almost 16 months

wxPython Users
1

Hi Everyone,

It’s been almost 16 months since we had an update. I don’t see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don’t see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.

So I think it is fair to say “we are very close to one more dead open source project”.

I do not want to see this die. I’m willing to help with some money (I don’t have the skills to write C or C ++ code). I have limited resources but I can help with some cash.

But first we need to find someone willing to take over with the skills (even if we have the money). I know many believe the desktop program is dead. But it’s not - with the web being hacked daily and mobile apps not suited to data entry - the desktop is NOT dead. GitHub just came out with Electron - because they believe the desktop is not dead.

We could just move to other tech and say bye to wxPython but what about the investment we all made. If wxPython would just moved to Python 3 we all could get many more years just using wxPython.

Please guys don’t let this die. Let’s do something - organize a way to fund a programmer and move forward. Let’s not lose our investment in wxPython.

Johnf

2

I think the way forward would be to actually get a Phoenix
release out the door, even if incmplete so we can start
porting to that (and Python 3).

One of the big things would be to convince Debian to include Phoenix ...

Once we got that I'd be one reporting bugs like crazy (for
which some money could be spent, eventually).

Karsten

···

On Mon, Apr 25, 2016 at 08:51:29AM -0700, John Fabiani wrote:

It's been almost 16 months since we had an update. I don't see Robin
answering question any longer - at least none I noticed. The Phoenix
project just has nightly builds but I don't see new code (not exactly true
there is something supporting python 3.5). The messages and questions have
slowed to a crawl.

So I think it is fair to say "we are very close to one more dead open
source project".

I do not want to see this die. I'm willing to help with some money (I
don't have the skills to write C or C ++ code). I have limited resources
but I can help with some cash.

But first we need to find someone willing to take over with the skills
(even if we have the money). I know many believe the desktop program is
dead. But it's not - with the web being hacked daily and mobile apps not
suited to data entry - the desktop is NOT dead. GitHub just came out with
Electron - because they believe the desktop is not dead.

We could just move to other tech and say bye to wxPython but what about the
investment we all made. If wxPython would just moved to Python 3 we all
could get many more years just using wxPython.

Please guys don't let this die. Let's do something - organize a way to
fund a programmer and move forward. Let's not lose our investment in
wxPython.

--
GPG key ID E4071346 @ eu.pool.sks-keyservers.net
E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346

3

I share the same concerns here. I’m sad to see great developers such as Robin and Andrea Gavana disappear from the scene. I don’t know how I can help but I see an opportunity here for skilled developers to pick up the wxPython code and perhaps create a commercial version of it if licensing permits. I don’t mind that. Perhaps similar to the QT suite.

···

On Monday, April 25, 2016 at 8:51:32 AM UTC-7, johnf wrote:

Hi Everyone,

It’s been almost 16 months since we had an update. I don’t see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don’t see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.

So I think it is fair to say “we are very close to one more dead open source project”.

I do not want to see this die. I’m willing to help with some money (I don’t have the skills to write C or C ++ code). I have limited resources but I can help with some cash.

But first we need to find someone willing to take over with the skills (even if we have the money). I know many believe the desktop program is dead. But it’s not - with the web being hacked daily and mobile apps not suited to data entry - the desktop is NOT dead. GitHub just came out with Electron - because they believe the desktop is not dead.

We could just move to other tech and say bye to wxPython but what about the investment we all made. If wxPython would just moved to Python 3 we all could get many more years just using wxPython.

Please guys don’t let this die. Let’s do something - organize a way to fund a programmer and move forward. Let’s not lose our investment in wxPython.

Johnf

4

Robin does not have too much time for wxPython, so a proper release is probably some time away.
Nonetheless, Phoenix is quite usable. I have been using it with Python 3.4 and 3.5
for several months now and I did not run into problems.
(I'm using my own build, as the nightly builds do not yet have the Metafile classes.)
Just give Phoenix a try. Keep your code 2/3 neutral in the beginning until you're confident that Phoenix is working for you.

Regards,

Dietmar

···

On 25.04.2016 17:51, John Fabiani wrote:

It's been almost 16 months since we had an update. I don't see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don't see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.

5

I think “giving it a try” is not a sustainable solution for this situation. The wxPython community has to do something now or, unfortunately, many would start migrating to other frameworks. I want to stick with wxPython but the situation isn’t looking good unless we act. I’m happy to contribute to pay a developer maintain the code.

···

On Mon, Apr 25, 2016 at 11:14 AM, Dietmar Schwertberger maillist@schwertberger.de wrote:

It’s been almost 16 months since we had an update. I don’t see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don’t see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.
On 25.04.2016 17:51, John Fabiani wrote:

Robin does not have too much time for wxPython, so a proper release is probably some time away.

