories

[Matias N.]

Hi,
just wanted to mention that there's also an automatic mirror on Gitlab at:

https://gitlab.com/nuttx

Best,
Matias

On Tue, Dec 10, 2019, at 12:24, Nathan Hartman wrote:
> On Tue, Dec 10, 2019 at 10:16 AM Gregory Nutt  wrote:
> >
> >
> >> I am continuing to accept changes to the (deprecated) NuttX repositories
> >> on Bitbucket.org. Can someone please inform me when INFRA is ready to
> >> instantiate the Apache Github repositories? I will then make the the
> >> Bitbucket repositories read-only and prohibit forks and PRs. Let's try
> >> to make that transition as "knife-edge" as possible.
> >
> >
> > After the bit bucket repos are made read only, I suggest to keep them 
> > available indefinitely, both for historical reference and for anyone who 
> > wishes to create something based on the original BSD-licensed code.
> >
> > I will follow the recommendation of vote of the PPMC. My plan was to remove 
> > them after a grace period to: make it more difficult for people to create 
> > something based on the original BSD-licensed code and instead encourage 
> > people to use the Apache licensed code, and , more importantly, to make it 
> > more difficult to fork the BSD licensed code under a different, license 
> > that is incompatible with Apache. Yours and mine are two differing 
> > positions. I will follow the will of the PPMC in all things.
> 
> Fair enough; it's just a suggestion. :-)
> 
> Nathan
> 

Re: State of Bitbucket Repsitories

[Gregory Nutt]

just wanted to mention that there's also an automatic mirror on Gitlab at:

https://gitlab.com/nuttx

There are dozens of mirrors of the NuttX Bitbucket repositories. Google 
nuttx+mirror to see some of them.


An alternative disposition of the Bitbucket repositories might be to 
make them manual mirrors of the Apache repositories.  AFAIK Bitbucket 
does not support automated mirroring without a plug-in. But periodic 
manually mirroring of the Apache repositories would help by assuring 
that the Apache licensed code is pushed out to all NuttX mirrors as well.


I really don't know what is the best thing to do.  I think, ideally, 
every user and every mirror should transition to the Apache code and 
that there should not be some creeping unsupported BSD version.  If 
someone decides to start supporting that stranded BSD version, then you 
would have a BSD fork to deal with.  And trademark issues too.  There 
cannot be a different OS called NuttX; there can be only one.  I just 
cannot imagine a positive scenario with retaining an unsupported BSD 
version of the RTOS.  I doubt the ASF would appreciate that either?


Still just brainstorming..

Re: State of Bitbucket Repsitories

[Nathan Hartman]

On Tue, Dec 10, 2019 at 10:43 AM Gregory Nutt  wrote:
>
>
> > just wanted to mention that there's also an automatic mirror on Gitlab at:
> >
> > https://gitlab.com/nuttx
> >
> There are dozens of mirrors of the NuttX Bitbucket repositories. Google
> nuttx+mirror to see some of them.
>
> An alternative disposition of the Bitbucket repositories might be to
> make them manual mirrors of the Apache repositories.  AFAIK Bitbucket
> does not support automated mirroring without a plug-in. But periodic
> manually mirroring of the Apache repositories would help by assuring
> that the Apache licensed code is pushed out to all NuttX mirrors as well.
>
> I really don't know what is the best thing to do.  I think, ideally,
> every user and every mirror should transition to the Apache code and
> that there should not be some creeping unsupported BSD version.  If
> someone decides to start supporting that stranded BSD version, then you
> would have a BSD fork to deal with.  And trademark issues too.  There
> cannot be a different OS called NuttX; there can be only one.  I just
> cannot imagine a positive scenario with retaining an unsupported BSD
> version of the RTOS.  I doubt the ASF would appreciate that either?
>
> Still just brainstorming..

I think we need to ask our mentors what is the generally accepted
thing to do with the project's earlier incarnation.

Nathan

Re: State of Bitbucket Repsitories

[Gregory Nutt]

I think we need to ask our mentors what is the generally accepted
thing to do with the project's earlier incarnation.


100% agreed.  We need guidance.

