A little bit else I would like to discuss

[azrael]

Sometimes I really get confused when looking out for a modul for some
kind of need. Sometimes I get frightened when I get the resaults. 8
wraper for this, 7 wrapers for that, 10 modules for anything. Between
them are maybe some kind of small differences, but to work with any of
the modules, I have to spend 10 hours of reading the help, If there is
any at all.

I think that there should be a list on python.org of supported or
sugested modules for some need. For example Database access. Or GUI
Building. It is a complete pain in the ass. Let's face the true, TK is
out of date. There should be another one used and appended to the
standard Python Library. One that plays well with other platforms. And
if it does't, let's make it play better.

Why will Microsoft's products kick the ass of open source. Because
anyone does what he wants. Let's say There are 5 GUI libraries
competing against each other. Think about it what could these 5 teams
acomplish if they would work together. Or maybe a framework for RAD
GUI devbelopment. after 10 years of python, there is still not one
application for GUI Building that can beat Visual Studio.

Anyone I talk to says: "Oh my god, Python and GUI"

There are a dozen of modules that should be, if you ask me, be
implemented into Python. Like NumPy, SciPY, PIL. A lot of people I
talk to, are using them. But everyone has to download them manually
and hope that if someone is using their code, has also installed the
needed modules.



My solution would be:
Let's make an announcement on Python.org. something like this.

We need Ideas for creating the a standard library in Python for Image
processing. Tell us your problem. What do you think which feature
should be implemented. Which transformation, which algorithm.
Make a vote. This should be that not.

But how to start something like that. Anyone that sees a problem that
should be solved, becomes the project leader. Find the best Ideas and
make a good library.


It's is very good that people have a choice. But why not make some
standards. I know that there is already a standard python library, But
why not extending it. classify the standard library into subcategories
like Networking, DataBase, Computation, ......



One other thing. I am not looking for war on groups. I just would like
to discuss some things that drive me crazy while using Python.

This group has a lot of members. not to mention every Python Forum.
Why not using this number of people and accomplish something great. If
anyone of us would write 10 line of good code, it would result a very
great and powerfull environment.

 

[Philip Semanchuk]

Why will Microsoft's products kick the ass of open source. Because
anyone does what he wants. Let's say There are 5 GUI libraries
competing against each other. Think about it what could these 5 teams
acomplish if they would work together. Or maybe a framework for RAD
GUI devbelopment. after 10 years of python, there is still not one
application for GUI Building that can beat Visual Studio.

Let me know when Visual Studio tries to address building GUIs for
Windows, OS X, Gnome, and KDE. Until then, you're comparing apples to
oranges, or chalk and cheese if you're from that part of the world.

 

[Tim Golden]

Let me know when Visual Studio tries to address building GUIs for
Windows, OS X, Gnome, and KDE. Until then, you're comparing apples to
oranges, or chalk and cheese if you're from that part of the world.

While your point is obviously good, it only applies if you
even care about building for OS X, Gnome etc. If all you
want is build on Windows (or Gnome or KDE or whatever) then
the cross-platformness of any proposed toolkit is worthless
and may even be a drawback if it reduces its functionality
to a lowest common denominator.

I have nothing really to say about GUI-building whatsits in
general; I don't have cause to use them myself so I simply
stand back and let their proponents fight it out. I just
wanted to point out that playing the cross-platform card
does not mean the game is over.

TJG

 

[Martin]

Hi,

at first I wanted to file this under meta-discussions, but your lost
paragraph got me thinking...

Nobody is going to stop you from creating a large bundle of useful
extensions as long as you follow the licenses. In fact lots of people
may appreciate a bundle. But the Python core package will always stay
small and agile.

How does "small and agile" work with "batteries included"?

From my point of view:

agile::
Would describe faster extension of the standard lib (rrd, yaml should
IMHO already be in the standard lib). I'm pretty sure other people
want to see other modules, but that's what agile + standard lib would
mean for me. (Also I'm sometimes confused by the naming of modules but
that's a different story)

small::
just the opposite of "batteries included"

My absolute favorite would be

* python as just python (no standard lib)
* a (rather) fast moving standard lib available as an addon download
* possibly for each version

I guess that would mean *a lot* maintenance overhead so that will
probably never.