Nonetheless, Phoenix is quite usable. I have been using it with Python 3.4 and 3.5

for several months now and I did not run into problems.

(I’m using my own build, as the nightly builds do not yet have the Metafile classes.)

Just give Phoenix a try. Keep your code 2/3 neutral in the beginning until you’re confident that Phoenix is working for you.

Regards,

Dietmar

–

You received this message because you are subscribed to a topic in the Google Groups “wxPython-users” group.

To unsubscribe from this topic, visit https://groups.google.com/d/topic/wxpython-users/RO2yGI9-fNE/unsubscribe.

To unsubscribe from this group and all its topics, send an email to wxpython-users+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

6

Dear wxPythoneers,

···

On 25 April 2016 at 18:57, Emad Dlala wrote:

I share the same concerns here. I’m sad to see great developers such as Robin and Andrea Gavana disappear from the scene. I don’t know how I can help but I see an opportunity here for skilled developers to pick up the wxPython code and perhaps create a commercial version of it if licensing permits. I don’t mind that. Perhaps similar to the QT suite.

On Monday, April 25, 2016 at 8:51:32 AM UTC-7, johnf wrote:

Hi Everyone,

It’s been almost 16 months since we had an update. I don’t see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don’t see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.

So I think it is fair to say “we are very close to one more dead open source project”.

I do not want to see this die. I’m willing to help with some money (I don’t have the skills to write C or C ++ code). I have limited resources but I can help with some cash.

But first we need to find someone willing to take over with the skills (even if we have the money). I know many believe the desktop program is dead. But it’s not - with the web being hacked daily and mobile apps not suited to data entry - the desktop is NOT dead. GitHub just came out with Electron - because they believe the desktop is not dead.

We could just move to other tech and say bye to wxPython but what about the investment we all made. If wxPython would just moved to Python 3 we all could get many more years just using wxPython.

Please guys don’t let this die. Let’s do something - organize a way to fund a programmer and move forward. Let’s not lose our investment in wxPython.

Johnf

As you are surely aware, I have been using wxPython for very many years - more than 10 years actually. And beside the AGW library, I have contributed to the Phoenix build process, documentation and porting to Python 3.

I am saddened by the long time that has passed since Robin pitched in, but then again I haven’t been any more active myself. Work requirements have escalated quite a bit for me, and my family has greatly expanded in the last two years, so my time has been severely curtailed.

That being said, I still use wxPython (Classic) all the time, for all GUI-based applications we develop. I have hundreds of thousands LOC of wxPython/Python tools. One of Maersk Oil flagship applications would not have been possible without me being able to use wxPython as a GUI framework.

I’d be willing to give a shot at trying and push Phoenix forward - at least until Robin (who is always the lead developer and BDFL of wxPython) has the time to do some more work on wxPython. Depending on the requirements, I may be able to commit around an hour per day to this task. I am unclear how any financial investments might be used in this regard, and in any case I would be reluctant to accept anything until I can prove that I can devote the time I promised to phoenix development and that I am actually able to do the technical work. I kind of despise Python 3 for all the nuances it introduced for such little gains, but I recognize that there are now quite a lot of developers using it as their main programming language version.

Of course, all I said is just nonsense if the community has better ideas on how to push the wxPython development forward. And I may also need some help in scheduling/prioritizing the bulk of the work so that our time is not wasted in tiny details (at least at the beginning).

Suggestions/comments are more than appreciated :slight_smile: .

Andrea.

7

Hellooooooo Andrea,

I feel so much better now just to know that you’re still a wxPyhton user. This gives me a lot more hope :slight_smile:

Guys, we should really, really figure out how we can arrange for Andrea to push Phoenix forward.

I’m also a fan of Python 2.7 (not yet into Python 3) and I see Python 2.7 alive and active for many years to come so I’d recommend to have a Phoenix version that compatible with Python 2.7. Again it’s just a suggestion and I leave the experts talk and decide.

Welcome again Andrea! Hope Robin will follow :slight_smile:

Thanks,

Emad

···

On Mon, Apr 25, 2016 at 11:44 AM, Andrea Gavana andrea.gavana@gmail.com wrote:

Dear wxPythoneers,

–

You received this message because you are subscribed to a topic in the Google Groups “wxPython-users” group.

To unsubscribe from this topic, visit https://groups.google.com/d/topic/wxpython-users/RO2yGI9-fNE/unsubscribe.

To unsubscribe from this group and all its topics, send an email to wxpython-users+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

On 25 April 2016 at 18:57, Emad Dlala wrote:

I share the same concerns here. I’m sad to see great developers such as Robin and Andrea Gavana disappear from the scene. I don’t know how I can help but I see an opportunity here for skilled developers to pick up the wxPython code and perhaps create a commercial version of it if licensing permits. I don’t mind that. Perhaps similar to the QT suite.

