web.py superiority and it outstanding simplicity is mostly hidden!

21 views
Skip to first unread message

Tzury

unread,
Aug 23, 2007, 4:40:25 AM8/23/07
to web.py
People simply don't know how great and powerful yet simple our
framework is.
If you think web.py is a great python-web-framework, and disagree with
the terms used at this blog's comments regarding web.py, you may
consider saying it out loud over there:
http://blog.ianbicking.org/2007/08/21/the-shrinking-python-web-framework-world/

lui...@gmail.com

unread,
Aug 23, 2007, 8:29:11 AM8/23/07
to web.py

> consider saying it out loud over there:http://blog.ianbicking.org/2007/08/21/the-shrinking-python-web-framew...


Webpy is great, but it cannot be compared to Django or TG.
This is simply because webpy is the "anti-framework". It's different,
it's not a full stack framework and it's not a bunch of components
glued together.
It's just a simple, lightweight, consistent way to make web apps in
python that can be delployed everywhere.
I believe this is what makes it great.

I'm sure Django, TG, Pylons, tec... are great too, and many people are
happy with them.
They just aren't for me.

Luis


Ben Wilson

unread,
Aug 23, 2007, 8:40:49 AM8/23/07
to we...@googlegroups.com
On 8/23/07, Tzury <Afro.S...@gmail.com> wrote:
>
> People simply don't know how great and powerful yet simple our
> framework is.
> If you think web.py is a great python-web-framework, and disagree with
> the terms used at this blog's comments regarding web.py, you may
> consider saying it out loud over there:

The general theme of that blog post is that the Python frameworks
field is starting to congeal into a few popular frameworks, and that
other developers should adopt the more popular frameworks to unify the
field and push back on RoR inroads.

I have to say that I generally agree with the blog, but I prefer to
use webpy because it meets my requirements. The author is not claiming
all other frameworks are inferior, per se, but that popularity trumps.
Logically, it's argumentum ad populum. Politically, it makes sense to
rally around the flag and repel the RoR invaders (so to speak). My
caveat is that flag should be critically reviewed and re-engineered as
necessary to make it pythonic.

--
Ben Wilson
"Words are the only thing which will last forever" Churchill

Tzury

unread,
Aug 23, 2007, 8:40:49 AM8/23/07
to web.py
What I didn't like was a comment of Resig reffer to web.py as one
programmer using his own framework. That is a minimizing action for
"great framework for those who hates framework"

lui...@gmail.com

unread,
Aug 23, 2007, 9:58:17 AM8/23/07
to web.py

I don't think this is a disdainful comment.
I guess the community realised that unless there is a one and only
"blessed" framework, Python will keep on losing positions against
other solutions, such as RoR.
When GvR blessed Django, I was surprised, because it seemed to me that
he was unfair with other frameworks, but if you think twice, it makes
a lot of sense. He just had to make a decision.

Django seems to be the logical choice because:
1) It's actively developed by its authors.
2) It's coherent and it's not in constant flux, like TG (they change
templates systems all the time, ORMs and now they're planning to
switch from CherryPy to Pylons...) . So if I were GvR, I wouldn't bet
on that experiment...

As for Webpy, it's great. But it's not a framework.
As Aaron said, it's the "anti-framework", and I guess this leaves it
out of the competition for the all-in-one python framework.

I started playing with it a few days ago, and I like it very much.
It's simple and lightweight, and it gives me total freedom to use
whatever components I want.
I find it very similar to mod_python's publisher handle, but it can be
deployed everywhere, on any server and with any available technology
such as fastcgi, scgi, mod_wsgi, etc, etc...
And it just requires one file, that could be placed anywhere.

For the same reasons I find it beautiful, others might find it
unsuitable.
Freedom of choice is a great thing. And it's good to have a powerful
secret weapon, don't you think?

Luis

Angelo Gladding

unread,
Aug 25, 2007, 8:12:07 PM8/25/07
to we...@googlegroups.com
Resig & jQuery

What I didn't like was a comment of Resig reffer to web.py as one programmer using his own framework.

Resig is the creator of jQuery. Browse the group's mailing list for a taste of what he's talking about.

jQuery -- 3207 members; "high" activity; 44095 posts
web.py -- 606 members; "low" activity; 3850 posts

I posted a question regarding an unclear use of a basic function and not only was the question solved immediately by Resig, many other people had chimed in. I pointed out that the documentation was unclear. Resig fixed it on the spot. This kind of activity is what a large, supportive community can offer a framework (or an anti-framework).

Ian Bicking & Uniformity

So people want a one-size-fits-all framework to compete with RoR? Why? The ability for a designer to pick up RoR and output a "Web 2.0" app overnight is fabulous. Does Python need to conform to this ideology as well? If so Django is probably the best fit but why must we chose one and stick with it?

