Thanks, but I will revert to Pharo 5.0 (pre spur VM)  again...
until the problems with 7.x are solved... or 8 is really stable.

I was regularly telling colleagues with whom I converse on Twitter
(e.g. Swift, C# and other programmers) 
that they should try Smalltalk, especially Pharo,
to see for themselves that programming in Smalltalk 
has many advantages, which however only become 
obvious when actually using Smalltalk. 
So I wrote tweets like
"try it for a week, and you'll be hooked" etc.
So people perhaps downloaded Pharo, but then
to get frozen in a mere minute..

if the *first experience* of those new to Smalltalk
is a freezing Pharo immediately after going full screen
or moving the window.
they might (probably will) think 
"ok, agreed, nothing is perfect, but so much for Smalltalk then, wasting my
time" 
never to return to Smalltalk again.  
Of course, that's really not good at all.

How can one promote Smalltalk if this kind of things happen?
Please note that this is not meant as a negative comment,
because I know that there are not many people
working on Pharo, but those that do, do the best they can!
However, it is unfortunately the status quo.
Very bad, because it bounces away those
who would really like to try Pharo
and might become enthusiastic Smalltalkers.

"Under the hood", need to know? 

New users know nearly nothing about VMs and how they
work. They shouldn't have to.
Let alone hacking VMs etc. to get Pharo going.
(respect btw for those building VMs, this is
not trivial, but this is exactly the reason why it is not a "user thing")

(I am profiling myself here as a typical Smalltalk
user, application programmer. 
Not a tool builder, I know almost nothing about VMs
(Do I have to?) and the deepest system related classes
in the Smalltalk deep sea. Initially, I live on the surface, and
only dive deeper when needed, so to speak.
e.g. looking for base classes like Morph.
If I go deeper, I might run out of mental oxygen :o) 
It is nearly impossible to know all classes living
in a Smalltalk image. (Seems this ocean metaphor is a 
nice one, we know more about (the surface of) the moon
 than about what is in the the depths of our oceans :o) )

All they (should care about) is learning/using
Smalltalk from an application programmer's perspective.

Perhaps it might be a good idea
to make it clear on the Pharo website
that version 5.0 (non spur VM) is the most recent stable version.
when using Pharo for safe reliable application building.
Focus on stable releases, add new things only after that.
(this might sound not too academic, 
but down-to-earth industrial, and yes it is) 

And    Please   Keep   It  Simple: 
just a no frills one-click-image will do.
no hassle with launchers, loaders etc. 
For example you can still download
Blender (www.blender.org)
A large 3d application in a zip file
which is a single complete app directory with everything in it.

Thanks
TedvG






Ben Coman wrote
> On Tue, 30 Apr 2019 at 17:44, Stephan Eggermont <

> stephan@

> > wrote:
>>
>> TedVanGaalen <

> tedvga@

> > wrote:
>> > Crying victory too soon:
>> > Alas, it happened again pharo screen frozen,
>> > partly drawn, when going to full screen
>> > ( see image.)
>> > frustrating.
>> > The strange thing however:
>> >  the Seaside/Zinc server
>> >  kept running that is: I could enter  and save data in
>> >  in my test app running in the web browser!
>> > No idea why.
> 
> If you hit 
> <F2>
>  do you see an option for Debug Options > Dump XXX
> maybe(?) provide some extra info.
> 
>>
>> That is consistent with other reports. Looks like the event loop
>> receiving
>> keyboard and mouse events gets stopped and not restarted. The http server
>> runs in its own process
> 
> Its been a long time since I've played with Seaside.
> IIUC from Seaside you can get a console-like interface to execute
> commands(?)
> Maybe try...  WorldMorph installNewWorld
> 
> Also maybe try commenting out the call to #interCyclePause:.
> (btw, your CPU will max out)
> 
> cheers -ben





--
Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Reply via email to