On Monday, April 25, 2016 at 8:51:32 AM UTC-7, johnf wrote:

Hi Everyone,

It’s been almost 16 months since we had an update. I don’t see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don’t see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.

So I think it is fair to say “we are very close to one more dead open source project”.

I do not want to see this die. I’m willing to help with some money (I don’t have the skills to write C or C ++ code). I have limited resources but I can help with some cash.

But first we need to find someone willing to take over with the skills (even if we have the money). I know many believe the desktop program is dead. But it’s not - with the web being hacked daily and mobile apps not suited to data entry - the desktop is NOT dead. GitHub just came out with Electron - because they believe the desktop is not dead.

We could just move to other tech and say bye to wxPython but what about the investment we all made. If wxPython would just moved to Python 3 we all could get many more years just using wxPython.

Please guys don’t let this die. Let’s do something - organize a way to fund a programmer and move forward. Let’s not lose our investment in wxPython.

Johnf

As you are surely aware, I have been using wxPython for very many years - more than 10 years actually. And beside the AGW library, I have contributed to the Phoenix build process, documentation and porting to Python 3.

I am saddened by the long time that has passed since Robin pitched in, but then again I haven’t been any more active myself. Work requirements have escalated quite a bit for me, and my family has greatly expanded in the last two years, so my time has been severely curtailed.

That being said, I still use wxPython (Classic) all the time, for all GUI-based applications we develop. I have hundreds of thousands LOC of wxPython/Python tools. One of Maersk Oil flagship applications would not have been possible without me being able to use wxPython as a GUI framework.

I’d be willing to give a shot at trying and push Phoenix forward - at least until Robin (who is always the lead developer and BDFL of wxPython) has the time to do some more work on wxPython. Depending on the requirements, I may be able to commit around an hour per day to this task. I am unclear how any financial investments might be used in this regard, and in any case I would be reluctant to accept anything until I can prove that I can devote the time I promised to phoenix development and that I am actually able to do the technical work. I kind of despise Python 3 for all the nuances it introduced for such little gains, but I recognize that there are now quite a lot of developers using it as their main programming language version.

Of course, all I said is just nonsense if the community has better ideas on how to push the wxPython development forward. And I may also need some help in scheduling/prioritizing the bulk of the work so that our time is not wasted in tiny details (at least at the beginning).

Suggestions/comments are more than appreciated :slight_smile: .

Andrea.

8

Wow your efforts would be awesome! When you get to a point where money can be accepted just shout out. I’ll bet there is enough to cover some of the pain that would come from taking time away from other duties. Also I can test and help with many other items. Just let me know!

I use Dabo which is built on wxPython and I can promise that shortly after Phoenix goes stable I’ll get Dabo working.

I believe that the current 3.x is fine for 2.7 and there is no need to attempt to have Phoenix support python 2.7. Once everything is stable and working for python 3.x then we can talk about adding features etc. Including my request of supporting some sort of CSS.

Just shout out! I’ll help!

Johnf

···

On Monday, April 25, 2016 at 11:44:43 AM UTC-7, Infinity77 wrote:

Dear wxPythoneers,

On 25 April 2016 at 18:57, Emad Dlala wrote:

I share the same concerns here. I’m sad to see great developers such as Robin and Andrea Gavana disappear from the scene. I don’t know how I can help but I see an opportunity here for skilled developers to pick up the wxPython code and perhaps create a commercial version of it if licensing permits. I don’t mind that. Perhaps similar to the QT suite.

On Monday, April 25, 2016 at 8:51:32 AM UTC-7, johnf wrote:

Hi Everyone,

It’s been almost 16 months since we had an update. I don’t see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don’t see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.

So I think it is fair to say “we are very close to one more dead open source project”.

I do not want to see this die. I’m willing to help with some money (I don’t have the skills to write C or C ++ code). I have limited resources but I can help with some cash.

But first we need to find someone willing to take over with the skills (even if we have the money). I know many believe the desktop program is dead. But it’s not - with the web being hacked daily and mobile apps not suited to data entry - the desktop is NOT dead. GitHub just came out with Electron - because they believe the desktop is not dead.

We could just move to other tech and say bye to wxPython but what about the investment we all made. If wxPython would just moved to Python 3 we all could get many more years just using wxPython.

Please guys don’t let this die. Let’s do something - organize a way to fund a programmer and move forward. Let’s not lose our investment in wxPython.

Johnf

As you are surely aware, I have been using wxPython for very many years - more than 10 years actually. And beside the AGW library, I have contributed to the Phoenix build process, documentation and porting to Python 3.