Re: State of Bitbucket Repsitories

[Flavio Junqueira]

I don't think you need to get rid of the earlier incarnation of the project. As 
per the ASF legal documentation, we need to make sure that the copyright 
holders sign either CLAs or software grants:

https://www.apache.org/foundation/how-it-works/legal.html 


This incubator document also talks about the initial code import:

https://incubator.apache.org/guides/transitioning_asf.html 


Finally, as an example, we still have the zookeeper code in sourceforge from 
before Apache, and I don't that's wrong:

https://sourceforge.net/p/zookeeper/code/ 


-Flavio

> On 10 Dec 2019, at 16:50, Gregory Nutt  wrote:
> 
> 
>> I think we need to ask our mentors what is the generally accepted
>> thing to do with the project's earlier incarnation.
> 
> 100% agreed.  We need guidance.
> 
> 

Re: State of Bitbucket Repsitories

[Gregory Nutt]

I don't think you need to get rid of the earlier incarnation of the project. ...


Skimming the documents, it appears that ASF has no position on the 
disposition of the earlier incarnations of the.  I think that 
maintaining them as mirrors of the Apache code would be best option for 
the community, for the NuttX trademark, and some protection against a 
fork with a different, incompatible license. That way, there is only one 
NuttX, Apache NuttX.


Certainly there are other forks of NuttX all over the internet (not 
mirrors, but usually older clones), but I think people would always want 
to go to the authoritative source and currently the world believes that 
the Bitbucket repository is the authoritative source.  Hopefully that 
will change over time.

Re: State of Bitbucket Repsitories

[Gregory Nutt]

I don't think you need to get rid of the earlier incarnation of the 
project. ...


Skimming the documents, it appears that ASF has no position on the 
disposition of the earlier incarnations of the.  I think that 
maintaining them as mirrors of the Apache code would be best option 
for the community, for the NuttX trademark, and some protection 
against a fork with a different, incompatible license. That way, there 
is only one NuttX, Apache NuttX.


Certainly there are other forks of NuttX all over the internet (not 
mirrors, but usually older clones), but I think people would always 
want to go to the authoritative source and currently the world 
believes that the Bitbucket repository is the authoritative source.  
Hopefully that will change over time.
Certainly, there is no benefit to keep an old static version of the last 
of the earlier incarnation.  Anyone can recover that at will by checking 
out that last version from GIT.  If we are smart, we can tag the initial 
Apache import to make that easier.

Re: [Test] Joined

[Disruptive Solutions]

[Test] Joined

Verstuurd vanaf mijn iPhone

> Op 10 dec. 2019 om 01:00 heeft david.sidr...@gmail.com het volgende 
> geschreven:
> 
>  [Test] Joined
> 
> Is it alive?
> 
> David

Re: State of Bitbucket Repsitories

[Alan Carvalho de Assis]

Hi Greg,

On 12/10/19, Gregory Nutt  wrote:
>>
>> Certainly there are other forks of NuttX all over the internet (not
>> mirrors, but usually older clones), but I think people would always
>> want to go to the authoritative source and currently the world
>> believes that the Bitbucket repository is the authoritative source.
>> Hopefully that will change over time.
> Certainly, there is no benefit to keep an old static version of the last
> of the earlier incarnation.  Anyone can recover that at will by checking
> out that last version from GIT.  If we are smart, we can tag the initial
> Apache import to make that easier.
>

What if we could keep the bitbucket for a while and use the git
"post-update" to updated the clone repository to the new apache
repository?

More info:

https://stackoverflow.com/questions/19434605/git-redirections

https://confluence.atlassian.com/bitbucketserver/using-repository-hooks-776639836.html

What do you think?

If you don't agree, then as an way to be helpful for new users, I
think it should better only removing the repository content and
putting a README file instructing the user to clone the new apache
repository.

BR,

Alan

Re: State of Bitbucket Repsitories

[Alan Carvalho de Assis]

The right bitbucket url for hooks:

https://www.atlassian.com/git/tutorials/git-hooks

On 12/10/19, Alan Carvalho de Assis  wrote:
> Hi Greg,
>
> On 12/10/19, Gregory Nutt  wrote:
>>>
>>> Certainly there are other forks of NuttX all over the internet (not
>>> mirrors, but usually older clones), but I think people would always
>>> want to go to the authoritative source and currently the world
>>> believes that the Bitbucket repository is the authoritative source.
>>> Hopefully that will change over time.
>> Certainly, there is no benefit to keep an old static version of the last
>> of the earlier incarnation.  Anyone can recover that at will by checking
>> out that last version from GIT.  If we are smart, we can tag the initial
>> Apache import to make that easier.
>>
>
> What if we could keep the bitbucket for a while and use the git
> "post-update" to updated the clone repository to the new apache
> repository?
>
> More info:
>
> https://stackoverflow.com/questions/19434605/git-redirections
>
> https://confluence.atlassian.com/bitbucketserver/using-repository-hooks-776639836.html
>
> What do you think?
>
> If you don't agree, then as an way to be helpful for new users, I
> think it should better only removing the repository content and
> putting a README file instructing the user to clone the new apache
> repository.
>
> BR,
>
> Alan
>

Re: State of Bitbucket Repsitories

[Gregory Nutt]

The right bitbucket url for hooks:

https://www.atlassian.com/git/tutorials/git-hooks


I don't think I can trigger any scripts in the Bitbucket repository 
based on events in a Github repository.


