How do we petition Google to mention web2py?

[weheh]

Google's page about App Engine http://code.google.com/appengine/docs/whatisgoogleappengine.html
states the following:

"""
...
The Python Runtime Environment
With App Engine's Python runtime environment, you can implement your
app using the Python programming language, and run it on an optimized
Python interpreter. App Engine includes rich APIs and tools for Python
web application development, including a feature rich data modeling
API, an easy-to-use web application framework, and tools for managing
and accessing your app's data. You can also take advantage of a wide
variety of mature libraries and frameworks for Python web application
development, such as Django.

The Python runtime environment uses Python version 2.5.2. Additional
support for Python ...
"""

Anybody know who has to be petitioned at Google to get a link to
web2py on this page?

[mdipierro]

I do not know but I guess you can email Guido van Rossum. He works
there.

Massimo

On Jul 5, 11:58 am, weheh <richard_gor...@verizon.net> wrote:
> Google's page about App Enginehttp://code.google.com/appengine/docs/whatisgoogleappengine.html

[weheh]

What's his email? Send it to me privately if you have it. I'm sure he
gets tons of email, but I'll be happy to compose a brief letter.

[Yarko Tymciurak]

he's fully aware of web2py...  he personally likes using django, and heads up (or at least is a lead player) on gae...

I think it would serve web2py better to not ask for recognition, but rather show useful applications on gae, and show increasing support for bigtables...   for example, reitveld (code-review) is django on gae, used throughout google, has been ported to private servers, is available for projects on code.google.com, etc.

Something this significant (in terms of usefulness, complete design, showing broad usability) would do _much_ more for getting recognition for web2py...

an email might even work counter to your intentions... do good work instead of emails!

[Richard]

you could try contacting Guido through his blog: http://neopythonic.blogspot.com/

[weheh]

Agreed with Yarko that until and unless web2py is as effective with
Google Apps as Django, then I would hold off any petitioning. However,
if it is as good, then I don't understand why Google would necessarily
show such a clear bias towards Django on the Google Apps info page. In
addition, assuming there are web2py Google apps to point at (I intend
to develop one) then in that case a petition would be in order,
especially if it came from the community of which I am a part.

[Yarko Tymciurak]

I thought I explained that perfectly well - it may be as good, but it has not the length and bredth of track record.

We can do things; we can do things quickly; we can show how you would do things;  sometimes we find that
things you would want to do just a little different aren't so easy to do, because we hadn't run into, thought of. or otherwise encourntered that.  Sometimes people want to do things different "just because they want to test if they can", but sometimes people want to do things moderately differently because they are sitting with a client, want to tweak the prototype for the client to match what he is asking for (and may or may not later suggest a better / different / more efficient approach).

If stage 0 is an idea;
stage 1 an implementation that works for me;
stage 2 an implementation I can show others how to use
stage 3 an implementation that can adapt to the needs of the real world under commercial demands
stage 4 is wide use across more than one industry
stage 5 is evolution and standardization, and adoption by major institutions
stage 6 is a standard...

Then django (regardless of technical / age / etc. arguments) is firm stage 4, having entered stage 5 (by virtue of GAE);

Most other frameworks (and not just python ones, and including ones with large backing by corporations) are lucky if they make it into stage 3, and rarely make it to stage 4 (things change too much in the internet - and we are embarking on yet another wave of major changes, mobile and rich clients being the _least_ of them!)

django is unusual in this regard, and - even as you can argue that things can be made better, you cannot deny this is the state of django - quite an accomplishment, and a testament to the power of media (e.g. news industry), and the effects of surviving the demands of that industry.

web2py, in my humble and simplistic painting of the scenario here, is firmly in stage two, poised at the brink of "entering" stage 3 (which it has not done yet in a long-term, varied, consistent way, without saying anything about if it can or can't... what the forces are propelling it, what are the forces resisting it at this moment).

What I said was "go yonder into the stage 3 world, and bring lots of feedback into these here forums".
When that has happened over the next year(s), then I'll talk about how we're poised to enter stage 4.

In the meantime, the quickest way to get noticed by the likes of google, guido is to continue to do useful things, continue to be part of conferences, and perhaps push some of the early stage 3 efforts into wide and heavy use _on_ gae.  This will get early attention.

All else is, in effect, premature.

Happy designing!

- Yarko

[weheh]

Well put, Yarko.

[Yarko Tymciurak]

:-) 

I'll also say that the people who participated in the "fire of the news industry" were skilled, survived it, and manage to show that industry they could do useful things... were / are the driving factor behind django being where it is.  Python benefited.  We all did.

I applaud them! 

On Sun, Jul 5, 2009 at 9:06 PM, weheh <richard...@verizon.net> wrote:

Well put, Yarko.

[Yarko Tymciurak]

(you can tell I appreciate Massimo - just look at the number of posts on web2py ;-))

[weheh]

I never doubted it.

[JohnMc]

A round about way with the Google is 'summer of code'. If there is one
or several students making pitches for project acceptance that uses
web2py for the project, then some form of recognition would be
automatic.

JohnMc

On Jul 5, 10:26 pm, weheh <richard_gor...@verizon.net> wrote:
> I never doubted it.

[Hans Donner]

And don't forget, web2py was already mentioned in a google appengine blogpost

http://googleappengine.blogspot.com/2009/05/web2py-support-new-datastore-backend.html

That's how I ended up here...

[Yarko Tymciurak]

yes - that was a nice, prominent mention....