Wasn't WSGI designed to provide a uniform base in which frameworks of all sorts could build interswappable components for? That said, isn't web.py really pushing the framework ecosystem in the right direction? I'd like to think so at least :). Maybe Aaron, Anand, or anyone else with more knowledge of WSGI can chime in.

Documentation & webpy.org

As for Guido's initial grade of "F" for documentation.. How far have we come since then? Are we at least passing by now? Sadly, I think not.. Where's `teh communicator`? Is it in progress through Google's Summer of Code? Where do we stand with the other ideas? As for the core documentation, where does that stand?

As for Resig's "one programmer using his own framework", how can we deny his statement short of proof? Can the following list be expanded? Can openlibrary.org be pushed to this list? In my opinion Micropledge's "fitting in" with the minimalist approach may be a potential turn off as people generally desire freedom rather than constraint.

Who uses web.py?

reddit.com, one of the top 1000 sites according to Alexa, uses web.py to serve its millions of daily page views. "It's the anti-framework framework. web.py doesn't get in your way," explains founder Steve Huffman. (Disclosure: web.py creator Aaron Swartz was also a founder of reddit.)

colr.org, a color scheme picking site, is built in web.py.

Yandex, a Russian traffic provider whose homepage alone receives 70 million daily page views, uses web.py for certain projects.

LShift has used web.py to build websites for Expro and publisher Dorling Kindersley. "web.py allows us to do what we do best," they report. "It does the webapp thing brilliantly, and without requiring us to compromise on flexibility and originality."

micropledge, a web app that collects funding for software ideas, is built in web.py. "We've enjoyed fitting in with its minimalist approach," says developer Ben Hoyt.

Aaron, might you help me retrieve my account password for the newly implemented wiki? Username: angelogladding. I don't remember registering @ webpy.org but I did have an account previously with infogami.com. Any help would be appreciated.

What do we, the web.py community, have to do to boost our reputation in the overall Python web community? I believe something close to `the communicator` as well as more accessible documentation (easier to browse and edit) will be a good start. Furthermore, I believe we need to describe the benefits of this "anti-framework" while comparing and contrasting other current solutions. We should also push for work on the "infrastructure" improvements listed on the ideas page as well. Essentially they are modules, components, plugins, whatever you want to call them. These seem to be the backbone of other successful frameworks. Seeing as openlibrary.org is an open project built on web.py, is there any code that can be offered from it to the greater community?

Anyone else agree, disagree, have any comments or ideas? I look forward to rallied support for taking web.py to the next level.

On 8/23/07, Tzury <Afro.S...@gmail.com> wrote:

bubblboy

unread,
Aug 25, 2007, 11:20:40 PM8/25/07
to we...@googlegroups.com

Actually, in my opinion that is exactly what web.py is; "a developper
using his own framework". Except that it is not one developper but a lot
of us, making our own frameworks, customized for every web application.
I prefer web.py to most other frameworks because I can think of a
website as a real python program, not as a "django program" or anything.
Maybe I am just repeating statements, but to me this comment describes
very well what web.py is about.

I hope I did not offend anyone by expressing myself this way ;) no harm
intended.

Greetings,
b^4

Angelo Gladding

unread,
Aug 26, 2007, 7:39:33 AM8/26/07
to we...@googlegroups.com
Actually, in my opinion that is exactly what web.py is; "a developper using his own framework". Except that it is not one developper but a lot of us, making our own frameworks, customized for every web application.

Unfortunately, I don't believe that was the tone the originator was using.. But well put and I agree!

As for my previous request on the status of `teh communicator`, I was just the second official download of the 0.01 of seddit which is the Google SOC project I was really looking for. Sooo, Aaron, is it usable and are there plans for a hosted instance @ webpy.org?
--
Angelo Gladding

Aaron Swartz

unread,
Aug 28, 2007, 9:19:54 PM8/28/07
to we...@googlegroups.com
Sorry, infogami.com accounts weren't carried over to the new site.

I think we still have a long way to go on documentation.

Aaron Swartz

unread,
Aug 28, 2007, 9:20:33 PM8/28/07
to we...@googlegroups.com
> As for my previous request on the status of `teh communicator`, I was just
> the second official download of the 0.01 of seddit which is the Google SOC
> project I was really looking for. Sooo, Aaron, is it usable and are there
> plans for a hosted instance @ webpy.org?

It still needs some work, but I'd love to host it for webpy.org when it's ready.

Adam Atlas

unread,
Aug 29, 2007, 4:26:16 PM8/29/07
to we...@googlegroups.com

On 26 Aug 2007, at 07.39, Angelo Gladding wrote:
> As for my previous request on the status of `teh communicator`, I
> was just the second official download of the 0.01 of seddit which
> is the Google SOC project I was really looking for. Sooo, Aaron, is
> it usable and are there plans for a hosted instance @ webpy.org?

Aaron, speaking of SOC projects -- what happened with the web stats
analysis package? I was interested in that... it looks like someone
picked it up, but I can't find anything on it (other than the
unchanged original proposal). Did that actually get made?