Finally: Ignore me, Python is still the language that let's me get
stuff done in the fastest and most enjoyable way. Thanks to GvR for
the initial idea and all the volounteers for making it even better


--
http://soup.alt.delete.co.at
http://www.xing.com/profile/Martin_Marcher
http://www.linkedin.com/in/martinmarcher

You are not free to read this message,
by doing so, you have violated my licence
and are required to urinate publicly. Thank you.

Please avoid sending me Word or PowerPoint attachments.
See http://www.gnu.org/philosophy/no-word-attachments.html

 

[Martin]

Oh yeah and ignore my typos also

Hi,

at first I wanted to file this under meta-discussions, but your lost
paragraph got me thinking...



How does "small and agile" work with "batteries included"?

From my point of view:

agile::
Would describe faster extension of the standard lib (rrd, yaml should
IMHO already be in the standard lib). I'm pretty sure other people
want to see other modules, but that's what agile + standard lib would
mean for me. (Also I'm sometimes confused by the naming of modules but
that's a different story)

small::
just the opposite of "batteries included"

My absolute favorite would be

* python as just python (no standard lib)
* a (rather) fast moving standard lib available as an addon download
* possibly for each version

I guess that would mean *a lot* maintenance overhead so that will
probably never.

Finally: Ignore me, Python is still the language that let's me get
stuff done in the fastest and most enjoyable way. Thanks to GvR for
the initial idea and all the volounteers for making it even better


--
http://soup.alt.delete.co.at
http://www.xing.com/profile/Martin_Marcher
http://www.linkedin.com/in/martinmarcher

You are not free to read this message,
by doing so, you have violated my licence
and are required to urinate publicly. Thank you.

Please avoid sending me Word or PowerPoint attachments.
See http://www.gnu.org/philosophy/no-word-attachments.html

--
http://soup.alt.delete.co.at
http://www.xing.com/profile/Martin_Marcher
http://www.linkedin.com/in/martinmarcher

You are not free to read this message,
by doing so, you have violated my licence
and are required to urinate publicly. Thank you.

Please avoid sending me Word or PowerPoint attachments.
See http://www.gnu.org/philosophy/no-word-attachments.html

 

[r]

Sometimes I really get confused when looking out for a modul for some
kind of need. Sometimes I get frightened when I get the resaults. 8
wraper for this, 7 wrapers for that, 10 modules for anything. Between
them are maybe some kind of small differences, but to work with any of
the modules, I have to spend 10 hours of reading the help, If there is
any at all.

I don't follow you here??

I think that there should be a list on python.org of supported or
sugested modules for some need. For example Database access. Or GUI
Building. It is a complete pain in the ass. Let's face the true, TK is
out of date. There should be another one used and appended to the
standard Python Library. One that plays well with other platforms. And
if it does't, let's make it play better.

There is a list of third party modules on the Python.org site and it
is huge! Tk is very simple i agree, but i would hate to double the
size of my python installer download just to have this or that module
built-in. Trust me you really don't want that. But i will agree a
better classification of modules would be nice

Why will Microsoft's products kick the ass of open source. Because
anyone does what he wants. Let's say There are 5 GUI libraries
competing against each other. Think about it what could these 5 teams
acomplish if they would work together. Or maybe a framework for RAD
GUI devbelopment. after 10 years of python, there is still not one
application for GUI Building that can beat Visual Studio.

Anyone I talk to says: "Oh my god, Python and GUI"

I will agree with you here. Just think what could be accomplished if
all the linux distro developers could band together to create one
golden linux distro. This level of collaboration would create a distro
so powerful the vibrations of it will cause Bill Gates house to slide
into lake Washington .

Tk could use a polishing i much agree and so could IDLE. Maybe we
should start a revolution here? i am game!

There are a dozen of modules that should be, if you ask me, be
implemented into Python. Like NumPy, SciPY, PIL. A lot of people I
talk to, are using them. But everyone has to download them manually
and hope that if someone is using their code, has also installed the
needed modules.

Again i very much disagree here. We don't want to make Python a piece
of elephant-size MS bloatware. Not everybody needs or wants all these
libraries and maintaining this extra code-base puts exrta strain on
the Python dev team. We need them to concentrate on other more
important issues.