[mdipierro]

I think this is a chicken and egg story. Before we compare with
Django, let's compare with Turbogears. According to google trends we
are as popular as they are (we going up and they going down). Same is
true for Cherrypy. Yet TG was allotted 8 hours of tutorials at Pycon
2009. Cherrypy had one talk. The web2py talks were rejected in 2008
and 2009. This is despite the fact pycon used web2py for their
registration software.

We all know what happened. A member of the SQLAlchemy community voted
against the web2py talks based on the argument that I would be doing
advertising and not a technical talk (pretending to forget I am a
prof. and not a salesman).

Why is that? I think there are multiple answers.

One is that the python community still perceives web2py as a one man
project. You people need to brag more about it and about your
contribution.

Another reason is that there are commercial interests behind web
frameworks and the bigger they are the more money have been invested
into them. Google has clearly made an investment into Django before
they ever heard of web2py. Remember that it is never about the best
technology, it always about getting the most users. Right now Django
has the most users and most people will back it up just for that
reason.

I think it is important to show Google and Guido that there is a
community behind web2py. That can be accomplished with email exchanges
but it should also be accomplished (and more effectively) by showing
off applications.

Web2py is growing exponentially (whatever system you use to measure
it: email, site visitors, group members), slowly but exponentially. It
will grow faster if Google were to acknowledge us but is not going to
happen until the community makes its voice heard.

As I said, it is a chicken and egg story.

Massimo

On Jul 6, 12:19 am, Yarko Tymciurak <yark...@gmail.com> wrote:
> yes - that was a nice, prominent mention....
>

> On Sun, Jul 5, 2009 at 11:32 PM, Hans Donner <hans.don...@pobox.com> wrote:
>
> > And don't forget, web2py was already mentioned in a google appengine
> > blogpost
>

> >http://googleappengine.blogspot.com/2009/05/web2py-support-new-datast...

>
> > That's how I ended up here...
>

[Hans Donner]

Hi Massimo,

this is perhaps also fuelled by the fact that all source/distributions
are coming from your trunk. Don't get me wrong, I think ity's very
good that yo oversee the project and keep it from straying off track.
And yes, everybody can send you patches and you says thanks in your
commits when you apply them.

We need more visibility on this, eg like Django on
http://code.djangoproject.com/#Gettinginvolved, but also the
ticket/feature requests discussed earlier. This will create better
visibility that more people are(already) contributing.

H

[Yarko Tymciurak]

I think you are starting to mix two different things here - django has been in commercial use for some years.  Runs info for a news organization.

Your TG story is another thing, but the _fact_ remains (and I've had this private conversation quite recently with you) that solid, well designed, non-trivial, and commercially in-use applications written in web2py are.... what?  not advertised?  non-existent?

Even the _church_ I want to build a free site for asks for "what other sites are heavy users of this?"  ---- THAT is a chicken & egg problem.

I think focusing on being "loud" will maybe anger, maybe make some Python community people make room.
But more importantly, it will waste time and resources, and not really budge Google/Guido.

It will do _nothing at all_ for my church situation, much less many, many other such situations for consulting and getting some smaller sites up and running.  Effort is much better spent making something substantial, and showing results rather than tutorials to the rest of the world.

Sure, some communication is important.  But that's not primarily what is needed now.
I welcome, of course,  other views and experiences - but this (my) experience and observation is not opinion, it is what I _have experienced_.  The only thing that is opinion is how widespread and significant this is.

What is needed are good designs, and showing  useful things, and getting them _used_, collecting statistics.

Forget about email campaigns for "airtime" to teach web2py - at least for now.  That needs to come by demand (and is).

What needs to happen is results, and statistics.

Not email.

How many people are using pyforums, for example?  In what applications?  What are limitations found?  What is the total user base?  Where does someone find that information.   And how about the survey tool?  Who is using that (the code was lost, so that is a rhetorical question - to make a point - it only exists in executable form as a demo on GAE, so it is ... a demo;  how is that organization using it that the original was donated to?)   Or reddish?   Or....

You see - we show ideas.   We need to transition into making things _others (non-developers)_  find useful, and make use of them.  This is what is missing.


You want certification?   Shelve the notion of developer certification (it is of little immediate use, without _commercial demand for developers_  existing first).   Instead, lets talk about certification of _applications_ and _modules_ for web2py.

A quick grading of the sites listed as running / supporting web2py on www.web2py.com, and I am left without anything I can even proudly show a non-technical church manager for _free_ work.  This borders on shameful.

"I use web2py; call me for web2py consulting"  sites, with one or two pages that are not showing anything but simple content --- doesn't show off anything, is less than sophmorish. 

Please get these OFF of the listings on the back pages of web2py.com!

Better to show some significant student projects.  

Setup tests so that the appliance that run are current, or are shown (at least) which version they are valid and tested up to.  

These are the things that need focus on.

web2py did registration for pycon2009.  It was asked to do that again for pycon2010.
It wasn't "great" (UI's, etc. were rushed; the app was written in a month, people hadn't considered all the requirements ... e.g. formal reciepts, etc.) - but it was reliable, and worked well enough (with a little support for users - as is normal for new software that sees the light of "real users" - and THAT is the whole point: how easy it was to adapt in the presence of sudden, large scale, first time use - it was a _great_ experience for us, who supported and extended tables on the live system...).

Things that work.
Things that people want to use.
Things that people are _willing_ to use, _do_ use, and find _useful_.