I am saddened by the long time that has passed since Robin pitched in, but then again I haven’t been any more active myself. Work requirements have escalated quite a bit for me, and my family has greatly expanded in the last two years, so my time has been severely curtailed.

That being said, I still use wxPython (Classic) all the time, for all GUI-based applications we develop. I have hundreds of thousands LOC of wxPython/Python tools. One of Maersk Oil flagship applications would not have been possible without me being able to use wxPython as a GUI framework.

I’d be willing to give a shot at trying and push Phoenix forward - at least until Robin (who is always the lead developer and BDFL of wxPython) has the time to do some more work on wxPython. Depending on the requirements, I may be able to commit around an hour per day to this task. I am unclear how any financial investments might be used in this regard, and in any case I would be reluctant to accept anything until I can prove that I can devote the time I promised to phoenix development and that I am actually able to do the technical work. I kind of despise Python 3 for all the nuances it introduced for such little gains, but I recognize that there are now quite a lot of developers using it as their main programming language version.

Of course, all I said is just nonsense if the community has better ideas on how to push the wxPython development forward. And I may also need some help in scheduling/prioritizing the bulk of the work so that our time is not wasted in tiny details (at least at the beginning).

Suggestions/comments are more than appreciated :slight_smile: .

Andrea.

9

I also would be happy to help financially support the continuation of wxPython. I have countless hours invested and would like to continue with it. Even an update of Classic to 3.1.0 would help. I could use the 4k display and MediaCtrl updates in wxWidgets 3.1.0.

Phoenix is out of the question for me as is. It does not support required controls, MediaCtrl being a big one.

The Desktop is not dead. It has slowed some but is by no means dead. I have over a 1000 downloads every week of my apps on just Windows. Not counting Mac and Linux.

10

Hi,

nice thread! some months ago I asked the same… actually I have many systems in Python 2.7+wxpython and I need to migrate into Python 3, for example latest release of Ubuntu 16.04 use Python 3.5 like main version and Python 2.7 for compatibility, I think the others distros will do the same in the next months.

So due to the lack of answers here about next release I start searching, Kivy is good for mobiles apps but still a littleboy for desktop software… now I just can read about Pyside will be developed by the same company of QT and with the same license (not dual like pyqt) so now this lib could be a good option if I need to port all my systems to Python 3.x.

Meanwhile I can contribute with testing …for example I reported this bug in Ubuntu:

https://bugs.launchpad.net/ubuntu/+source/wxwidgets3.0/+bug/1388847

more than 1 year ago and still is not fixed :frowning: looks like distros dont want to help with wxpython …

So I’ll be reading news here, hope wxpython can take again his place in the Opensource software, or I’ll need to wait for Pyside+qt5

···

2016-04-26 14:36 GMT-05:00 tsmorton tim@preservedwords.com:

I also would be happy to help financially support the continuation of wxPython. I have countless hours invested and would like to continue with it. Even an update of Classic to 3.1.0 would help. I could use the 4k display and MediaCtrl updates in wxWidgets 3.1.0.

Phoenix is out of the question for me as is. It does not support required controls, MediaCtrl being a big one.

The Desktop is not dead. It has slowed some but is by no means dead. I have over a 1000 downloads every week of my apps on just Windows. Not counting Mac and Linux.

–

You received this message because you are subscribed to the Google Groups “wxPython-users” group.

To unsubscribe from this group and stop receiving emails from it, send an email to wxpython-users+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Saludos / Best regards

Mario Lacunza
Email:: mlacunza@gmail.com
Personal Website:: http://www.lacunza.biz/
Hosting:: http://mlv-host.com/
Mascotas Perdidas:: http://mascotas-perdidas.com/
Skype: mlacunzav

Lima - Peru

11

This is a pet peeve of mine. I know my opinion is not universally
shared, but here it is anyway.
When one starts an open source project, one envisions that it will
solve some problem. Once that problem has been solved, the project
is finished. The open source world seems to have a SERIOUS problem
with any project being declared “finished”. Why do we feel the need
to churn features endlessly? If wxPython provides a Python
interface to wxWidgets, and that works, then why should we expect
there to be a continuous stream of pointless features?
Now, it’s true that there are known, unsolved bugs in wxPython, so
in that sense there is still work to be done, but I’m not sure
there’s anything wrong with a “dead open source project”. After
all, the entire user interface universe upon which wxWidgets is
based is itself dead. Microsoft isn’t doing anything new with GDI,
and hasn’t for years. We’ve already missed two revolutions (Windows
Forms and WPF), and the world is moving into a third (descriptive
HTML UIs with Javascript). Should wxPython really need to be
changing?

···

John Fabiani wrote:

      It's been almost 16 months since we had an update.  ...