My solution would be:
Let's make an announcement on Python.org. something like this.

We need Ideas for creating the a standard library in Python for Image
processing. Tell  us your problem. What do you think which feature
should be implemented. Which transformation, which algorithm.
Make a vote. This should be that not.

There is a library PIL but it could use some updating too.

But how to start something like that. Anyone that sees a problem that
should be solved, becomes the project leader. Find the best Ideas and
make a good library.

Motivation is hard to find, and it's more than that. People have to
make a living, but we never know until we try. I am with you! Lest
revolutionize Python!

It's is very good that people have a choice. But why not make some
standards. I know that there is already a standard python library, But
why not extending it. classify the standard library into subcategories
like Networking, DataBase, Computation, ......

Bloated_python == dead_Python

One other thing. I am not looking for war on groups. I just would like
to discuss some things that drive me crazy while using Python.

This group has a lot of members. not to mention every Python Forum.
Why not using this number of people and accomplish something great. If
anyone of us would write 10 line of good code, it would result a very
great and powerfull environment.

I agree, anybody else out there want to add to this converstation?

 

[Terry Reedy]

Sometimes I really get confused when looking out for a modul for some
kind of need. Sometimes I get frightened when I get the resaults. 8
wraper for this, 7 wrapers for that, 10 modules for anything. Between
them are maybe some kind of small differences, but to work with any of
the modules, I have to spend 10 hours of reading the help, If there is
any at all.

I think that there should be a list on python.org of supported or
sugested modules for some need. For example Database access. Or GUI
Building. It is a complete pain in the ass. Let's face the true, TK is
out of date. There should be another one used and appended to the
standard Python Library. One that plays well with other platforms. And
if it does't, let's make it play better.

You are welcome to volunteer to write entries for the Python wiki.

 

[Jason Scheirer]

Sometimes I really get confused when looking out for a modul for some
kind of need. Sometimes I get frightened when I get the resaults. 8
wraper for this, 7 wrapers for that, 10 modules for anything. Between
them are maybe some kind of small differences, but to work with any of
the modules, I have to spend 10 hours of reading the help, If there is
any at all.

I think that there should be a list on python.org of supported or
sugested modules for some need. For example Database access. Or GUI
Building. It is a complete pain in the ass. Let's face the true, TK is
out of date. There should be another one used and appended to the
standard Python Library. One that plays well with other platforms. And
if it does't, let's make it play better.

Why will Microsoft's products kick the ass of open source. Because
anyone does what he wants. Let's say There are 5 GUI libraries
competing against each other. Think about it what could these 5 teams
acomplish if they would work together. Or maybe a framework for RAD
GUI devbelopment. after 10 years of python, there is still not one
application for GUI Building that can beat Visual Studio.

Anyone I talk to says: "Oh my god, Python and GUI"

There are a dozen of modules that should be, if you ask me, be
implemented into Python. Like NumPy, SciPY, PIL. A lot of people I
talk to, are using them. But everyone has to download them manually
and hope that if someone is using their code, has also installed the
needed modules.

My solution would be:
Let's make an announcement on Python.org. something like this.

We need Ideas for creating the a standard library in Python for Image
processing. Tell  us your problem. What do you think which feature
should be implemented. Which transformation, which algorithm.
Make a vote. This should be that not.

But how to start something like that. Anyone that sees a problem that
should be solved, becomes the project leader. Find the best Ideas and
make a good library.

It's is very good that people have a choice. But why not make some
standards. I know that there is already a standard python library, But
why not extending it. classify the standard library into subcategories
like Networking, DataBase, Computation, ......

One other thing. I am not looking for war on groups. I just would like
to discuss some things that drive me crazy while using Python.

This group has a lot of members. not to mention every Python Forum.
Why not using this number of people and accomplish something great. If
anyone of us would write 10 line of good code, it would result a very
great and powerfull environment.

Now hold on here, Microsoft has a pretty schizophrenic opinion on GUIs
itself. Over the years you've had direct Win32 calls, the MFC
libraries, then WinForms and later WPF on .NET. If I were a C++/C#
developer new to Windows, which one should I go with? Or how about I
want to be cross-platform and start looking into Fox or QT or even
GTK?

