The Future of TG2

[Michael Pedersen]

Before I talk about the future, I'd like to look at the past a little bit.

TurboGears has a good history. It's one of the older web frameworks in the
Python world, and it managed to have a decent following for a long time. You
can see that by looking at the history of the message counts for this ML.
They're over at http://groups.google.com/group/turbogears-trunk/about

The strongest period of time on the list was from Nov 2008 to Mar 2009.
Ironically, Mar 2009 was when I first started looking at TurboGears. I'm not
sure what happened, but traffic on the ML took a major dive that month. From
there, a slow but steady decline is visible. In fact, most of 2010 shows
traffic to be at an all time low, with May not even having a single post.

My ultimate goal is to fix that. I want TG to become what it once was in
popularity. I want to make TG into something people look at and say "Wow".
As it stands right now, we're a long way from that. We can fix that, though.
To that end, here's my current plan:

First: Complete the migration from the current server onto
beta.turbogears.org. Move the live tickets into SF.net's Allure platform.
Give us a new face, and start using our own product.
Second: Release 2.0.4. We have some bug fixes in place already, we just have
to complete any remaining tickets and do the release.
Third: Release 2.1.1. Same deal, we just have to complete the mandatory
minimum tickets for it.

I hope to have *all* of that done by the end of April.

Once we've accomplished that, it's time to begin working towards 2.2.0. This
is where the work will become difficult. I have a number of goals for 2.2.0.

- Bring testing coverage to 100%
- Improve documentation. Overall goal is to make docs into a book in a
few major parts: Tutorial (take a project from idea to maintenance),
alternatives and extensions (to help get projects moving quickly), and
finally a reference section at the end.
- Close out all bugs that can be closed without introducing backward
incompatibility *or* deprecation warnings.
- New features will be limited to things that don't introduce backward
incompatibility *or* deprecation warnings

I want to release 2.2.0 by the end of this year. Between now and then, I
plan to release incremental improvements to 2.1, so that we can enjoy the
benefits of the progress. I'm hoping that these releases will help to bring
TG2 back onto the radar for python web developers.

So, there you have it. That's my goals and plans. What do you all think?
--
Michael J. Pedersen
My IM IDs: Jabber/***@icelus.tzo.com, ICQ/103345809, AIM/pedermj022171
Yahoo/pedermj2002, MSN/***@hotmail.com

[Andy Bierman]

Post by Michael Pedersen
Before I talk about the future, I'd like to look at the past a little bit.

my 2 cents:

Flexibility has diluted the main appeal of TG - simplified WEB development.

When I started with TG1 there was only 1 set of documentation so there
was no way to get derailed by the subtle differences between branches.
Assuming 1.0 docs would work in 2.0 or 2.1 is problematic.

The original 20 minute wiki was a great tutorial and got me started
very fast. Now, TG demos almost work, but you probably need to google for
answers to get all the details to actually work.

I want to leverage the expertise of you developers wrt/
best current practices. I don't want to become an expert
in a dozen different opensource packages in order to use TG;
The 80/20 rule applies -- if I stick to commonly used features
I should not need to be an expert in that package.
TG actually shines in this area, but it is not obvious. You really have
to dig to find all the value.

All these branches causes confusion for users and makes people
think TG is not really stable. IMO, optimize for new projects
in TG 2.2. Optimizing for TG1 porting is not nearly as important.
I do not think there is a market for 1.5 or 2.0 (2.1 was clearly
making 2.0 obsolete before most people had a chance to look into it.)

There are so many extensions and widgets and tools that it is not
usually obvious what feature/package to use. Build out the core of
SO, genshi, tw2, etc. so rapid prototyping is a reality again,
like it was in TG1. Moving on the next bleeding edge gizmo before
the masses have deployed the last one is the definition of instability.

I am still planning to move my TG1 project to 2.1 but my day job
gets in the way. I suspect it will be way easier now because
work-in-progress back then is done now, and the docs have been updated.


thanks,
Andy

-

Post by Michael Pedersen
TurboGears has a good history. It's one of the older web frameworks in the Python world, and it managed to have a decent following for a long time. You can see that by looking at the history of the
message counts for this ML. They're over at http://groups.google.com/group/turbogears-trunk/about
The strongest period of time on the list was from Nov 2008 to Mar 2009. Ironically, Mar 2009 was when I first started looking at TurboGears. I'm not sure what happened, but traffic on the ML took a
major dive that month. From there, a slow but steady decline is visible. In fact, most of 2010 shows traffic to be at an all time low, with May not even having a single post.
My ultimate goal is to fix that. I want TG to become what it once was in popularity. I want to make TG into something people look at and say "Wow". As it stands right now, we're a long way from that.
First: Complete the migration from the current server onto beta.turbogears.org <http://beta.turbogears.org>. Move the live tickets into SF.net's Allure platform. Give us a new face, and start using
our own product.
Second: Release 2.0.4. We have some bug fixes in place already, we just have to complete any remaining tickets and do the release.
Third: Release 2.1.1. Same deal, we just have to complete the mandatory minimum tickets for it.
I hope to have *all* of that done by the end of April.
Once we've accomplished that, it's time to begin working towards 2.2.0. This is where the work will become difficult. I have a number of goals for 2.2.0.
* Bring testing coverage to 100%
* Improve documentation. Overall goal is to make docs into a book in a few major parts: Tutorial (take a project from idea to maintenance), alternatives and extensions (to help get projects
moving quickly), and finally a reference section at the end.
* Close out all bugs that can be closed without introducing backward incompatibility *or* deprecation warnings.
* New features will be limited to things that don't introduce backward incompatibility *or* deprecation warnings
I want to release 2.2.0 by the end of this year. Between now and then, I plan to release incremental improvements to 2.1, so that we can enjoy the benefits of the progress. I'm hoping that these
releases will help to bring TG2 back onto the radar for python web developers.
So, there you have it. That's my goals and plans. What do you all think?
--
Michael J. Pedersen
--
You received this message because you are subscribed to the Google Groups "TurboGears Trunk" group.
For more options, visit this group at http://groups.google.com/group/turbogears-trunk?hl=en.

...