There is a mirror plug-in for Bitbucket: 
https://marketplace.atlassian.com/apps/1211351/repository-mirror-plugin-for-bitbucket?hosting=server&tab=overview


But I think it is only usable if you have Bitbucket on your own server.  
Also, I see only references to support mirroring of Bitbucket 
repositories on other remotes.

Re: State of Bitbucket Repsitories

[Nathan Hartman]

On Tue, Dec 10, 2019 at 1:07 PM Gregory Nutt  wrote:

> > If we are smart, we can tag the initial
> > Apache import to make that easier.
>
+1 to tagging the initial import

Nathan

RE: State of Bitbucket Repsitories

[David Sidrane]

The history will be intact no matter where it is. It will remain BSD until
the license changes. The change should be on a branch than a single commit.
Then it is a knife edge


-Original Message-
From: Gregory Nutt [mailto:spudan...@gmail.com]
Sent: Tuesday, December 10, 2019 7:44 AM
To: dev@nuttx.apache.org
Subject: Re: State of Bitbucket Repsitories


> just wanted to mention that there's also an automatic mirror on Gitlab at:
>
> https://gitlab.com/nuttx
>
There are dozens of mirrors of the NuttX Bitbucket repositories. Google
nuttx+mirror to see some of them.

An alternative disposition of the Bitbucket repositories might be to
make them manual mirrors of the Apache repositories.  AFAIK Bitbucket
does not support automated mirroring without a plug-in. But periodic
manually mirroring of the Apache repositories would help by assuring
that the Apache licensed code is pushed out to all NuttX mirrors as well.

I really don't know what is the best thing to do.  I think, ideally,
every user and every mirror should transition to the Apache code and
that there should not be some creeping unsupported BSD version.  If
someone decides to start supporting that stranded BSD version, then you
would have a BSD fork to deal with.  And trademark issues too.  There
cannot be a different OS called NuttX; there can be only one.  I just
cannot imagine a positive scenario with retaining an unsupported BSD
version of the RTOS.  I doubt the ASF would appreciate that either?

Still just brainstorming..

Candidate Committers

[Gregory Nutt]

Given that we want to open up the first committer spots to all 
significant NuttX contributors, I made a list.  I went through the top 
20 or so all time contributors to NuttX and made a list of potential 
committers.  I have excluded already identified initial committers.  I 
have also excluded anyone who has not contributed in a long time (i.e., 
essentially former NuttX users).


I have already contacted some people who were not interested in 
participating on the PMC or being committers.   Of the people I have 
contacted, the following have an interest in being a committer and/or a 
member of the PPMC.


   Masayuki Ishikawa 
   Kenneth Pettit 
   Brennan Ashton 

There are eight more people whom I have not yet contacted.   I have made 
no concerted effort to contact them individually, I am having the 
relevant conversations as targets of opportunity.  I prefer not to name 
them here without first contacting them.


I understand that for some projects, all committers are PPMC members, 
and other projects PPMC members are only a subset of the committers.  I 
leave this to the PPMC.


I did make a commitment to keep a low profile while the PPMC is forming 
and I will do that.  But there a a few topics that I must hand-off to 
the PPMC (like this one) and a few that I must deal with myself (like 
the hand-off of the Bitbucket repositories to INFRA).


Greg

Re: Candidate Committers

[Alan Carvalho de Assis]

HI Greg,

I think these guys are good options and it is very nice to see them aboard!

BR,

Alan

On 12/10/19, Gregory Nutt  wrote:
> Given that we want to open up the first committer spots to all
> significant NuttX contributors, I made a list.  I went through the top
> 20 or so all time contributors to NuttX and made a list of potential
> committers.  I have excluded already identified initial committers.  I
> have also excluded anyone who has not contributed in a long time (i.e.,
> essentially former NuttX users).
>
> I have already contacted some people who were not interested in
> participating on the PMC or being committers.   Of the people I have
> contacted, the following have an interest in being a committer and/or a
> member of the PPMC.
>
> Masayuki Ishikawa 
> Kenneth Pettit 
> Brennan Ashton 
>
> There are eight more people whom I have not yet contacted.   I have made
> no concerted effort to contact them individually, I am having the
> relevant conversations as targets of opportunity.  I prefer not to name
> them here without first contacting them.
>
> I understand that for some projects, all committers are PPMC members,
> and other projects PPMC members are only a subset of the committers.  I
> leave this to the PPMC.
>
> I did make a commitment to keep a low profile while the PPMC is forming
> and I will do that.  But there a a few topics that I must hand-off to
> the PPMC (like this one) and a few that I must deal with myself (like
> the hand-off of the Bitbucket repositories to INFRA).
>
> Greg
>
>

Re: Candidate Committers

[Ishikawa, Masayuki (SHES)]

Dear Greg,

Thanks for adding me to the list.