The only difference between this and the number of Python GUIs is that
your version of Visual Studio sticks around for a few years. You could
standardize on a single version of the GUI you're using (say, WX) and
declare it to have a 5 year lifespan in your organization, install QT
builder and use that, or any of a large number of options. Don't
expect other people to make the decision for you, though, as an open
source environment brings about a plurality of options. Python has a
far broader scope as a tool for all kinds of tasks than a GUI-Centric
Microsoft development environment, and assuming every Python developer
wants to focus on making Python a vehicle of platform-specific GUIs
for producing shrink-wrapped software is a little off base.

Anyone I talk to says: "Oh my god, Python and GUI"

Have these people needed to spend much time getting to know any one
GUI environment? They all suck in every language at first, then the
Stockholm syndrome sets in and it's not so bad.

The problem here also is that there have been attempts to get
something going that sputtered and failed (a sort of anygui module).
It's relatively easy to standardize on how to talk to a relational
database (that PEP is nice and solid), but looking at a grand unified
way of talking to so many different paradigms of GUI library, writing
bindings, testing, etc. is a much larger, less sensical beast. The
cross-platform UIs like Swing or QT all feel slightly out-of-place on
many systems because the interface elements and design decisions are
different from platform to platform.

If you want a platform-integrated GUI, use Glade/GTK/Python or QT
Designer/Python in Linux, XCode/Interface Builder/PyObjC in OSX, and
even consider IronPython and the .Net GUI stuff on Windows. Heck, I've
done some Swing and Jython and it was not as bad as writing Swing in
Java.

Might I also suggest that if you are struggling to make a proper GUI
in Python, and a GUI is what you need, that Python is probably not the
right tool for you to be using to get the job done.

 

[azrael]

While your point is obviously good, it only applies if you
even care about building for OS X, Gnome etc. If all you
want is build on Windows (or Gnome or KDE or whatever) then
the cross-platformness of any proposed toolkit is worthless
and may even be a drawback if it reduces its functionality
to a lowest common denominator.

I have nothing really to say about GUI-building whatsits in
general; I don't have cause to use them myself so I simply
stand back and let their proponents fight it out. I just
wanted to point out that playing the cross-platform card
does not mean the game is over.

TJG

OK. If I don't use GUIs, is there anything other that I do. Is there
anything else where I can contribute. Maybe there should be a better
networking, crypto, database connection library. Or whatever.

Forget about Crossplatform Gui Building. That is something that the
developers from wxWidgets think about. Where is the point of python in
crossplatform compability if I download on windows machine a piece of
code writen for GTK and I don't have anything installed to support it.
Again I am going to make some downloads. Python is bundled with TK,
which is out of date.

 

[azrael]

The Python core will not ship with another GUI toolkit. TK may be
removed from the core in some distant future but we - the Python core
development team - will *not* make a choice between GTK, WX and Qt.


In your opinion all third party are bad. You like to have one monolithic
block of software. That's a typical Microsoft approach. Lot's of people
from the open source community prefer small and loosely coupled pieces
of software. One of the greatest ruler of all time once said "divide et
impera" -- divide it in small pieces to make it much easier to conquer
every piece by its own. We are following the same path here.

Nobody is going to stop you from creating a large bundle of useful
extensions as long as you follow the licenses. In fact lots of people
may appreciate a bundle. But the Python core package will always stay
small and agile.

Christian

I understand that. completely. I don't think that third party is bad.
It is great that someone has an Idea and decides to make a new
project. But why not adopting best practices or maybe even make a
small effort to construct better support for some popular and commonly
used modules. If these modules always stay third party, there will
come a day when they stop being developed. At this point , Python
looses a good module from beeing developed any further. And some time
later, someone else starts again from scratch and make something new.

I came to Python because it has some features that always annoyed me
in C++ or Java. The diference between Java and Python that Java is
today way much often used thaen python, has a much larger Library
collection. I am mainly afraid that Python will become just a small
Scripting languages. There are a lot of people that say that they
would start Python, but they dont't want to because they don't want to
get their results in Comand line. I don't say that Comandline is bad,
but I think about the common user.