Lack of those is what results in what has been referred to as this "looking like a one-man show..." --- NOT lack of people speaking out (to think speaking out is what is missing would miss a critical point).  

Mind,  having people talk about their experiences w/ web2py (e.g. Brazil conference) is valid speaking out, but it is based on ... you can guess what I'll say ..... EXPERIENCE WITH MAKING SOMETHING INTERESTING WORK USING WEB2PY....

Focus on that.

Identify _real_ things that would make your life easier in accomplishing that with web2py (that will drive continuous improvement of web2py).

Then take your results, and present about them, email about them, push for _them_ to get into conferences.

To your church.
To your local python group.
To you community.
Then get broader.

We here are all _interested_ in web2py, but why?  Because it's easy to build things.

That's right - we can make things with it.

Make more and more significant things.

Emails to google or pycon.... Pshah!  waste of talent and energy in my opinion.

Who has built that wiki for documentation?  Not significant enough?  Ok - then lets use MoinMoin w/ rst default.
We need effort on testing, in particular setting up testing of appliances.
Is anyone up for updating the appliances app?
Probably what we need is a project list where people sign up, and people join in - and where we can visibly critique, and try to make projects more significant, or ensure that the most significant projects get most attention.  Good graduate projects, and more.

- Yarko

[Yarko Tymciurak]

I agree - a mainline release process would be of benefit.

I think we need to look at the maturity of that google code mercurial option, as that would give us some important tools - like "code-review" per submission, so it would at least be a process...

and could lead us into what Hans mentions...

[mdipierro]

Yarko. Many more people than you know are using web2py.

I know of companies who do and I know of government organizations who
do, although I am not saying their names without their permission.

For various reason these users tend to use web2py do develop
applications for internal use and not public web sites. Our users are
naturally reserved about what they do. We pitch web2py as an
Enterprise framework in fact.

We should not confuse apples and oranges. Some frameworks are popular
because they have been used to develop public blogs, wikies and things
like that. While web2py, for example, has been used to develop a
complex system for a major bank. If you read posts on this list you
find people are using web2py to build systems for medical records and
accounting systems.

One of the most iPhone apps is called iFart. By that analogy
popularity is not a measure of quality nor a measure of success.

It is good that web2py runs on the Google App Engine and I would like
to see more applications but I am happier with one major bank a web2py
app than 100 GAE apps to post pictures of cats.

I would be like Google to acknowledge web2py more on their site. For
this to happen the community has to speak up. One way to do it is,
when asking questions on their list, make clear that we are using
web2py.

Massimo

> A quick grading of the sites listed as running / supporting web2py onwww.web2py.com, and I am left without anything I can even proudly show a

[weheh]

Hey, Massimo, what've you got against cats? ;^)

And Yarko says:
"""Your TG story is another thing, but the _fact_ remains (and I've
had this
private conversation quite recently with you) that solid, well
designed,
non-trivial, and commercially in-use applications written in web2py
are....
what? not advertised? non-existent? """

I just announced my first app, which has been used commercially now
for a few weeks and has made hundreds of paid-for reservations for the
club "SPiN New York". Talk about high profile, SPiN has been mentioned
in Conde Nast, People Magazine, Jimmy Fallon Late at Night, Good
Morning America, etc. Go look at their website. Susan Sarandon is one
of the backers. Not that she's very web saavy, but you can bet she's
conscious of keeping her image clean. Which means, she doesn't want
her website to crash.

My web2py app was a ground-up implementation of a calendar-based
reservations system which is every bit as sophisticated as Google
calendar, with a GUI interface and a comprehensive api. It supports
all the "gnarly" functionality that Google claims as a reason why you
should use Google Cal and not build one yourself. (As a matter of
fact, it was a little gnarly, especially with the repeat events, but
web2py made it a lot easier to build IMO).

Listen. I'm not asking for Google to tell everyone to use web2py. I'm
just asking for Google to acknowledge web2py in the same breath as
Django on their "what is GAE" page. They don't have to endorse it.
Just say it works with GAE. For that matter, they should also mention
other frameworks that work with GAE. As I said before, I think if the
concensus here is that web2py works with GAE as well as Django does,
then pull the trigger and let's start a petition to get some
visibility and mention on the GAE page. I, for one, am 100% behind
building commercial-grade, heavy-lifting, production websites in
web2py. I may not be the world's most sophisticated web developer, but
it sure does the trick for me. As such, I want to see it thrive so
that my approach will survive.

And to do such a petition takes almost no effort at all in the
internet era.

[mdipierro]

weheh. You think you absolutely have a point.

here are some other public apps that use web2py (although as I said
most web2py apps are not designed to be public):

http://www.appliedstacks.com/NewestFirst/web2py

Massimo

[Yarko Tymciurak]

Massimo -

Sounds good.

On Mon, Jul 6, 2009 at 8:51 AM, mdipierro <mdip...@cs.depaul.edu> wrote:

Yarko. Many more people than you know are using web2py.

I know of companies who do and I know of government organizations who
do, although I am not saying their names without their permission.

 While this is understandable for occaisional users, it makes little sense that there is _no one_ to talk about.

For various reason these users tend to use web2py do develop
applications for internal use and not public web sites. Our users are
naturally reserved about what they do. We pitch web2py as an
Enterprise framework in fact.

If I - anyone - tries to sell someone, the first questions that come out are something like this:

• Who else uses this?  How much do they use this?
• Is there enough users that if something happens to you, we will be able to find someone to maintain / update?
• How widely is it used?  How much traffic has this seen?
  