Berwyn Hoyt

unread,
Sep 4, 2007, 7:32:29 PM9/4/07
to we...@googlegroups.com
As Angelo and Aaron both mentioned (see below), web.py could do with better docs.  But I just don't see it happening without some fan support.

I'm thinking of creating a microPledge project to raise money to create the docs like they raised $17,000 for Ruby on Rails by merely accepting donations (though I don't imagine that we'll make it that high because Ruby has a much bigger following).  I know that using a new microPledge feature you can now "pledge now, and pay later" to get an indication of the level support "out there", but I'd like to open this for discussion here before I even start the project.

As for who would do the documentation, we at microPledge would be willing to do it as a job, and you can assess the quality of our docs.  But we'd be happy if others quoted the job, too.  After all, the object is to make the docs, not the money.

What do you think?

- Berwyn

Tommi Raivio

unread,
Sep 5, 2007, 1:13:25 AM9/5/07
to we...@googlegroups.com
I have given this a lot of thought, right from the beginning. Some
selected pointers:

* What is the average user of web.py like? Does she enjoy reading source
code? Can he program?
* How does one measure a great documentation? What goals it should promote?
* Is less more in this particular case?

bubblboy

unread,
Sep 5, 2007, 2:42:32 AM9/5/07
to we...@googlegroups.com
Berwyn Hoyt wrote:
> As Angelo and Aaron both mentioned (see below), web.py could do with better
> docs. But I just don't see it happening without some fan support.
>
> I'm thinking of creating a microPledge <http://micropledge.com> project to raise
> money to create the docs like they raised $17,000 for Ruby on Rails
> <http://www.pledgie.com/campaigns/34> by merely accepting donations (though I
> don't imagine that we'll make it that high because Ruby has a much bigger
> following). I know that using a new microPledge feature you can now /"pledge
> now, and pay later" /to get an indication of the level support "out there", but
> I'd like to open this for discussion here before I even start the project.
>
> As for who would do the documentation, we at microPledge would be willing to do
> it as a job, and you can assess the quality of our docs
> <http://micropledge.com/help>. But we'd be happy if others quoted the job,
> too. After all, the object is to make the docs, not the money.
>
> What do you think?
>
> - Berwyn
>
>>
>> Documentation & webpy.org <http://webpy.org>

>>
>> As for Guido's initial grade of "F" for documentation.. How far have we come
>> since then? Are we at least passing by now? Sadly, I think not.. Where's `teh
>> communicator`? Is it in progress through Google's Summer of Code? Where do we
>> stand with the other ideas <http://webpy.org/ideas>? As for the core
>> documentation, where does that stand?
>>
>> As for Resig's "one programmer using his own framework", how can we deny his
>> statement short of proof? Can the following list be expanded? Can
>> openlibrary.org <http://openlibrary.org> be pushed to this list? In my opinion
>> Micropledge's "fitting in" with the minimalist approach may be a potential
>> turn off as people generally desire freedom rather than constraint.
>>
>>
>> Who uses web.py?
>>
>> reddit.com <http://reddit.com/>, one of the top 1000 sites according to

>> Alexa, uses web.py to serve its millions of daily page views. "It's the
>> anti-framework framework. web.py doesn't get in your way," explains
>> founder Steve Huffman. (Disclosure: web.py creator Aaron Swartz was also a
>> founder of reddit.)
>>
>> colr.org <http://colr.org/>, a color scheme picking site, is built in web.py.
>>
>> Yandex <http://yandex.ru/>, a Russian traffic provider whose homepage

>> alone receives 70 million daily page views, uses web.py for certain projects.
>>
>> LShift <http://www.lshift.net/> has used web.py to build websites for
>> Expro <http://exproretail.com/> and publisher Dorling Kindersley
>> <http://travel.dk.com/>. "web.py allows us to do what we do best," they

>> report. "It does the webapp thing brilliantly, and without requiring us to
>> compromise on flexibility and originality."
>>
>> micropledge <http://micropledge.com/>, a web app that collects funding for

>> software ideas, is built in web.py. "We've enjoyed fitting in with its
>> minimalist approach," says developer Ben Hoyt.
>>
>> Aaron, might you help me retrieve my account password for the newly
>> implemented wiki? Username: angelogladding. I don't remember registering @
>> webpy.org <http://webpy.org> but I did have an account previously with
>> infogami.com <http://infogami.com>. Any help would be appreciated.