 

[azrael]

There is no need to make it elephant size. Python takes only 14 MB if
I am not wrong. Compare 10 GB of VS package in compare with that.
nothing. Python enthought edition is something really sweet. For
starters, Why does Python not have a build in library to handle
images. I don't get this. Why? PIL is great, but it could really need
some polish. I think that the last programming language I can remind
of that could not handle images, If I am not wrong was QBASIC.

All I hear when I talk to people who own or eork in SW companies is
"Python? Isn't that that small scripting language. This will never
bring a real application." I am tired of hearing this. Even Microsoft
implemented Python. Only because the GUI building I am thinking about
of moving to IronPython. But this great comunity holds me bac on
CPython.

We need better and not out of date libraries and modules. Python was
build because of bad experiences of bad programing languages. I sak
The Python Development team to listen to us users an developers.
listen to our problems.

 

[Damon]

Let me know when Visual Studio tries to address building GUIs for  
Windows, OS X, Gnome, and KDE. Until then, you're comparing apples to  
oranges, or chalk and cheese if you're from that part of the world.

Right now.

Use Visual Studio to program a .net application using Windows Forms.
Run it in Mono on Windows, or OS X, or Linux. You can even do it with
IronPython. Or, if you prefer a native look on each application, do it
with wx.net. The great thing about IronPython in .net is that you can
just link in any other .net library in whatever language it's written
in, without needing any sort of wrapper.

Admittedly, I haven't tried this, I'm just going by the hype I read on
the Mono and wx.net pages. Also, you can quibble whether it's Visual
Studio that tries to address this, or the Mono effort.

Damon.

 

[Damon]

In your opinion all third party are bad. You like to have one monolithic
block of software. That's a typical Microsoft approach. Lot's of people
from the open source community prefer small and loosely coupled pieces
of software. One of the greatest ruler of all time once said "divide et
impera" -- divide it in small pieces to make it much easier to conquer
every piece by its own.

"Divide et impera" was a strategy used by the monolithic Roman empire
to divide and subjugate others. I hope that's not what happens to
Python!

The original poster complained about needing to go off to third-party
sites to hunt for software. I wonder if the Python team has ever
considered following the lead of miktex or R, and setting up a
centralized (mirrored) repository of packages? Let anyone who wants,
submit packages.

* Like R, every time there is a new version of Python, the repository
should rebuild the packages, for all supported platforms, and make
available all those that compile cleanly. R also forces you to write
properly structured documentation for every exposed function, before
the repository will accept it.

* Like miktex, when the user does "import foo", the Python interpreter
should (optionally) look for foo in its cached list of packages, and
download the latest version if necessary.

This works fine for R and miktex, and it has made my life very easy.

Damon.

 

[David Cournapeau]

The original poster complained about needing to go off to third-party
sites to hunt for software. I wonder if the Python team has ever
considered following the lead of miktex or R, and setting up a
centralized (mirrored) repository of packages? Let anyone who wants,
submit packages.

Of course they have, that's what pypi is for. Pypi does not solve all
the problems, though.

* Like R, every time there is a new version of Python, the repository
should rebuild the packages, for all supported platforms, and make
available all those that compile cleanly. R also forces you to write
properly structured documentation for every exposed function, before
the repository will accept it.

* Like miktex, when the user does "import foo", the Python interpreter
should (optionally) look for foo in its cached list of packages, and
download the latest version if necessary.

This works fine for R and miktex, and it has made my life very easy.

Yes, R in particular is very nice for this. But it is a complex
problem - and it is even more complex for python compared to R and
miktex, because python is more general. In particular, in R (and
miktex I guess ?), you rarely care about having several versions
installed at the same time (interpreter and/or packages), but in
python, this is often useful. This complicates matter quite a bit.
Having automatic download at import who annoy many users, me included.

For the time being, my opinion is that the only solution for
deployment on desktop apps on windows/mac os X is something like
Enthought distribution, which is self contained. But when you start
installing something outside Enthought, it can become tricky for
"average users" (the one who don't care about the technicalities).

cheers,

David

 

[Terry Reedy]

The original poster complained about needing to go off to third-party
sites to hunt for software. I wonder if the Python team has ever

The 'Python team' is everyone who volunteers to help.
How about you?

considered following the lead of miktex or R, and setting up a
centralized (mirrored) repository of packages? Let anyone who wants,
submit packages.