We should not confuse apples and oranges. Some frameworks are popular
because they have been used to develop public blogs, wikies and things
like that. While web2py, for example, has been used to develop a
complex system for a major bank. If you read posts on this list you
find people are using web2py to build systems for medical records and
accounting systems.

You are not hearing me - I am not talking about populrity - I am talking about something else.

E.g. - "Medical system I am not at liberty to disclose;  type of use:  records;  estimated data size:  over 10M records;  estimated user base:  > 1000 system uses;  frequency of use:  daily 24/7"  is a FAR sight different than "people are developing for major medical system with web2py!  Isn't that great?!"

The first I (you - anyone) can take to a client (assuming it is accompanied by _some_ disclosable data, e.g. "City of Sao Paulo; http://www.here_i_am_a_web2py_site.gov"); 

the second is no more than so much talk.

One of the most iPhone apps is called iFart. By that analogy
popularity is not a measure of quality nor a measure of success.

I ask for data, concrete evidence, and you talk to me about popularity...  this (iFart) example is completely irrelevant here.

 

It is good that web2py runs on the Google App Engine and I would like
to see more applications but I am happier with one major bank a web2py
app than 100 GAE apps to post pictures of cats.

Me too.  In fact, look at this page:

http://www.joomla.org/about-joomla.html

Notice "who uses Joomla"  (a php cms that is in my face at at least one organization:  "Yarko look at this:  who uses web2py?  what can I show the board of directors to compare? They are being told they should go with Joomla" ---  Get real: NO ONE is asking about popularity!!!)

Notice the balance:  only ONE  "internal, we can't show you" in a sea of 9 sites.
Notice: while there are no statistics here, the _sense_ is:  "kinds of uses".  Want to see that?  I particularly liked the Harvard site:
http://gsas.harvard.edu/

Click around it; hover over things; "feels" responsize.  It's a good "first impression".

"Yarko - is there a web2py site like this we can look at with the Board of Directors?"

Ummm- I'm trying to find something.... please just be patient with that contract....  Don't go to php just yet...

 

I would be like Google to acknowledge web2py more on their site. For
this to happen the community has to speak up. One way to do it is,
when asking questions on their list, make clear that we are using
web2py.

That (mention what you are using, why you are asking) is at least a sane suggestion.

Doesn't address my issue in the least, but at least it makes some sense.
 

Massimo

Yarko

[Yarko Tymciurak]

Hey weheh -

On Mon, Jul 6, 2009 at 1:33 PM, weheh <richard...@verizon.net> wrote:

Hey, Massimo, what've you got against cats? ;^)

......

I just announced my first app, which has been used commercially now
for a few weeks and has made hundreds of paid-for reservations for the
club "SPiN New York". Talk about high profile, SPiN has been mentioned
in Conde Nast, People Magazine, Jimmy Fallon Late at Night, Good
Morning America, etc.

I remember - I was just with a board member that week who was asking "is there something we can look at"; I clicked over to the site, and it was run in (I think it was) Ruby-on-Rails, and you explained just your piece, just the reservation piece...

Now - I don't diminish your work, or that it is used - but I also don't have something I can give a board of directors and say "sure - look around this system, see if you like it's behavior" --- they might wind up looking for a RoR consultant!

I looked at events:  that doesn't look like web2py;

(it does show a version number that is only displayed in "show source" intentionally - it would be nice to have such a marker in webpy basic layout)

I looked at membership.... doesn't look web2py either;

Go look at their website. Susan Sarandon is one
of the backers. Not that she's very web saavy, but you can bet she's
conscious of keeping her image clean. Which means, she doesn't want
her website to crash.

I looked around, but could not find anything that I could point a board to look around on their own to get to know web2py.

I couldn't use this.
 

My web2py app was a ground-up implementation of a calendar-based
reservations system which is every bit as sophisticated as Google
calendar, with a GUI interface and a comprehensive api. It supports
all the "gnarly" functionality that Google claims as a reason why you
should use Google Cal and not build one yourself. (As a matter of
fact, it was a little gnarly, especially with the repeat events, but
web2py made it a lot easier to build IMO).

Sounds perhaps like material for an article to a Python mag somewhere...  

Listen. I'm not asking for Google to tell everyone to use web2py. I'm
just asking for Google to acknowledge web2py in the same breath as
Django on their "what is GAE" page. They don't have to endorse it.
Just say it works with GAE. For that matter, they should also mention
other frameworks that work with GAE. As I said before, I think if the
concensus here is that web2py works with GAE as well as Django does,
then pull the trigger and let's start a petition to get some
visibility and mention on the GAE page. I, for one, am 100% behind
building commercial-grade, heavy-lifting, production websites in
web2py. I may not be the world's most sophisticated web developer, but
it sure does the trick for me. As such, I want to see it thrive so
that my approach will survive.

Good motivation.

What I said is "I don't think they will - they use code-review publicly and internally, gae+django;  heck, the Android project took reitveld (the code-review project) and rewrote it in java for Android development use.   They don't see anything near that scale, and won't feel comfortable _saying_ that 'you can use web2py with our google stuff' because they want to see what they would be associating with first.

Who can blame them?   Any small board of directors I've talked to wants the same!

PyCon said "ok" to web2py for strength of Massimo's enthusiasm, and support of 2-3 people who were behind it. It was a fight, but the final determinant:  it was late, and they had no other viable choices before them.  It worked out, so they were happy in the end.

