On Wed, 26 Jul 2006, Jorge Vargas wrote:
> On 7/26/06, Martin <[EMAIL PROTECTED]> wrote:
>>
>> I mean I am concerned that I could end up getting stuck in
>> idiosyncrasies of _javascript_ and browser dependent behaviour. My
>> instinct is to awoid hands on approach to _javascript_ technology and
>> instead focus on GWT provided offcourse that GWT has no major bugs
>> offcourse. With Eclipse, it's debugging and productivity of Java the
>> reasons are compelling (provided GWT works well).
GWT does look nice.
You're right to be afraid of vanilla JS, of course, but it's not necessary
to jump all the way to something like GWT (though I've been grumbling for
years there's no reason that sort of thing couldn't be done, so it's nice
to see people actually going that way seriously). No doubt GWT's approach
will have its own costs, especially in the near-term, which will differ
from those of the Mochikit / TG approach. At this point, both approaches
seem likely to be reasonable trade-offs.
my biggest fear of it is that during coding/debug your running a native java app (with SWT) and when run in production your not only on a diferent machine but language and code, and even though it's google that's scary.
On the other hand GWT create a code of each browser so it takes out all the crapy branching of plain _javascript_ but then you actually have X diferent codes, so if an error does gets out you need to figure out on which one was in order to report a bug.
Of course, using GWT means you don't get to use Python. The productivity
of Java-the-language isn't usually regarded as compelling by Python users
;-)
hehehe, yea and the best part is that GWT isn't 100% of java's API, which IMO it's is best feature, is just a subset.
For the future:
http://pyjamas.pyworks.org/
http://pyworks.org/mailman/listinfo/py-gwt
http://wiki.python.org/moin/SummerOfCode
I'll look into those.
Compilation of Python to JS in a (now defunct) Python web framework has
already been done, of course.
well that shows how practical it is :)
Of more practical interest, though, is that the PyPy people have a JS backend for RPython. PyPy is run by some very
serious and talented core Python people: this is not a throw-away project,
so it seems quite plausible that in the not-too-distant future TG will
move in that direction, or at least grow support for that way of working.
TG has already shown itself very capable of keeping up with new
developments, and I'm sure it'll continue to do that nicely even after 1.0
and the constraints of strict backwards-compatibility. As a useable web
framework, though, this is of course vapourware, unlike GWT!
I don't agree that GWT is not vapourware, GWT is there because someone at google wanted it to be, not because people use it everyday, on the other hand TG is here because of Kevin but is still here and growing because we are using it, and will keep using it.
And hey, Guido (creator of Python) is doing web development at Google, so
who knows where that might lead!-)
who knows, but I don't think he is developing a webframework....
> The main objective of mochikit is to take out browser to take out all the
> platform dependancies.
...and add some of the missing basic stuff that should have been in JS
from the start. Which involves making JS look more like Python, of course
:-) (FWIW, Brendan Eich obviously shares that sentiment, in spades, as
you'll see if you read his plans for JS 2).
I believe that back then those where not necessary, _javascript_'s history is too hard to follow, as for JS2 there will probably be something better before that is finish/widly addopted.
for example this http://weblogs.mozillazine.org/roadmap/archives/008865.html
ones we get python on firefox nothing will stop us :D
> about ecplise many people say that pydev (one of
> eclipse python's plugin) is the best IDE for python outhere, and recently
> it's author made a blogentry or video I don't know about setting it up to
> work with TurboGears.
A nice IDE it may be, but Python doesn't gain the same benefits from
Eclipse that Java does, AFAICT. Python seems too dynamic to fit perfectly
into the semantic editing thing. Net of everything, Python still comes
out well on top, of course ;-)
IMO the best IDE for python is none,
the only 2 features of an ide that are usefull within python
- is code compleating which almost all text editors have today.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "TurboGears" 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
-~----------~----~----~----~------~----~------~--~---