The Phoenix project just has nightly builds but I don’t see
new code (not exactly true there is something supporting
python 3.5). The messages and questions have slowed to a
crawl.

      So I think it is fair to say "we are very close to one more

dead open source project".

-- Tim Roberts, Providenza & Boekelheide, Inc.

timr@probo.com

12

As you are surely aware, I have been using wxPython for very many years - more than 10 years actually. And beside the AGW library, I have contributed to the Phoenix build process, documentation and porting to Python 3.

I am saddened by the long time that has passed since Robin pitched in, but then again I haven’t been any more active myself. Work requirements have escalated quite a bit for me, and my family has greatly expanded in the last two years, so my time has been severely curtailed.

That being said, I still use wxPython (Classic) all the time, for all GUI-based applications we develop. I have hundreds of thousands LOC of wxPython/Python tools. One of Maersk Oil flagship applications would not have been possible without me being able to use wxPython as a GUI framework.

I’d be willing to give a shot at trying and push Phoenix forward - at least until Robin (who is always the lead developer and BDFL of wxPython) has the time to do some more work on wxPython. Depending on the requirements, I may be able to commit around an hour per day to this task. I am unclear how any financial investments might be used in this regard, and in any case I would be reluctant to accept anything until I can prove that I can devote the time I promised to phoenix development and that I am actually able to do the technical work. I kind of despise Python 3 for all the nuances it introduced for such little gains, but I recognize that there are now quite a lot of developers using it as their main programming language version.

Of course, all I said is just nonsense if the community has better ideas on how to push the wxPython development forward. And I may also need some help in scheduling/prioritizing the bulk of the work so that our time is not wasted in tiny details (at least at the beginning).

Suggestions/comments are more than appreciated :slight_smile: .

Andrea.

While I have resisted moving to Python 3 as well, it seems that there is an official / unofficial EOL for Python 2 in 2020 - PEP 373 -- Python 2.7 Release Schedule

So if we want to use wxPython in future versions of Python, then there is definitely a reason to port to Python 3 for that reason alone. I personally don’t care if wxPython continually gets new features, but I do think it’s one of the easiest GUI toolkits for Pytohn around and I will use it for as long as I can.

Mike

13

I agree with Mike. We need to support the effort of upgrading wxPython.

···

On Tue, Apr 26, 2016 at 2:50 PM, Mike Driscoll kyosohma@gmail.com wrote:

As you are surely aware, I have been using wxPython for very many years - more than 10 years actually. And beside the AGW library, I have contributed to the Phoenix build process, documentation and porting to Python 3.

I am saddened by the long time that has passed since Robin pitched in, but then again I haven’t been any more active myself. Work requirements have escalated quite a bit for me, and my family has greatly expanded in the last two years, so my time has been severely curtailed.

That being said, I still use wxPython (Classic) all the time, for all GUI-based applications we develop. I have hundreds of thousands LOC of wxPython/Python tools. One of Maersk Oil flagship applications would not have been possible without me being able to use wxPython as a GUI framework.

I’d be willing to give a shot at trying and push Phoenix forward - at least until Robin (who is always the lead developer and BDFL of wxPython) has the time to do some more work on wxPython. Depending on the requirements, I may be able to commit around an hour per day to this task. I am unclear how any financial investments might be used in this regard, and in any case I would be reluctant to accept anything until I can prove that I can devote the time I promised to phoenix development and that I am actually able to do the technical work. I kind of despise Python 3 for all the nuances it introduced for such little gains, but I recognize that there are now quite a lot of developers using it as their main programming language version.

Of course, all I said is just nonsense if the community has better ideas on how to push the wxPython development forward. And I may also need some help in scheduling/prioritizing the bulk of the work so that our time is not wasted in tiny details (at least at the beginning).

Suggestions/comments are more than appreciated :slight_smile: .

Andrea.

While I have resisted moving to Python 3 as well, it seems that there is an official / unofficial EOL for Python 2 in 2020 - http://legacy.python.org/dev/peps/pep-0373/

So if we want to use wxPython in future versions of Python, then there is definitely a reason to port to Python 3 for that reason alone. I personally don’t care if wxPython continually gets new features, but I do think it’s one of the easiest GUI toolkits for Pytohn around and I will use it for as long as I can.

Mike

–

You received this message because you are subscribed to a topic in the Google Groups “wxPython-users” group.

To unsubscribe from this topic, visit https://groups.google.com/d/topic/wxpython-users/RO2yGI9-fNE/unsubscribe.

To unsubscribe from this group and all its topics, send an email to wxpython-users+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

14

I completely agree with you. I do not want to see wxPython or Phoenix development die. Like many others, I have been using wxPython for a long time, and have many wxPython programs that I maintain and use daily, and continue to develop. I’ve had good experiences with Phoenix, but have done only light testing with Phoenix and Python3. I am not eager to switch to another GUI toolkit.