Every board I've talked to is as risk averse.

And to do such a petition takes almost no effort at all in the
internet era.

As Massimo suggested: Mention you are working w/ web2py;  stop trying to "control" if Google likes you or will talk about you; worry instead about what you can do (you can control that).

If you are not willing to  see the realities I uncover before you,  then you will net see them and you will not be able to address them.

Start building a portfolio you (we) can show others.  (and as Massimo so correctly points out, that is not "fluff" popularity, etc.... although if you build somthing that takes on as much as twitter or facebook, that probably won't hurt - popularity that brings a large user base is, at least, data!).

- Yarko

[mdipierro]

I understand what you are trying to say but when you say "no one is
using web2py" gives people the wrong impression and not the correct
one. It is not fair to member of this community who use web2py.

You should also assume that there are lots of people on this list who
are considering web2py but have not yet made their mind about it.

As you pointed out Pycon 2009 used web2py to register 800 users and
handled thousands of monetary transactions.
http://www.klasproducts.com/ uses web2py to sell hundreds of products
http://www.spinyc.com/ uses web2py in their intranet
http://www.whitepeaksoftware.com/ uses web2py for their web site
http://diarywiz.com/ is powered by web2py

There are just some of those we know about and choose to advertise
what they do.

Of course more people will come out eventually. As I said most of the
work is done on intranet apps.

There is no question that Django is more popular and Drupal is
probably much more popular than web2py. They have been around longer,
came first and had more time to build a community. Moreover Drupal is
a designed to build CMS therefore apps made with Drupal are naturally
publicly available.

I do not think discussion about how popular we are vs how popular they
are is healthy. Moreover despite what some people think I do not care
much about it. I would like the discussion to focus more on what we do
better (feature wise), what they do better, what can we do to improve
our features, what can we do to promote web2py features.

Some people are not interested in features as are only interested in
learning the system that will give them the highest chance of getting
hired. Some people are interested in features as a way to get the job
done better and quicker. I am working for this second class of users.
I am interested in building a community for these people.

Massimo

On Jul 6, 2:18 pm, Yarko Tymciurak <yark...@gmail.com> wrote:
> Hey weheh -
>

[Jonathan Lundell]

On Jul 5, 2009, at 11:59 PM, mdipierro wrote:

> Why is that? I think there are multiple answers.
>
> One is that the python community still perceives web2py as a one man
> project. You people need to brag more about it and about your
> contribution.

To the extent that the "one-man project" characterization is true (and
in a sense it's undeniable), it may be helpful to make the analogy of
Linux, where there are many, many developers, but Linus is the
gatekeeper for the release tree, and is the final arbiter of every
patch to that tree, as well as the final authority on overall design
and direction questions.

Of course, Linux is a bigger project, and there are a lot of project
leaders for various parts of the tree. But there's a real sense in
which it's a "one-man project", and it manages to be quite popular,
thank you very much.

[JohnMc]

"Some people are not interested in features as are only interested in
learning the system that will give them the highest chance of getting
hired. Some people are interested in features as a way to get the job
done better and quicker. I am working for this second class of users.
I am interested in building a community for these people.

Massimo"

What you are describing is a SWOT (strength, weakness, opportunity,
threats) analysis.

S:
- works with existing DB's with minimal restrictions.
- Provides WWW std outputs (RSS, XML, etc)
- DAL to abstract away DB complexity.
- GAE friendly.
- High level of customization possible.
- Appliances to do plug and play.
- Built in ticketing.
- etc.

W:
- Python powered. (Rankles me too, but Python is usually not the first
language uttered when discussing web platforms even with Django's
popularity.)
- Minimal third party tools.
- Fear of 'death by truck' for the framework.
- Development is outpacing documentation.

O:
- DAL possibly support schemeless DB's (CouchDB or BerkelyDB)?
(future?)
- Ability to run in a Android environment.
- General portability of the framework.


T:
- Not in the top three in their language class as a framework.
- PHP on the low end. Django/Rails in the midrange. Java/Glassfish/
JBoss/PlOne/Zodb on the high end.

Least that is what I would report to my boss were he to ask. And
please I did this as impartial as I could. For example 'minimal third
party tools' would be valid complaint. The counter argument of course
is many tools are already built in and anything one could consider
lacking could be developed or modified from existing tools.

Were I to make one recommendation for a feature -- Make the DAL
capable of inter-operating with CouchDB. Rationale --

* CouchDB is an up and comer in the DB arena. It plays into the
BigTable and S3 space as a substitute for those that need to keep data
inhouse.
* CouchDB has a simple interface and there are tools that permit one
to integrate django to CouchDB. But it is tacked on. A seamless
interface between Web2Py and CouchDB would be a reasonable first for
any framework.
* One would garner more support from the Mozilla Foundation in the way
of recognition and public relations than one might get from Google.

Just one opinion.

JohnMc

On Jul 6, 3:13 pm, mdipierro <mdipie...@cs.depaul.edu> wrote:
> I understand what you are trying to say but when you say "no one is
> using web2py" gives people the wrong impression and not the correct
> one. It is not fair to member of this community who use web2py.
>
> You should also assume that there are lots of people on this list who
> are considering web2py but have not yet made their mind about it.

<snip>

[Tim Michelsen]

mdipierro :

Some people are interested in features as a way to get the job
> done better and quicker. I am working for this second class of users.
> I am interested in building a community for these people.