>>
>> What do we, the web.py community, have to do to boost our reputation in the
>> overall Python web community? I believe something close to `the communicator`
>> as well as more accessible documentation (easier to browse and edit) will be a
>> good start. Furthermore, I believe we need to describe the benefits of this
>> "anti-framework" while comparing and contrasting other current solutions. We
>> should also push for work on the "infrastructure" improvements listed on the
>> ideas page as well. Essentially they are modules, components, plugins,
>> whatever you want to call them. These seem to be the backbone of other
>> successful frameworks. Seeing as openlibrary.org <http://openlibrary.org> is
>> an open project built on web.py, is there any code that can be offered from it
>> to the greater community?
>>
>> Anyone else agree, disagree, have any comments or ideas? I look forward to
>> rallied support for taking web.py to the next level.
>>
>> On 8/23/07, *Tzury* <Afro.S...@gmail.com <mailto:Afro.S...@gmail.com>>
>> wrote:
>>
>>
>> What I didn't like was a comment of Resig reffer to web.py as one
>> programmer using his own framework. That is a minimizing action for
>> "great framework for those who hates framework"
>>
>>
>> angelog...@gmail.com
>> http://angelogladding.com
>> (626) 755 - 1417
>>

Maybe it is just me, but I too often thought of working on the webpy
docs a little bit. When I look something up online and see it is not
there or explained in a way that is hard to understand for inexperienced
people, I often thought "alright lemme fix that small part". This is
where I have to be honest; I really just can not be arsed to log in. I
know it sounds stupid, but this feeling of "let's fix that" always comes
in a whim and whenever I have to think hard to remember my password I
already feel like it's been too much trouble to begin with. I wonder if
maybe it is possible to allow anonymous edits to the documentation,
maybe with a captcha?

Really, I know it is conceited and stupid and that I should just take
the time to remember my password once and log in. If it were just me, I
wouldn't even mention it. Though, I can not help but think that maybe
other people feel the same way. Many times have I seen a wiki project I
wanted to work on a little bit, but subsequently dropped the thought
because it required you to register for it. Am I the only one?

Alright, that's my arrogant 2 cents, I hope it does not sound too
annoying :)

Greetings,
b^4

Adam Atlas

unread,
Sep 5, 2007, 8:53:55 AM9/5/07
to we...@googlegroups.com
Maybe we should make some screencasts. Those seem to be popular with
these web frameworks.

Anand

unread,
Sep 5, 2007, 9:20:08 AM9/5/07
to we...@googlegroups.com

On 05-Sep-07, at 6:23 PM, Adam Atlas wrote:

>
> Maybe we should make some screencasts. Those seem to be popular with
> these web frameworks.

I think screencasts are good for newcomers. Web.py is so simple that
there is no need for screencasts.
It is almost impossible to write hello world in a single file in any
other web framework, so they need screencasts.

I think making good documentation with nice examples will be good
enough.

Berwyn Hoyt

unread,
Sep 7, 2007, 1:24:05 AM9/7/07
to we...@googlegroups.com
Ok, I hear you. The deafening silence indicates that you folks aren't particularly interested in supporting a project to make docs for web.py. I guess I won't set one up then.


--------------------

As Angelo and Aaron both mentioned (see below), web.py could do with better docs. But I just don't see it happening without some fan support.

I'm thinking of creating a microPledge project to raise money to create the docs like they raised $17,000 for Ruby on Rails by merely accepting donations (though I don't imagine that we'll make it that high because Ruby has a much bigger following). I know that using a new microPledge feature you can now "pledge now, and pay later" to get an indication of the level support "out there", but I'd like to open this for discussion here before I even start the project.

As for who would do the documentation, we at microPledge would be willing to do it as a job, and you can assess the quality of our docs. But we'd be happy if others quoted the job, too. After all, the object is to make the docs, not the money.


What do you think?

- Berwyn

gregp...@gmail.com

unread,
Sep 7, 2007, 11:21:27 AM9/7/07
to web.py
On Sep 7, 1:24 am, Berwyn Hoyt <berh...@gmail.com> wrote:
> Ok, I hear you. The deafening silence indicates that you folks aren't particularly interested in supporting a project to make docs for web.py. I guess I won't set one up then.

Maybe current users of web.py are content just to look at the code for
documentation? I remember when I started using web.py I gave up on
the documentation pretty quickly and just started reading the code
instead. Obviosly to bring web.py beyond early adaptors, it would
need better documentation.

Features like authentication, openid, and a few other things seem like
the priority for web.py to me. Maybe we could pledge towards that?
Whatever happened with those summer of code projects BTW?

-Greg


Angelo Gladding

unread,
Sep 7, 2007, 6:08:34 PM9/7/07
to we...@googlegroups.com
I'm thinking of creating a microPledge project to raise money to create the docs like they raised $17,000 for Ruby on Rails by merely accepting donations (though I don't imagine that we'll make it that high because Ruby has a much bigger following).  I know that using a new microPledge feature you can now "pledge now, and pay later" to get an indication of the level support "out there", but I'd like to open this for discussion here before I even start the project.

As for who would do the documentation, we at microPledge would be willing to do it as a job, and you can assess the quality of our docs.  But we'd be happy if others quoted the job, too.  After all, the object is to make the docs, not the money.