Like others have expressed, I doubt I could actually do the C++ work myself, but I’d be willing to try to pitch in resources. I should admit that I’m somewhat disappointed that neither Enthought nor ContinuumIO have made this happen. If it is only a matter of paying for Robin’s time (or Enthought or ContinuumIO) o do the work, would it be too impolite to find out what is needed?

Anyway, as you say it have been a very long time. I think we have to start considering the possibility that relying on Robin to do this development may not be working out very well for wxPython users. Assuming for the moment that Robin is not going to make a Phoenix release, does anyone have a sense for how much work that would be?

Cheers,

–Matt Newville

···

On Monday, April 25, 2016 at 10:51:32 AM UTC-5, johnf wrote:

Hi Everyone,

It’s been almost 16 months since we had an update. I don’t see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don’t see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.

So I think it is fair to say “we are very close to one more dead open source project”.

I do not want to see this die. I’m willing to help with some money (I don’t have the skills to write C or C ++ code). I have limited resources but I can help with some cash.

But first we need to find someone willing to take over with the skills (even if we have the money). I know many believe the desktop program is dead. But it’s not - with the web being hacked daily and mobile apps not suited to data entry - the desktop is NOT dead. GitHub just came out with Electron - because they believe the desktop is not dead.

We could just move to other tech and say bye to wxPython but what about the investment we all made. If wxPython would just moved to Python 3 we all could get many more years just using wxPython.

Please guys don’t let this die. Let’s do something - organize a way to fund a programmer and move forward. Let’s not lose our investment in wxPython.

Johnf

15

Hi,

···

On 27 April 2016 at 05:28, Matthew Newville matt.newville@gmail.com wrote:

On Monday, April 25, 2016 at 10:51:32 AM UTC-5, johnf wrote:

Hi Everyone,

It’s been almost 16 months since we had an update. I don’t see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don’t see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.

So I think it is fair to say “we are very close to one more dead open source project”.

I do not want to see this die. I’m willing to help with some money (I don’t have the skills to write C or C ++ code). I have limited resources but I can help with some cash.

But first we need to find someone willing to take over with the skills (even if we have the money). I know many believe the desktop program is dead. But it’s not - with the web being hacked daily and mobile apps not suited to data entry - the desktop is NOT dead. GitHub just came out with Electron - because they believe the desktop is not dead.

We could just move to other tech and say bye to wxPython but what about the investment we all made. If wxPython would just moved to Python 3 we all could get many more years just using wxPython.

Please guys don’t let this die. Let’s do something - organize a way to fund a programmer and move forward. Let’s not lose our investment in wxPython.

Johnf

I completely agree with you. I do not want to see wxPython or Phoenix development die. Like many others, I have been using wxPython for a long time, and have many wxPython programs that I maintain and use daily, and continue to develop. I’ve had good experiences with Phoenix, but have done only light testing with Phoenix and Python3. I am not eager to switch to another GUI toolkit.

Like others have expressed, I doubt I could actually do the C++ work myself, but I’d be willing to try to pitch in resources. I should admit that I’m somewhat disappointed that neither Enthought nor ContinuumIO have made this happen. If it is only a matter of paying for Robin’s time (or Enthought or ContinuumIO) o do the work, would it be too impolite to find out what is needed?

Anyway, as you say it have been a very long time. I think we have to start considering the possibility that relying on Robin to do this development may not be working out very well for wxPython users. Assuming for the moment that Robin is not going to make a Phoenix release, does anyone have a sense for how much work that would be?

It can be quite some work. To be honest, I think there is only one person who can move swiftly and nimbly through the entire wxPython codebase - after all,Robin created wxPython. For anybody else, it’s going to be a slower, more error-prone approach.

Although we have introduced a lot of automation in the Phoenix building/compilation/testing/documentation generation/continuous integration process, there are still quite a few places where the programmer intervention in the only way: most notably patches, Python docstrings, testing-testing-testing, clear and reproducible bug reports. Not to mention the support of various versions/architectures on Windows, 1,489 different Linux distributions (each of them with their funny assumptions about HIG for user interfaces) and Mac. I would tend to say that it’s quite a task for a single person.

I have started playing with the latest Phoenix - forked, compiled, built the docs, playing with the demo, looked at the bugs and PRs on GitHub. I believe that fixing the bugs in the Python code or adapting it to support Python 3 should not be a daunting task, but I am less sure when it comes to the C++/Python integration part. More so, if a bug is found in the mother library (wxWidgets), then it needs to be demonstrated with a C++ sample, submitted as a bug report, a patch needs to be created, and maybe it will get accepted. We have no control over that.

