Newbie question about the future of TurboGears

[Hannes]

Hi,

After a lot of research on Python frameworks and briefly trying
Django, Web2Py and (very briefly) Pylons and Flask, Eventually I
decided to learn TG2.1 as I always wanted to go deeper into SQLAlchemy
based full-stack frameworks, particularly Pylons. So I was surprised
so see that Pylons is discontinued and is going to be absorbed in
Pyramid. And I was even more frustrated when my first impression with
Pyramid was less than satisfactory (Maybe too complicated or poorly
documented for a beginner?). Hence I chosen TG. and so far so good. I
am really enjoying it :)

Now I am wondering where TG is aimed to go from here? And does it
really worth to learn TG, while it is based on a stagnant Pylons?

I know that mako, SQLAlchemi, paster, etc don't go anywhere but
nevertheless each framework has its own quirks that take a while for a
developer to master and I guess it is not a good practice to spend one
month on learning conventions that will be abandoned at the end of the
second month, hence I am really looking for you guys in TG project to
give me a prospect of TG project. I hope this prospect at least can
convince me that learning TG is less time-wasting than learning Pylons
1.0 :)

My apologies if this topic has already been raised elsewhere.
Cheers

[Michael Pedersen]

Ah, the questions I wasn't looking to answer *just* yet. I suppose it was only a matter of time, though.

TurboGears has two major releases out right now: TG1 and TG2. TG1 is converging into a 1.5 version. I don't know what will happen for that version after 1.5 is out. Christoph is leading that effort, so any questions about the future of TG1 are best left directed to him.

TG2, on the other hand, I've got a road map for, at least mentally. Specific details, of course, are subject to change, but the general layout of the land is that we have three major projects already on the plate for TG2.

1. Project cleanup. This will be 2.2 when it's done. This involves a bit more website work (specifically: getting buildbot working), getting our testing coverage to 100%, and getting our documentation re-done to the point that all issues are cleared up and it covers the entire lifecycle, along with alternatives and extensions.
2. Performance enhancement. The system works fairly well right now, but it can be improved. That all by itself will be a full version bump, mainly because we could introduce backward compatibility issues. I'm trying very hard to be very careful about that.
3. Extension development. We need more extensions to help make websites more easily. We have some good ones now, but expanding the repertoire will be very useful. In addition, we'll be bringing some of the more major ones into the TG process (automated testing, documentation, etc).

I'm not sure if performance enhancement should be before or after extension development, honestly. I don't want to break extensions, but I don't want to delay extensions. I'm holding off on making that choice until we get 2.2 release candidates out the door, though.

After that, the future is not so clear to me. Right now, Mark is working on Orion, which is a TG type stack on top of Pyramid. We should also begin looking at working on Python 3 compatibility somewhere in there.

Pylons, and TG, are far from dead. Very far. Pylons has gone into maintenance mode, which only means we now have a stable platform to work with, and can file bug reports as needed. They've already committed to keeping bug fixes being released. TG, on top of Pylons, is still undergoing active development and will be maintained for years to come. Hell, I'm hoping to make a career out of it, personally.

So, is TG worth your time? Only you can decide that. I know it's worth my time, though. I'm not going anywhere.

--
You received this message because you are subscribed to the Google Groups "TurboGears" group.
To post to this group, send email to turbo...@googlegroups.com.
To unsubscribe from this group, send email to turbogears+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/turbogears?hl=en.

--
Michael J. Pedersen
My IM IDs: Jabber/pede...@icelus.tzo.com, ICQ/103345809, AIM/pedermj022171
          Yahoo/pedermj2002, MSN/pederm...@hotmail.com
My LinkedIn Profile: http://www.linkedin.com/in/michaeljpedersen

[Alessandro Molina]

On Thu, Apr 28, 2011 at 12:55 PM, Hannes <esi...@gmail.com> wrote:
> So I was surprised
> so see that Pylons is discontinued and is going to be absorbed in
> Pyramid.

Actually Pylons has just gone in legacy mode.
This means just that instead of having Pylons2.0 (which was intended
to be a rewrite of pylons) we will have Pyramid.

But Pylons1.0 will continue to exist. It won't get new features, but
will be bug-fixed.
New features will be demanded to higher levels frameworks that work
over pylons, like TG.

> And I was even more frustrated when my first impression with
> Pyramid was less than satisfactory (Maybe too complicated or poorly
> documented for a beginner?). Hence I chosen TG. and so far so good. I
> am really enjoying it :)
>

Personally I got to "hate" Zope (it was the reason that caused me to
switch from Python to RoR for web development long time ago) and I
have a strange feeling about Pyramid. They technically did an
excellent job and It is indeed a really good project but to me it
"smells bad", like a dish that I might recognize that it is well made
but I really don't like how it tastes :D
TG has been the first framework that made my say "woah! This is
exactly how a web framework should be!" and so personally I'll be
commited to maintaining and improving it for long time.

I think that if Pylons someday won't fit TG anymore for any kind of
reason we will have to take a decision without breaking compatibility
with existing applications, but this will be far away in the future as
currently the Pylons team is still maintaining Pylons itself.

[Michael Pedersen]

On Thu, Apr 28, 2011 at 11:30 AM, Alessandro Molina <alessand...@gmail.com> wrote:

Personally I got to "hate" Zope (it was the reason that caused me to
switch from Python to RoR for web development long time ago) and I
have a strange feeling about Pyramid. They technically did an
excellent job and It is indeed a really good project but to me it
"smells bad", like a dish that I might recognize that it is well made
but I really don't like how it tastes :D

The only thing I can add to this is to ask a favor: Let's not bash other frameworks. Each of them has their own strengths and weaknesses that make them suited for different tasks.
 

TG has been the first framework that made my say "woah! This is
exactly how a web framework should be!" and so personally I'll be
commited to maintaining and improving it for long time.

I had the exact same reaction. I tried others, but all of them failed to fit my mental model of how things should work. TG fit, perfectly. I look at it, and the code makes sense. I know that I can go through the core, and see it make sense.

I'm glad to be helping TG move forward.

--

[Alessandro Molina]

On Thu, Apr 28, 2011 at 5:49 PM, Michael Pedersen <m.ped...@icelus.org> wrote:
> On Thu, Apr 28, 2011 at 11:30 AM, Alessandro Molina
> <alessand...@gmail.com> wrote:
>>
>> Personally I got to "hate" Zope (it was the reason that caused me to
>> switch from Python to RoR for web development long time ago) and I
>> have a strange feeling about Pyramid. They technically did an
>> excellent job and It is indeed a really good project but to me it
>> "smells bad", like a dish that I might recognize that it is well made
>> but I really don't like how it tastes :D
>
> The only thing I can add to this is to ask a favor: Let's not bash other
> frameworks. Each of them has their own strengths and weaknesses that make
> them suited for different tasks.
>

Sorry, didn't mean to bash any of them,
was just trying to express the fact that my mindset is far from their
way to do things but I recognize that they are excellent projects