Berwyn,

I'm glad to hear some more enthusiasm on the topic. I wonder, though, if web.py is ready for that level of commitment. A little bit of digging shows that there were several books written for RoR by the time the pledgie fundraising occurred. Because I can't backtrack at all at rubyonrails.com, I can only assume that at the point these books were written, rails was at least at 1.0 and production worthy out of the box. It is my strong opinion that with Django, Turbogears, etc. out there we're going to need more widespread support to justify a financial input at this time.

I truly believe the organic approach is best for now. In other words, we need a reliable wiki and related discussion. Optimally they would coexist on the same website even though we already have conversations on this list. Aaron has mentioned this from the beginning but we've yet to realize any such solution. That said, and part of the reason I've withheld responding thus far, I'm going to begin a hosted solution to this problem. I'll probably post back by Sunday with my progress. I may or may not make use of Seddit's code considering it uses Prototype and Postgre and I've yet to see it in action.

** Sorry for postponing this long enough for you to become disheartened. That's the direct effect of a low activity project. Don't lose faith.
    • What is the average user of web.py like? Does she enjoy reading source code? Can he program?
    • How does one measure a great documentation? What goals it should promote?
    • Is less more in this particular case?
      Tommi,

      I don't think less is ever more when it comes to good documentation. When coding, absolutely. When looking for help, not so much. Great documentation = python.org. I believe the current one-pager documentation is actually quite useful but it could always grow and expand and that's where it's lacking -- depth. Where do we outline design idioms, the concept of REST, an HTTP roundup (headers, seeother, redirect) etc.

      This is where I have to be honest; I really just can not be arsed to log in. I
      know it sounds stupid, but this feeling of "let's fix that" always comes
      in a whim and whenever I have to think hard to remember my password I
      already feel like it's been too much trouble to begin with. I wonder if
      maybe it is possible to allow anonymous edits to the documentation,
      maybe with a captcha?

      b^4 or should I say b*4 :),

      What you represent is the reality of micro contributions to an open source (free) project. The lower we can set the barrier to entry and the better we can make small contributions stick to each other, the stronger the overall documentation will be. I'm going to be using OpenID for the community pages I'll be building. Anonymous is just being too lazy.

      Maybe we should make some screencasts. Those seem to be popular with
      these web frameworks.

      Adam,

      But how would we make an anti-screencast for our anti-framework? I've never been too successful at making one of these. All the pretty ones seem to come from Mac software.

      I think screencasts are good for newcomers. Web.py is so simple that
      there is no need for screencasts.
      It is almost impossible to write hello world in a single file in any
      other web framework, so they need screencasts.

      I think making good documentation with nice examples will be good
      enough.

      Anand,

      Screencasts are good for marketing and definitely good for newcomers who want a quick drilldown of the standard featureset. Good documentation should easily be 99% of the focus.

      Aaron,

      Will Jottit be using web.py?

      Until Sunday,
      Angelo

      --
      Angelo Gladding

      Berwyn Hoyt

      unread,
      Sep 7, 2007, 7:52:30 PM9/7/07
      to we...@googlegroups.com
      Hi Angelo.  You've probably hit the nail on the head about organic growth right now.  I'm looking forward to seeing your collab solution.  I guess you've decided google offerings are not appropriate.

      Greg, as for pledging to authentication and openid projects, the thing to do would be to enter all the projects and make a web.py tag so that the /tags/web.py page will pull them all up.  If you want these projects, go ahead and create them.  I could make a master project that serves as a focal point and links all of them together if desired (though maybe the web.py wiki is better for that.

      BTW: is there an official verdict? When making tags and names for web.py, should we make the tag web.py or webpy without a dot like in this group name?

      Drew Newberry

      unread,
      Sep 7, 2007, 8:14:28 PM9/7/07
      to web.py
      On Sep 7, 3:08 pm, "Angelo Gladding" <angelogladd...@gmail.com>
      wrote:

      > I truly believe the organic approach is best for now. In other words, we
      > need a reliable wiki and related discussion. Optimally they would coexist on
      > the same website even though we already have conversations on this list.
      > Aaron has mentioned this from the beginning but we've yet to realize any
      > such solution. That said, and part of the reason I've withheld responding
      > thus far, I'm going to begin a hosted solution to this problem. I'll
      > probably post back by Sunday with my progress. I may or may not make use of
      > Seddit's code considering it uses Prototype and Postgre and I've yet to see
      > it in action.

      I'm not sure if Seddit is really the tool to solve the "documenting
      process." When I set out to create it, I had the idea of something
      more like a mailing list. You'll search the archives if you have a
      question, but usually the first place to find your answer will be in
      the documentation. And for documentation, a wiki is certainly the best
      tool for the job. I believe seddit will work well for your "related
      discussion" aspect of the wiki.

      I'm interested to see how your hosted solution is coming. I'd be
      interested to talk with you, and see if maybe we can't get seddit
      integrated when it's ready to go live. I think it can solve atleast
      half of the problem.

