Hi,
I realise this thread is getting long already, but I'd like to chip
in. First a couple of thoughts about life on internet mailing lists:
I think what percious is getting at is a complaint I often have. If
you post a message saying "Hi, I've written this, it's really alpha,
but I'd love some comments" then you get no replies. If you post
"TurboGears sucks, you're all fools" you get about 20 really long
replies. This is just life though - those of you who've worked in big
corporates will know this well. It happens on a bigger scale too - we
never hear about the hundreds of things politicians get right, we just
hear again and again about the odd slip up.
Somewhere along the way someone used the word "idiot". I don't think
anyone meant offense by this, it was just a shorter way of saying
"less experienced user". I think we all agree that anyone who
downloads a Python web development library is far from an idiot by any
reasonable standard. You should have seen my former next door
neighbour, that's a real idiot! Again, it's just life that you can
write a long, well-considered message, drop in one bit of emotive
language, and suddenly everyone focuses on that, not all your good
points.
Now, considering TurboGears in particular:
I've generally had the impression that TG is not targeted at people
new to web development ("idiots" for short ;-). Sure, it's a worthy
goal to try and accomodate such people. But I think the main target of
TG is people who already know some Python, SQL, JavaScript, etc. And
given that there's limited development resource, I think it's
important to understand our main audience.
There are several design decisions that have come up many times.
Whether to have a favored JavaScript library is one. Another is
whether to favor raw SQLAlchemy or Elixir. What it may be worth doing,
and apoligies if someone has already done this, is having a page that
summarises such previous discussions. If someone posts on such an
issue, we can refer them to that page, and invite them to post again
if they feel they have something substantive to add.
I've come to think that TurboGears is actually taking a very difficult
approach to framework development. Pylons mostly avoids these issues
by being fully pluggable and not favoring anything (ok, that's not
100% accurate, but you get the idea). Django mostly avoids these
issues by being fully monolithic (ok, the less tightly bound
components like templates are pluggable). While I think the TG
approach of "pluggable with defaults" is wonderful for an app
developer, it's an absolute PITA for the framework developers. As
Florent mentioned, dealing with upstream bugs takes a lot more work
than fixing bugs in a monolithic framework. New versions of upstream
packages can cause problems. And there's the question of
documentation, deferring to the upstream package documentation is
often less than optimal.
Did anyone read to the end? :-)
Paul
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"TurboGears Trunk" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/turbogears-trunk?hl=en
-~----------~----~----~----~------~----~------~--~---