* Like R, every time there is a new version of Python, the repository
should rebuild the packages, for all supported platforms, and make
available all those that compile cleanly. R also forces you to write
properly structured documentation for every exposed function, before
the repository will accept it.

* Like miktex, when the user does "import foo", the Python interpreter
should (optionally) look for foo in its cached list of packages, and
download the latest version if necessary.

This works fine for R and miktex, and it has made my life very easy.

If you have an idea on how to improve the current package index, help
improve it. Appoint yourself the 'test with new release' manager.

tjr

 

[Terry Reedy]

We need better and not out of date libraries and modules. Python was
build because of bad experiences of bad programing languages. I sak
The Python Development team to listen to us users an developers.
listen to our problems.

I ask you to help update a module or package that you care about.
Join the team and contribute.

tjr

 

[Marco Mariani]

I know that there is already a standard python library, But
why not extending it. classify the standard library into subcategories
like Networking, DataBase, Computation, ......

If the standard library where that huge, python 3.0 would have been late
by a couple of years.

Why not using this number of people and accomplish something great. If
anyone of us would write 10 line of good code, it would result a very
great and powerfull environment.

I totally want to write my 10 lines of great image processing or speech
recognition software, but it's not how development works.

 

[Martin]

Martin wrote:
[typos igored as requested ]

How does "small and agile" work with "batteries included"?

The Python slogan says "batteries included", not "fusion reactor included".

I'd be fine with a fusion reactor, my objections would be if skynet
was included

The rules are:

...

It takes at least 1.5 years to get a new feature into an extension and
at least 3 years to remove or change a feature. That's a major shop
stopper for every fast moving piece of Python software.

True, I thought a bit more about it and decided for myself to be happy
with the way it is since I prefer quality to features


/Martin




--
http://soup.alt.delete.co.at
http://www.xing.com/profile/Martin_Marcher
http://www.linkedin.com/in/martinmarcher

You are not free to read this message,
by doing so, you have violated my licence
and are required to urinate publicly. Thank you.

Please avoid sending me Word or PowerPoint attachments.
See http://www.gnu.org/philosophy/no-word-attachments.html

 

[Simon Hibbs]

All I hear when I talk to people who own or eork in SW companies is
"Python? Isn't that that small scripting language. This will never
bring a real application." I am tired of hearing this. Even Microsoft
implemented Python. Only because the GUI building I am thinking about
of moving to IronPython. But this great comunity holds me bac on
CPython.

I don't agree at all that Python can't compete with Visual Studio for
GUI development. There are Python modules for Eclipse if you need an
enterprise class IDE. For GUI design the PyQT GUI toolkit includes QT
Designer - a fully featured graphical GUI builder every bit as capable
as the one in Visual Studio. In fact I much prefer working in QT
Designer for GUI layout, the Signals/Slots model for hooking up to the
application logic is just so much more elegant and robust.

If Eclipse is a bit heavyweight for you, and you're after a Pyhton
equivalent to VB, I can't recommend Eric enough. It's a Python IDE
written in Python using the PyQT toolkit, and integrates directly with
QTDesigner. IMHO it's far superior to anything Microsoft has to offer,
with full native support for MacOS X, Linux and Windows.

The only fly in the ointment in licensing. The QT toolkit will be
fully GPL and LGPL from version 4.4, but PyQT itself has a dual GPL/
commercial licensing structure. Still, it's relatively cheap and well
worth it if you're goingt to do commercial development.

Simon Hibbs

 

[Basilisk96]

* Like R, every time there is a new version of Python, the repository
should rebuild the packages, for all supported platforms, and make
available all those that compile cleanly. R also forces you to write
properly structured documentation for every exposed function, before
the repository will accept it.

A very good idea indeed. I would love to start using Py3k today, but I
am still on 2.5 because I depend on win32 binaries for the majority of
my library packages. I can build some myself, but not all. A
repository of prebuilt binaries that stay in step with currently
available language releases would be most welcome.

Just my $0.02,
-Basilisk96