I like that statement. Keen to see what follows up.
I mean, you add neice little features every week and asure to maintain
backward compatibility.
But I really miss a roadmap with release dates.

mdipierro :

> One is that the python community still perceives web2py as a one man
> project. You people need to brag more about it and about your
> contribution.

True. And I guess this applies even for memebers of this list.

Such posts as
http://groups.google.com/group/web2py/msg/61d0e1356ac1252a
are interesting.
But will not change a lot until something changes in the project set up.

I agre with Hans here:

> We need more visibility on this, eg like Django on
> http://code.djangoproject.com/#Gettinginvolved, but also the
> ticket/feature requests discussed earlier. This will create better
> visibility that more people are(already) contributing.

For my topic of contribution I would suggest you to install pydocweb on
your server:
http://groups.google.com/group/sphinx-dev/browse_thread/thread/b77ff5ed73faf410
http://code.google.com/p/pydocweb/
http://gminick.wordpress.com/

Although written in Django, it offers the great opportunity to get nice
shiny and valid docstrings with doctest more sooner than later.

a good notice: I was asked by my potential users to continue my app
development today.

Best regards,
Timmie

[Yarko Tymciurak]

On Mon, Jul 6, 2009 at 3:13 PM, mdipierro <mdip...@cs.depaul.edu> wrote:

I understand what you are trying to say but when you say "no one is
using web2py" gives people the wrong impression and not the correct
one. It is not fair to member of this community who use web2py.

Massimo - "no one is using..." are your words, not mine - what I _did_ say is there is a gross shortage
of sites you can point to that serve signicifant users (PyCon registration being the one clear exception).

This doesn't need to be fair or unfair - it just data - what is (I hesitate to say "fact" as there may be unknown data).

You should also assume that there are lots of people on this list who
are considering web2py but have not yet made their mind about it.

There are people who consider using it for developing projects. 

Do you deny it would help them, if they are to present to their partners, their clients, their boards, etc. 
if we had more evidence collected and available for them (us) to use as resources?

Do you realy believe this is not useful?  Data is not useful?  Ach!

Serving that community - even if it gets uncomfortable - if fair.
 

As you pointed out Pycon 2009 used web2py to register 800 users and
handled thousands of monetary transactions.

This is significant.  Cooperated w/ django app for site info.
 

http://www.klasproducts.com/ uses web2py to sell hundreds of products

This is important, if small quantity.  It is a nice  site.
 

http://www.spinyc.com/ uses web2py in their intranet

This is significant, but anecdotal (nothing you can show someone for them to experience; perhaps a mockup that is open would be possible?). 
Runs with RorR as main site (and all we can see / point to).

 

http://www.whitepeaksoftware.com/  uses web2py for their web site

For purposes of showing web2py look and feel, this is just a website front done in web2py;
The storefront is something else; the blog is wordpress.  There is nothing of the "benefits" of web2py to show here I could see.

That is not to criticize - no sense re-inventing wheels; use what is available, and build what is new and interesting.

But from the standpoint of being a showcase example for this community to draw on - for people wanting to show sites
that take advantage of the benefits of web2py, I do not feel this belongs on this list.

We need to be more discerning.

 

http://diarywiz.com/ is powered by web2py

This is a real application example.  I'm not sure how it is significant in terms of what it does new (for example, like survey tool did),
and I'm not sure that I would show this as an example to clients, boards, etc....

It seems like an appropriate example for prospective developers.
 

There are just some of those we know about and choose to advertise
what they do.

So that is 2 for "showcase" - one very low usage ("hundreds"), one seasonal but successful (near 1000 users).
 

Of course more people will come out eventually. As I said most of the
work is done on intranet apps.

Let's hear talk about them.  Even if someone can't name the company, or the specifics it is useful to hear kinds of things.
 

There is no question that Django is more popular and Drupal is
probably much more popular than web2py. They have been around longer,
came first and had more time to build a community. Moreover Drupal is
a designed to build CMS therefore apps made with Drupal are naturally
publicly available.

I am pointing out what will be helpful here, for web2py, for this community.
I am stating what I could use now, what I want

Those other things, comparing, are just distractions (inasmuch as they serve as guideposts, examples, lessons of aspects
we can benefit from).
 

I do not think discussion about how popular we are vs how popular they
are is healthy. Moreover despite what some people think I do not care
much about it. I would like the discussion to focus more on what we do
better (feature wise), what they do better, what can we do to improve
our features, what can we do to promote web2py features.

I am puzzled why you keep using the term "popular" - you must have some issue with this.
There is "fad"; there is popular because useful, reasons behind use (e.g. excercise is popular - it helps you stay healthy!  When the evidence smoking was bad for health became available, smoking became much less "popular").

Discussing what we can do better feature wise - and this is my point - I'd like to see balanced with
EVIDENCE / EXAMPLES of that "better" - "Here - look at this site, for example; or look at this example site."

People ask ME for that, and it's a fair request.

Some people are not interested in features as are only interested in
learning the system that will give them the highest chance of getting
hired. Some people are interested in features as a way to get the job
done better and quicker. I am working for this second class of users.
I am interested in building a community for these people.

Your own words:  "...get the job done better, quicker..."

It's what I hear ->  "You say this is better?  Show me."  Show me the jobs.

*sigh*

A little intentional focus on this kind of evidence, on
applied "better, quicker" can only help web2py.  