      As for seddit's progress. I don't plan to drop the ball on this one.
      I'll continue to work on it throughout the school year, and get it
      deployed. The summer just wasn't enough time to get specs written up,
      and write a fully function application.

      Also, just curious. Is there anything wrong with prototype and
      postgresql?

      > I don't think less is ever more when it comes to good documentation. When
      > coding, absolutely. When looking for help, not so much. Great documentation
      > = python.org. I believe the current one-pager documentation is actually
      > quite useful but it could always grow and expand and that's where it's
      > lacking -- depth. Where do we outline design idioms, the concept of REST, an
      > HTTP roundup (headers, seeother, redirect) etc.

      I agree. I came into writing seddit without ever using web.py. One of
      the most daunting pieces of writing code for web.py is that I had no
      clue where to start. I think some documentation with "best practices"
      would be very helpful. I know one of the neat things about web.py is
      the freedom you have to develop your own application, but it's still
      very helpful to give newcomer's somewhere to start. Let them get their
      feet wet before letting them loose.

      Drew.

      Tzury

      unread,
      Sep 8, 2007, 12:39:47 PM9/8/07
      to web.py
      Guys,
      Why not use infogami as the wiki platform for documentation? Am I
      missing something?

      Adam Atlas

      unread,
      Sep 8, 2007, 1:19:27 PM9/8/07
      to we...@googlegroups.com
      Well... infogami.com is pretty much abandoned (Aaron sold it to the
      same company that bought Reddit, I think, and they haven't been
      developing it any further). Which is a problem because of the
      relatively meager feature set (and the general lack of support).

      We used to be using webpy.infogami.com, but we moved the wiki to
      webpy.org (merging the small amount of content on the old Trac wiki
      into it) for those reasons. Webpy.org is now powered by "the new
      infogami", which I'm guessing is the third wikiish web app named
      "Infogami" written by Aaron. (Infogami.com was the second that I know
      of, and zpedia.org, another site by Aaron, was "powered by" an
      "Infogami" that appeared to be a still earlier iteration.) So if
      we're going by the brand name alone, we *are* using Infogami. And
      presumably a better one than the one running on infogami.com.

      Anand

      unread,
      Sep 8, 2007, 2:03:02 PM9/8/07
      to we...@googlegroups.com
      > We used to be using webpy.infogami.com, but we moved the wiki to
      > webpy.org (merging the small amount of content on the old Trac wiki
      > into it) for those reasons. Webpy.org is now powered by "the new
      > infogami", which I'm guessing is the third wikiish web app named
      > "Infogami" written by Aaron. (Infogami.com was the second that I know
      > of, and zpedia.org, another site by Aaron, was "powered by" an
      > "Infogami" that appeared to be a still earlier iteration.) So if
      > we're going by the brand name alone, we *are* using Infogami. And
      > presumably a better one than the one running on infogami.com.

      For your information, new Infogami powers the Open Library (http://
      demo.openlibrary.org) and I am working on infogami development.

      Tzury

      unread,
      Sep 8, 2007, 2:09:34 PM9/8/07
      to web.py
      I actually meant the *new infogami*, you know, the one that is used in
      the openlibrary.org
      Anyway, once there is an up and running application on the air, I
      think it would be easier for community developers who want to
      contribute for the documentation.

      Angelo Gladding

      unread,
      Sep 9, 2007, 6:30:06 PM9/9/07
      to we...@googlegroups.com
      Hi Angelo. You've probably hit the nail on the head about organic growth right now. I'm looking forward to seeing your collab solution. I guess you've decided google offerings are not appropriate.

      We'll see if I can convert words to action in a meaningful enough way. I look forward to finishing something useful. Google's products are great. I'm so satisfied to see a GGroup and a GCode for an OSS project because I know it's bound to just work. However, when you want to step it up a notch it only makes sense to branch out to a custom solution. Web service interoperability just isn't there yet. That's a conversation for another day.

      Greg, as for pledging to authentication and openid projects, the thing to do would be to enter all the projects and make a web.py tag so that the /tags/web.py page will pull them all up. If you want these projects, go ahead and create them. I could make a master project that serves as a focal point and links all of them together if desired (though maybe the web.py wiki is better for that.

      Yes, yes, and yes. This seems to be a much better appropriation of funds than for documentation. An integration of OpenID with this summer's session project would make the barrier to entry of social apps a cake walk.

      BTW: is there an official verdict? When making tags and names for web.py, should we make the tag web.py or webpy without a dot like in this group name?

      I've thought about this one myself as well. `Webpy` and `webpy` just don't look right. In light of Pythonic naming convention, my vote is to maintain the period.


      I'm not sure if Seddit is really the tool to solve the "documenting
      process." When I set out to create it, I had the idea of something
      more like a mailing list. You'll search the archives if you have a
      question, but usually the first place to find your answer will be in
      the documentation. And for documentation, a wiki is certainly the best
      tool for the job. I believe seddit will work well for your "related
      discussion" aspect of the wiki.