On 2019/12/11 9:41, "Gregory Nutt"  wrote:

Given that we want to open up the first committer spots to all 
significant NuttX contributors, I made a list.  I went through the top 
20 or so all time contributors to NuttX and made a list of potential 
committers.  I have excluded already identified initial committers.  I 
have also excluded anyone who has not contributed in a long time (i.e., 
essentially former NuttX users).

I have already contacted some people who were not interested in 
participating on the PMC or being committers.   Of the people I have 
contacted, the following have an interest in being a committer and/or a 
member of the PPMC.

Masayuki Ishikawa 
Kenneth Pettit 
Brennan Ashton 

There are eight more people whom I have not yet contacted.   I have made 
no concerted effort to contact them individually, I am having the 
relevant conversations as targets of opportunity.  I prefer not to name 
them here without first contacting them.

I understand that for some projects, all committers are PPMC members, 
and other projects PPMC members are only a subset of the committers.  I 
leave this to the PPMC.

I did make a commitment to keep a low profile while the PPMC is forming 
and I will do that.  But there a a few topics that I must hand-off to 
the PPMC (like this one) and a few that I must deal with myself (like 
the hand-off of the Bitbucket repositories to INFRA).

Greg

Re: Candidate Committers

[Duo Zhang]

Hi, Greg. In ASF, usually the discussion about the committer candidates
should happen in the private@nuttx mailing list...

And also, the later vote for the new PPMC members and committers should
also happen in the private mailing list.

Thanks.

Ishikawa, Masayuki (SHES)  于2019年12月11日周三
上午9:05写道：

> Dear Greg,
>
> Thanks for adding me to the list.
>
> On 2019/12/11 9:41, "Gregory Nutt"  wrote:
>
> Given that we want to open up the first committer spots to all
> significant NuttX contributors, I made a list.  I went through the top
> 20 or so all time contributors to NuttX and made a list of potential
> committers.  I have excluded already identified initial committers.  I
> have also excluded anyone who has not contributed in a long time
> (i.e.,
> essentially former NuttX users).
>
> I have already contacted some people who were not interested in
> participating on the PMC or being committers.   Of the people I have
> contacted, the following have an interest in being a committer and/or
> a
> member of the PPMC.
>
> Masayuki Ishikawa 
> Kenneth Pettit 
> Brennan Ashton 
>
> There are eight more people whom I have not yet contacted.   I have
> made
> no concerted effort to contact them individually, I am having the
> relevant conversations as targets of opportunity.  I prefer not to
> name
> them here without first contacting them.
>
> I understand that for some projects, all committers are PPMC members,
> and other projects PPMC members are only a subset of the committers.
> I
> leave this to the PPMC.
>
> I did make a commitment to keep a low profile while the PPMC is
> forming
> and I will do that.  But there a a few topics that I must hand-off to
> the PPMC (like this one) and a few that I must deal with myself (like
> the hand-off of the Bitbucket repositories to INFRA).
>
> Greg
>
>
>
>

Re: Candidate Committers

[Gregory Nutt]

Hi, Greg. In ASF, usually the discussion about the committer candidates
should happen in the private@nuttx mailing list...

And also, the later vote for the new PPMC members and committers should
also happen in the private mailing list.


Okay... thanks!  Still learning how things work.

Greg

Re: Candidate Committers

[Justin Mclean]

Hi,

> Given that we want to open up the first committer spots to all significant 
> NuttX contributors, I made a list.  I went through the top 20 or so all time 
> contributors to NuttX and made a list of potential committers.  I have 
> excluded already identified initial committers.  I have also excluded anyone 
> who has not contributed in a long time (i.e., essentially former NuttX users).

This lis a little unusual, I'll explain why on the private list.

Thanks,
Justin

Update LittleFS to version 2.1.4

[Xiang Xiao]

Many feature is added, here has the detailed information:
https://github.com/ARMmbed/littlefs/releases

Should we use dev mail list for patch until the github workflow is ready?

Thanks
Xiang