Thinking about "I think this will be easier" is fun, creative; 
When it works in many hands / many situations, is work, not always so much "fun";
this is what builds confidence. 
This is nothing new.

 For _your_ comfort, I will say no more on this topic - regardles, I think that clear talk about this topic is in the best interests of web2py, the web2py community.

Massimo

Yarko

[Yarko Tymciurak]

On Mon, Jul 6, 2009 at 5:48 PM, Yarko Tymciurak <yar...@gmail.com> wrote:

On Mon, Jul 6, 2009 at 3:13 PM, mdipierro <mdip...@cs.depaul.edu> wrote:

.....

 

http://www.whitepeaksoftware.com/  uses web2py for their web site

For purposes of showing web2py look and feel, this is just a website front done in web2py;
The storefront is something else; the blog is wordpress.  There is nothing of the "benefits" of web2py to show here I could see.

That is not to criticize - no sense re-inventing wheels; use what is available, and build what is new and interesting.

But from the standpoint of being a showcase example for this community to draw on - for people wanting to show sites
that take advantage of the benefits of web2py, I do not feel this belongs on this list.

We need to be more discerning.

I am sure WhitePeaks is a fine development house.  Perhaps they have some examples they would be willing to share with the community?

[Joe Barnhart]

+1

On Jul 6, 3:48 pm, Yarko Tymciurak <yark...@gmail.com> wrote:

> On Mon, Jul 6, 2009 at 3:13 PM, mdipierro <mdipie...@cs.depaul.edu> wrote:
>
> > I understand what you are trying to say but when you say "no one is
> > using web2py" gives people the wrong impression and not the correct
> > one. It is not fair to member of this community who use web2py.
>
> Massimo - "no one is using..." are your words, not mine - what I _did_ say
> is there is a gross shortage
> of sites you can point to that serve signicifant users (PyCon registration
> being the one clear exception).
>
> This doesn't need to be fair or unfair - it just data - what is (I hesitate
> to say "fact" as there may be unknown data).
>
>
>
> > You should also assume that there are lots of people on this list who
> > are considering web2py but have not yet made their mind about it.
>
> There are people who consider using it for developing projects.
>
> Do you deny it would help them, if they are to present to their partners,
> their clients, their boards, etc.
> if we had more evidence collected and available for them (us) to use as
> resources?
>
> Do you realy believe this is not useful?  Data is not useful?  Ach!
>
> Serving that community - even if it gets uncomfortable - if fair.
>
>
>
> > As you pointed out Pycon 2009 used web2py to register 800 users and
> > handled thousands of monetary transactions.
>
> This is significant.  Cooperated w/ django app for site info.
>
>
>

> >http://www.klasproducts.com/uses web2py to sell hundreds of products

>
> This is important, if small quantity.  It is a nice  site.
>
>
>

> >http://www.spinyc.com/uses web2py in their intranet

>
> This is significant, but anecdotal (nothing you can show someone for them to
> experience; perhaps a mockup that is open would be possible?).
> Runs with RorR as main site (and all we can see / point to).
>
>
>
> >http://www.whitepeaksoftware.com/ uses web2py for their web site
>
> For purposes of showing web2py look and feel, this is just a website front
> done in web2py;
> The storefront is something else; the blog is wordpress.  There is nothing
> of the "benefits" of web2py to show here I could see.
>
> That is not to criticize - no sense re-inventing wheels; use what is
> available, and build what is new and interesting.
>
> But from the standpoint of being a showcase example for this community to
> draw on - for people wanting to show sites
> that take advantage of the benefits of web2py, I do not feel this belongs on
> this list.
>
> We need to be more discerning.
>
>
>

> >http://diarywiz.com/is powered by web2py

> Your own words:  "...get *the job done* better, quicker..."

>
> It's what I hear ->  "You say this is better?  Show me."  Show me the jobs.
>
> *sigh*
>
> A little intentional focus on this kind of evidence, on

> *applied* "better, quicker" can only help web2py.

[mdipierro]

Yarko,

we should not transform a strength into a weakness. As I stressed many
times web2py is mostly used to develop intranet apps in corporate
environments (small and medium size businesses).

There are people do not want to say what they are doing with web2py
and do not want to advertise that they are using web2py because they
are not developing open source projects, because their paid by an
employer who does not want them to publicly say what they are doing,
because
they do not know if they will meet their deadline, and for many other
reasons

While the fact that we want to have more information is a problem for
us, this is not a weakness of web2py. Those corporate users are
exactly our target users and they are entitled to keep their privacy.
I do not care if nobody says they are using web2py. I know we get more
tha 1000 visitors/day and more than 100downloads/day.

I see how everybody is eager to increase the users base.

I made a plot of #users/time from the data in the google group. The
data is nicely fit by an exponential. What is remarkable is that there
are no bumps in the data. It is very smooth. This means web2py is not
spreading because of major events (like my reddit posts) but it is
spreading by word of mouth.

So, we'll keep collecting information people want to send us and we
encourage them to do so. But, most importantly, we should all do our
best to talk about the tools we like to use and contribute develop.