      I'm interested to see how your hosted solution is coming. I'd be
      interested to talk with you, and see if maybe we can't get seddit
      integrated when it's ready to go live. I think it can solve atleast
      half of the problem.

      You seddit :). I'm already breaking DRY by rewriting a wiki system. My intentions were to discuss possible integrations with Seddit (either by me or by another volunteer) once I had a skeleton up and running. Who better than the original author? We can resume a private conversation for the interim if it'll provide more rapid output but we really should keep to the lists for external input and moderation. Thanks for chiming in though, seriously.

      As for seddit's progress. I don't plan to drop the ball on this one.
      I'll continue to work on it throughout the school year, and get it
      deployed. The summer just wasn't enough time to get specs written up,
      and write a fully function application.

      I saw all of your pre-dev docs, wireframes, etc. Just curious, was this out of obligation or necessity?

      Also, just curious. Is there anything wrong with prototype and postgresql?

      I used Prototype prior to jQuery. jQuery simply quadrupled my AJAX/enlivenment output. As for postgre, I've simply never used it and thus don't have it installed/configured. Do you have PG specific schema in your app? We can decide a DB upon integration at a later time.

      As for Infogami, I really don't know what to say. I used to use the hosted version. It was nice. The new dynamic version just isn't very smooth and shows no signs of eminent change. As for the *new infogami* a.k.a. OpenLibrary, Tzury would you care to explain? From their tech preview, I see some neat things going on but a) how is it the new infogami, b) how is it portable to other applications, and c) what are some of the new features that would be of interest to our needs?

      @ Greg: what is utilitymill's purpose and intent? OSS? I've got a ton of ideas if you're interested -- especially if you're willing to donate time/code to this project.

      So here's what I'm currently working on and towards:
      • OpenID consumption -- anyone who doesn't already have one should get one.. if you own a domain name and are not using it as a delegate, you may want to look into it (my OpenID: angelogladding.com; my provider: myopenid.com)
      • Wiki pages -- for documentation and user namespaces -- content types decide presentation ( i.e. Markdown, Python, etc.)
        • documentation may contain historic snapshots of code
        • user namespaces may contain profile information and sandboxes for code
      • Discussions -- merge IRC & Gmail, like forums, with AJAX (Seddit)
        • spawned just like a post to this list
        • additionally every wiki page maintains it's own thread (for annotations and discussion for improvement)
      • Track -- further down the road and only if supported by the community
        • integrated code versioning (I'm thinking take SVN down a decentralized path a la GIT, Bazaar, etc. if sandboxes are utilized)
      I just realized something as I refer back to Aaron's `teh communicator` spec over on the ideas page. The four items under "web apps" could be made available in one fell swoop if we can pull this all together. Furthermore, the pyWeek-esque contests would be trivial to coordinate in a more cohesive community environment like this ( e.g. target fellow developers based on skills according to profile page in a discussion forum, start a project wiki page, push code to personal sandboxes, finalize by merging with main trunk or experimental trunk). Of course why stop there? It would be just as easy to say that the Tools & Infrastructure upgrades could be managed as well. I suppose the moral of the story, IMO, is that a central hub of communication is a vastly important element in expanding a project with such potential diversity. GGroups just isn't cutting it.

      So go grab your OpenIDs, delegate 'em, and start brainstorming a (re)organizing of docs, discussion, and existing code projects. As promised, a progress update will come tonight.

      Angelo

      On 9/8/07, Tzury <Afro.S...@gmail.com> wrote:

      I actually meant the *new infogami*, you know, the one that is used in
      the openlibrary.org
      Anyway, once there is an up and running application on the air,  I
      think it would be easier for community developers who want to
      contribute for the documentation.


      On Sep 8, 8:19 pm, Adam Atlas < a...@atlas.st> wrote:
      > Well... infogami.com is pretty much abandoned (Aaron sold it to the
      > same company that bought Reddit, I think, and they haven't been
      > developing it any further). Which is a problem because of the
      > relatively meager feature set (and the general lack of support).
      >
      > We used to be using webpy.infogami.com , but we moved the wiki to

      > webpy.org (merging the small amount of content on the old Trac wiki
      > into it) for those reasons. Webpy.org is now powered by "the new
      > infogami", which I'm guessing is the third wikiish web app named
      > "Infogami" written by Aaron. ( Infogami.com was the second that I know

      Angelo Gladding

      unread,
      Sep 9, 2007, 8:05:26 PM9/9/07
      to we...@googlegroups.com
      Am I doing something wrong here? In browsing this thread from Google Groups' web interface Anand interjected between Adam and Tzury yet it never hit the thread in Gmail. It was a small but rather important comment as well. Here's the quote if anyone else has the same problem.

      For your information, new Infogami powers the Open Library ( http://demo.openlibrary.org) and I am working on infogami development.
      ~ Anand

      @ Anand: What is your position on web.py community pages? Do either you or Aaron have any intention of bringing that infogami to web.py? I see now that it's being used for code management. Please advise.