While I can do quite some work for Windows and Linux, I have zero access to and zero experience with Mac. So any bug that will need testing/debugging on that platform will leave me powerless. I will do some more work in the next coming days on my fork to see how far support for Python 3 has gotten, and potentially iron out some of the bugs on Python 2. In the meanwhile, I hope everybody knows that there are Phoenix snapshots built continuously here:

http://wxpython.org/Phoenix/snapshot-builds/

For all platforms/architectures I can think of. I think we should all try them out and see if there are strange things going on.

Andrea.

16

IMO the above is not meant to say wxPython is dead because
there are no new features being added.

I would rather think it alludes to the fact that it doesn't
seem to be kept relevant on current platforms as timely as
one would wish:

- Python 3
- "official" releases of Phoenix
- packages for, say, Debian or Ubuntu

and thusly risks to fall by the wayside DESPITE NOT being
dead as far as usefulness is concerned.

It is important to have available official releases and
packages for *end user* suitability.

Note that I am not whining about the above situation, merely
pointing out my understanding of where we stand with regard
to wxPython. Unfortunately, I lack the skills and time to
offer substantial help.

Karsten

···

On Tue, Apr 26, 2016 at 02:33:01PM -0700, Tim Roberts wrote:

John Fabiani wrote:

So I think it is fair to say "we are very close to one more dead open
source project".

This is a pet peeve of mine. I know my opinion is not universally
shared, but here it is anyway.

When one starts an open source project, one envisions that it will solve
some problem. Once that problem has been solved, the project is
finished. The open source world seems to have a SERIOUS problem with
any project being declared "finished". Why do we feel the need to churn
features endlessly? If wxPython provides a Python interface to
wxWidgets, and that works, then why should we expect there to be a
continuous stream of pointless features?

--
GPG key ID E4071346 @ eu.pool.sks-keyservers.net
E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346

17

Fabulous, Andrea!

I can help testing Phoenix in Windows 7 & 10 and report bugs. Also I can even help more with any patches for wxPython 3 that’s compatible with Python 2.7.

···

On Tue, Apr 26, 2016 at 10:47 PM, Andrea Gavana andrea.gavana@gmail.com wrote:

Hi,

–

You received this message because you are subscribed to the Google Groups “wxPython-users” group.

To unsubscribe from this group and stop receiving emails from it, send an email to wxpython-users+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

On 27 April 2016 at 05:28, Matthew Newville matt.newville@gmail.com wrote:

Hi Everyone,

It’s been almost 16 months since we had an update. I don’t see Robin answering question any longer - at least none I noticed. The Phoenix project just has nightly builds but I don’t see new code (not exactly true there is something supporting python 3.5). The messages and questions have slowed to a crawl.

So I think it is fair to say “we are very close to one more dead open source project”.

I do not want to see this die. I’m willing to help with some money (I don’t have the skills to write C or C ++ code). I have limited resources but I can help with some cash.

But first we need to find someone willing to take over with the skills (even if we have the money). I know many believe the desktop program is dead. But it’s not - with the web being hacked daily and mobile apps not suited to data entry - the desktop is NOT dead. GitHub just came out with Electron - because they believe the desktop is not dead.

We could just move to other tech and say bye to wxPython but what about the investment we all made. If wxPython would just moved to Python 3 we all could get many more years just using wxPython.

Please guys don’t let this die. Let’s do something - organize a way to fund a programmer and move forward. Let’s not lose our investment in wxPython.

Johnf

On Monday, April 25, 2016 at 10:51:32 AM UTC-5, johnf wrote:

I completely agree with you. I do not want to see wxPython or Phoenix development die. Like many others, I have been using wxPython for a long time, and have many wxPython programs that I maintain and use daily, and continue to develop. I’ve had good experiences with Phoenix, but have done only light testing with Phoenix and Python3. I am not eager to switch to another GUI toolkit.

Like others have expressed, I doubt I could actually do the C++ work myself, but I’d be willing to try to pitch in resources. I should admit that I’m somewhat disappointed that neither Enthought nor ContinuumIO have made this happen. If it is only a matter of paying for Robin’s time (or Enthought or ContinuumIO) o do the work, would it be too impolite to find out what is needed?

Anyway, as you say it have been a very long time. I think we have to start considering the possibility that relying on Robin to do this development may not be working out very well for wxPython users. Assuming for the moment that Robin is not going to make a Phoenix release, does anyone have a sense for how much work that would be?

It can be quite some work. To be honest, I think there is only one person who can move swiftly and nimbly through the entire wxPython codebase - after all,Robin created wxPython. For anybody else, it’s going to be a slower, more error-prone approach.