Massimo


On Jul 6, 5:48 pm, Yarko Tymciurak <yark...@gmail.com> wrote:

> On Mon, Jul 6, 2009 at 3:13 PM, mdipierro <mdipie...@cs.depaul.edu> wrote:
>
> > I understand what you are trying to say but when you say "no one is
> > using web2py" gives people the wrong impression and not the correct
> > one. It is not fair to member of this community who use web2py.
>
> Massimo - "no one is using..." are your words, not mine - what I _did_ say
> is there is a gross shortage
> of sites you can point to that serve signicifant users (PyCon registration
> being the one clear exception).
>
> This doesn't need to be fair or unfair - it just data - what is (I hesitate
> to say "fact" as there may be unknown data).
>
>
>
> > You should also assume that there are lots of people on this list who
> > are considering web2py but have not yet made their mind about it.
>
> There are people who consider using it for developing projects.
>
> Do you deny it would help them, if they are to present to their partners,
> their clients, their boards, etc.
> if we had more evidence collected and available for them (us) to use as
> resources?
>
> Do you realy believe this is not useful?  Data is not useful?  Ach!
>
> Serving that community - even if it gets uncomfortable - if fair.
>
>
>
> > As you pointed out Pycon 2009 used web2py to register 800 users and
> > handled thousands of monetary transactions.
>
> This is significant.  Cooperated w/ django app for site info.
>
>
>

> >http://www.klasproducts.com/uses web2py to sell hundreds of products

>
> This is important, if small quantity.  It is a nice  site.
>
>
>

> >http://www.spinyc.com/uses web2py in their intranet

>
> This is significant, but anecdotal (nothing you can show someone for them to
> experience; perhaps a mockup that is open would be possible?).
> Runs with RorR as main site (and all we can see / point to).
>
>
>
> >http://www.whitepeaksoftware.com/ uses web2py for their web site
>
> For purposes of showing web2py look and feel, this is just a website front
> done in web2py;
> The storefront is something else; the blog is wordpress.  There is nothing
> of the "benefits" of web2py to show here I could see.
>
> That is not to criticize - no sense re-inventing wheels; use what is
> available, and build what is new and interesting.
>
> But from the standpoint of being a showcase example for this community to
> draw on - for people wanting to show sites
> that take advantage of the benefits of web2py, I do not feel this belongs on
> this list.
>
> We need to be more discerning.
>
>
>

> >http://diarywiz.com/is powered by web2py

> Your own words:  "...get *the job done* better, quicker..."

>
> It's what I hear ->  "You say this is better?  Show me."  Show me the jobs.
>
> *sigh*
>
> A little intentional focus on this kind of evidence, on

> *applied* "better, quicker" can only help web2py.

[Yarko Tymciurak]

On Mon, Jul 6, 2009 at 7:00 PM, mdipierro <mdip...@cs.depaul.edu> wrote:

Yarko,

we should not transform a strength into a weakness.

Massimo, Massimo, Massimo -

Facts are neither strength or weakness (these words from you seem strange indeed).
 

As I stressed many
times web2py is mostly used to develop intranet apps in corporate
environments (small and medium size businesses).

There are people do not want to say what they are doing with web2py
and do not want to advertise that they are using web2py because they
are not developing open source projects, because their paid by an
employer who does not want them to publicly say what they are doing,
because
they do not know if they will meet their deadline, and for many other
reasons

I agree - but you would have me believe that is the _only_ user base, and that challenges credibility,
and raises questions.

Besides, even IN THE CORPORATION - to get a project ok'd, you STILL need some evidence like I'm calling for,
so your arguments seem to me circular, avoiding...
 

While the fact that we want to have more information is a problem for
us, this is not a weakness of web2py.

I agree - not a weakness, merely the current opportunity (if it is recognized).
 

Those corporate users are
exactly our target users and they are entitled to keep their privacy.
I do not care if nobody says they are using web2py. I know we get more
tha 1000 visitors/day and more than 100downloads/day.

Yes - well, your personal interests are teaching.  My personal interests are connecting teachings to practical uses.
We have complementary objectives.

Downloading is fine;  learning is fine;  being able to apply what you learn to the benefit of others is... divine! ;-)
 

I see how everybody is eager to increase the users base.

I am not - increasing the user base should be a symptom; a sign that we are doing something right.

The question remains: who is that user base?  Why is it increasing?  (The answers will affect interpretation, of course!)

Downloads of web2py measure NO increasing the "I use this great website and it really helps me / our organization.  What was that thing is was
built with again???" - that is the base to increase (indirectly) now.

I made a plot of #users/time from the data in the google group.

What that is, is a plot of potential web site writers... not "users", coders (or "would-be" coders, people considering if the cost of entry is good for them, their ideas).

 

The
data is nicely fit by an exponential. What is remarkable is that there
are no bumps in the data. It is very smooth. This means web2py is not
spreading because of major events (like my reddit posts) but it is
spreading by word of mouth.

For people not familiar with website programming, web2py has an unmistakably low threshold (and some conceptual roadblocks)
For people who know website programming, there is also some special leverage (and some flexigility barriers, although the important ones are lowering all the time - which is great).

You have hit a particular market - two classes of web-developers (and would-be developers).

Now, what kind of end-user impact will these groups achieve?
 

So, we'll keep collecting information people want to send us and we
encourage them to do so. But, most importantly, we should all do our
best to talk about the tools we like to use and contribute develop.

And why we like to use them, and how they help (or hinder) our end goals, and how best to work towards those end goals,
and how best to see them, and to look at how they have worked in practice (so me may stay rooted in reality).
 

Massimo

Yarko

[Yarko Tymciurak]

Ach - I see why you are so protective of the user base, and all your efforts, long hours and fine teachings.

Let it not get in the way of stepping back now and then, and looking, seeing...

there is no shame in reality...