      Angelo

      Anand

      unread,
      Sep 10, 2007, 7:12:34 AM9/10/07
      to we...@googlegroups.com
      On 10-Sep-07, at 5:35 AM, Angelo Gladding wrote:

      Am I doing something wrong here? In browsing this thread from Google Groups' web interface Anand interjected between Adam and Tzury yet it never hit the thread in Gmail. It was a small but rather important comment as well. Here's the quote if anyone else has the same problem.

      For your information, new Infogami powers the Open Library ( http://demo.openlibrary.org) and I am working on infogami development.
      ~ Anand

      @ Anand: What is your position on web.py community pages?

      I think we really need to have good documentation. Probably a web.py book on the wiki. Also the site deserves better organization and design.

      Do either you or Aaron have any intention of bringing that infogami to web.py?

      What do you mean? http://webpy.org runs on infogami.

      I see now that it's being used for code management. Please advise.

      I didn't understand what you meant by code management.

      Aaron Swartz

      unread,
      Sep 10, 2007, 1:44:20 PM9/10/07
      to we...@googlegroups.com
      > What do you mean? http://webpy.org runs on infogami.

      Yes, openlibrary.org and webpy.org are running the same version of infogami.

      Drew Newberry

      unread,
      Sep 11, 2007, 10:18:58 AM9/11/07
      to web.py
      On Sep 9, 3:30 pm, "Angelo Gladding" <angelogladd...@gmail.com> wrote:
      ...

      > >I'm not sure if Seddit is really the tool to solve the "documenting
      > > process." When I set out to create it, I had the idea of something
      > > more like a mailing list. You'll search the archives if you have a
      > > question, but usually the first place to find your answer will be in
      > > the documentation. And for documentation, a wiki is certainly the best
      > > tool for the job. I believe seddit will work well for your "related
      > > discussion" aspect of the wiki.
      >
      > > I'm interested to see how your hosted solution is coming. I'd be
      > > interested to talk with you, and see if maybe we can't get seddit
      > > integrated when it's ready to go live. I think it can solve atleast
      > > half of the problem.
      >
      > You seddit :). I'm already breaking DRY by rewriting a wiki system. My
      > intentions were to discuss possible integrations with Seddit (either by me
      > or by another volunteer) once I had a skeleton up and running. Who better
      > than the original author? We can resume a private conversation for the
      > interim if it'll provide more rapid output but we really should keep to the
      > lists for external input and moderation. Thanks for chiming in though,
      > seriously.
      >
      > As for seddit's progress. I don't plan to drop the ball on this one.
      >
      > > I'll continue to work on it throughout the school year, and get it
      > > deployed. The summer just wasn't enough time to get specs written up,
      > > and write a fully function application.
      >
      > I saw all of your pre-dev docs, wireframes, etc. Just curious, was this out
      > of obligation or necessity?

      This was out of necessity. I wasn't exactly sure how I wanted to
      approach this application, and by gettting some wireframes up, and
      writing everything down I was able to get a better idea of how the
      application was going to be implemented. I was also hoping to get some
      community feedback, but never got around to getting that stuff in a
      finalized state, and posting it up. ;)

      >
      > Also, just curious. Is there anything wrong with prototype and postgresql?
      >
      >
      >
      > I used Prototype prior to jQuery. jQuery simply quadrupled my
      > AJAX/enlivenment output. As for postgre, I've simply never used it and thus
      > don't have it installed/configured. Do you have PG specific schema in your
      > app? We can decide a DB upon integration at a later time.
      >

      Nothing is specific to postgresql that couldn't be rewritten in mysql.
      It would be trivial to add support for mysql. For now I'm thinking the
      best way for me to proceed is to get seddit into a working state and
      then report back with something usable for you guys to toy around
      with. Maybe adding some kinda of RESTful api to this thing would be
      helpful in integration? If anyone has thoughts on that let me know.
      One of the more useful things that seddit would be able to provide is
      information on individual users. Seddit will *eventually* keep track
      of what topics each user is knowledgeable in. I'm sure having
      information like that available would help when targeting users.

      Drew Newberry

      unread,
      Sep 11, 2007, 8:00:01 PM9/11/07
      to web.py
      I found an interesting blog post discussing much of the same issues we
      are discussing here, although more rails centric.

      http://nubyonrails.com/articles/is-free-beer-possible

      It's worth a read.

      Werner Hartnagel

      unread,
      Sep 14, 2007, 2:38:29 PM9/14/07
      to web.py
      Regarding the docs, a good start would be more Docstring inside the
      Code ...that would allow to create a up-to-date auto generated manual
      or explore with pydoc. I like to help with the Docs but i'm not a
      native english speaker and i'm still a novice with webpy. I think
      about writing a few tutorials on my blog in German.

      Reply all
      Reply to author
      Forward
      0 new messages