Although we have introduced a lot of automation in the Phoenix building/compilation/testing/documentation generation/continuous integration process, there are still quite a few places where the programmer intervention in the only way: most notably patches, Python docstrings, testing-testing-testing, clear and reproducible bug reports. Not to mention the support of various versions/architectures on Windows, 1,489 different Linux distributions (each of them with their funny assumptions about HIG for user interfaces) and Mac. I would tend to say that it’s quite a task for a single person.

I have started playing with the latest Phoenix - forked, compiled, built the docs, playing with the demo, looked at the bugs and PRs on GitHub. I believe that fixing the bugs in the Python code or adapting it to support Python 3 should not be a daunting task, but I am less sure when it comes to the C++/Python integration part. More so, if a bug is found in the mother library (wxWidgets), then it needs to be demonstrated with a C++ sample, submitted as a bug report, a patch needs to be created, and maybe it will get accepted. We have no control over that.

While I can do quite some work for Windows and Linux, I have zero access to and zero experience with Mac. So any bug that will need testing/debugging on that platform will leave me powerless. I will do some more work in the next coming days on my fork to see how far support for Python 3 has gotten, and potentially iron out some of the bugs on Python 2. In the meanwhile, I hope everybody knows that there are Phoenix snapshots built continuously here:

http://wxpython.org/Phoenix/snapshot-builds/

For all platforms/architectures I can think of. I think we should all try them out and see if there are strange things going on.

Andrea.

18

Where money can be spent to get Robin-like help:

   - Computer with Windows 7, 8, 8.1 and 10 maybe Vista? (I'm on Windows XP
   with python 3.5)
   - Appropriate Visual Studio Pro versions for said computers
   - A Mac computer with appropriate OS and software tools.
   - Computer with Linux OS(es)
   - Profession Git(hub) account/SourceForge accounts.

I am not sure if Robin said not too long ago (perhaps not in these forums)
that he finally got the a new-dash-updated-licensed-version of Visual
Studio Pro finally.

Who ever helps needs to know how Robin codes for distributing Python
language versions within wxPython and Phoenix version-ed code. That would
help a a lot.

19

I don’t understand your point, but I think Robin need to said something at this point.

···

Where money can be spent to get Robin-like help:

  • Computer with Windows 7, 8, 8.1 and 10 maybe Vista? (I’m on Windows XP with python 3.5)
  • Appropriate Visual Studio Pro versions for said computers
  • A Mac computer with appropriate OS and software tools.
  • Computer with Linux OS(es)
  • Profession Git(hub) account/SourceForge accounts.

I am not sure if Robin said not too long ago (perhaps not in these forums) that he finally got the a new-dash-updated-licensed-version of Visual Studio Pro finally.

Who ever helps needs to know how Robin codes for distributing Python language versions within wxPython and Phoenix version-ed code. That would help a a lot.

20

Hi Andrea,
it’s great news that you’re stepping forward! While I think you’re surely up to the task, my humble advice would be to set things in such a way to make it easier for other people to join in. We need some sort of roadmap, a list of priorities, and a call for volunteers.

For what it’s worth, my opinion would be:

  • to completely forget about wxPython classic and more in general Python 2.7 support, and just focus on Phoenix / Python 3: specifically, make clear that we won’t work on supporting wx3.1 on wxPython classic;
  • to aim to a first minimal working beta of Phoenix, dropping all the unfinished/problematic parts for now. I know it’s a shame to leave out the media ctrl and many other things: but I think that if we can come out with a first clean Phoenix version that “just works”, then the project will regain momentum and the missing widgets could be re-added later;
  • by “just working” I mean 1) targeting Python 3.5 and ready for Python 3.6; 2) easily pip-installing and working in a consistent way on all the supported platforms; 3) well documented (eg., a clear migration guide from wxPython classic);
  • then, we should focus on provinding support: if we are able to close tickets, release bugfixes etc., the project will live again;
  • then, and only them, we should start working on the missing parts.

Having said that, it’s still not clear to me how to give shape to this rebooting effort. Do you have commit privileges on the official Phoenix repo? If not, are you thinking to officially fork it, maintainging eg. a separate bug tracker, separate website, etc.?

Also, it would be nice to hear from Werner about this - AFAIK he has commit privileges on the Phoenix repo, and he is (was?) the unofficial “second in command” until last year…

Finally, of course, any word from Robin would be more than welcome at this point.

just my 2c

riccardo

···

Il giorno mercoledì 27 aprile 2016 07:47:12 UTC+2, Infinity77 ha scritto:

It can be quite some work. To be honest, I think there is only one person who can move swiftly and nimbly through the entire wxPython codebase - after all,Robin created wxPython. For anybody else, it’s going to be a slower, more error-prone approach.