Webware for Python

Randall Smith wrote:
> Being able to reload modules with the server running means more uptime.  
> Webware can also watch mtimes on the files to reload them when they are 
> changed.

Somebody correct me if I'm wrong, but doesn't this have races that 
can lead to devious non-reproducible bugs?  IMHO a few seconds of 
downtime are far cheaper than hunting a bug that doesn't really exist.

     -- Daniel Newby

On Wed, 2004-03-24 at 22:53, Randall Smith wrote:
> So I ponder.  Why isn't Webware more aggressive as a general
> webservices
> app server.  Is there a SOAP implementation?  Would it be difficult
> to
> create one?  

I think it should be quite straightforward -- there are (several?) soap 
implementations for python, and as far as integrating it into Webware goes,
you should just be able to use the XMLRPC implementation as a guide.

> 3. Manageable.  The web interface for managing the service is nice.
> Being able to reload modules with the server running means more
> uptime.  Webware can also watch mtimes on the files to reload them when they
> are changed.

Although Webware allows you to reload individual modules, dependency
problems can be a big issue, as dependencies do not get reloaded
automatically.  You'll also run into problems if you reload modules
containing modified class definitions while there are still objects
instances created with the previous version of the class.  In short, you
don't want to go there.

Luckily, restarting Webware doesn't mean you'll drop requests.  The
appserver only shuts down after it was processed all pending requests.  
Requests which come in while the appserver is restarting will be delayed
somewhat (in apache), as mod_webkit attempts to reconnect to the appserver.
mod_webkit will retry though (up to 10 times I think), so as long as the
appserver comes up within a few seconds, all requests will be served.

> So back to my question.  Is Webware intended to be a webservices
> (web/xmlrpc/soap) app server?  

Certainly.  The only reason Webware lacks soap support is that no one has
taken the initiative to work on it.  I'm sure if that functionality existed,
people would use it.  If you're excited about doing that, go for it!

On a somewhat related note, I think Ian Bicking has done some work to
allow the appserver to more easily support multiple protocols, listening
on several ports (i.e. not just http).  I don't know what the status of this
work is, though.

peace,
Jason

> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> Webware-discuss mailing list
> Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss
--
Jason D. Hildebrand
jason@...

On Wed, 2004-03-24 at 22:53, Randall Smith wrote:
> Is there a better app server that 'all the other' Pythoneers are using and 
> I'm just missing?  If not, ring the bell!  
> Webware is such a simple and powerful framework for building web apps.  

I agree wholeheartedly and I wanted to follow up on the idea of improving
visibility of the Webware project.  For people who aren't subscribed
to the list (and there are probably many), the website is the primary
source of information.

It'd be great to make the Webware website more attractive and useful.
I'd like to see more links to related projects (kits, packages, templating
solutions which Webware users find useful), and tips on using Webware.

There was an effort last year to revamp the Webware website.  We had
contributions for a logo, and I think we voted/decided on one.  Does anyone
have a copy of the logo we chose?

Tom von Schwerdtner (tvon@...) began working on a new website for
the project.  I saw a preview of it at one point, but I haven't seen a
post from him in over 6 months.  I contacted him off-list 3 weeks ago,
asking if he still had the site and whether someone could continue work
on it.  No response though, so I guess we should just start a new one.

Is anyone interested in volunteering some time to design a new website
for the project?  I'd be happy to help coordinate this effort.

peace,
Jason

Hallo,
Jason Hildebrand hat gesagt: // Jason Hildebrand wrote:

> I agree wholeheartedly and I wanted to follow up on the idea of improving
> visibility of the Webware project.  For people who aren't subscribed
> to the list (and there are probably many), the website is the primary
> source of information.
> 
> It'd be great to make the Webware website more attractive and useful.
> I'd like to see more links to related projects (kits, packages, templating
> solutions which Webware users find useful), and tips on using Webware.
>
> Is anyone interested in volunteering some time to design a new website
> for the project?  I'd be happy to help coordinate this effort.

This would be really great. I would like to extent this effort to
include the "dogbowl" effect: The next Webware site IMO should be run
by Webware. I know this is next to impossible to do on the Sourceforge
servers, so maybe some external hosting of the pages would be needed. 

But anyway, having a Webware site that also serves as an example site
of what could be done with Webware would make WW much more visible per
se. And it also could lead to a very cool example application like
Plone is for Zope. Well, not that big, but you get the idea, I hope.

And maybe this even could breath some new live into some kind of
UserKit? ;)

ciao
-- 
 Frank Barknecht                               _ ______footils.org__

Frank Barknecht wrote:

>Hallo,
>Jason Hildebrand hat gesagt: // Jason Hildebrand wrote:
>
>  
>
>>I agree wholeheartedly and I wanted to follow up on the idea of improving
>>visibility of the Webware project.  For people who aren't subscribed
>>to the list (and there are probably many), the website is the primary
>>source of information.
>>
>>It'd be great to make the Webware website more attractive and useful.
>>I'd like to see more links to related projects (kits, packages, templating
>>solutions which Webware users find useful), and tips on using Webware.
>>
>>Is anyone interested in volunteering some time to design a new website
>>for the project?  I'd be happy to help coordinate this effort.
>>    
>>
>
>This would be really great. I would like to extent this effort to
>include the "dogbowl" effect: The next Webware site IMO should be run
>by Webware. I know this is next to impossible to do on the Sourceforge
>servers, so maybe some external hosting of the pages would be needed. 
>
>But anyway, having a Webware site that also serves as an example site
>of what could be done with Webware would make WW much more visible per
>se. And it also could lead to a very cool example application like
>Plone is for Zope. Well, not that big, but you get the idea, I hope.
>
>And maybe this even could breath some new live into some kind of
>UserKit? ;)
>
>ciao
>  
>
I started this many times, but never had the time to get into it.  I do 
have some ideas and even a room on a hosted server we can work with.

-Aaron

Well as far as the logos go, we did have a some contributions, but I
don't remember there actually being any decision as to which would be
official.  I still have all the submitted logos on my website and you
can certainly use any image you wish.  The site with the logos is:

http://balac.compmed.ucdavis.edu/logos/

Jose


-----Original Message-----
From: webware-discuss-admin@...
[mailto:webware-discuss-admin@...] On Behalf Of Jason
Hildebrand
Sent: Friday, March 26, 2004 8:18 AM
To: Randall Smith
Cc: webware-discuss
Subject: [Webware-discuss] improving Webware visibility


On Wed, 2004-03-24 at 22:53, Randall Smith wrote:
> Is there a better app server that 'all the other' Pythoneers are using

> and
> I'm just missing?  If not, ring the bell!  
> Webware is such a simple and powerful framework for building web apps.


I agree wholeheartedly and I wanted to follow up on the idea of
improving visibility of the Webware project.  For people who aren't
subscribed to the list (and there are probably many), the website is the
primary source of information.

It'd be great to make the Webware website more attractive and useful.
I'd like to see more links to related projects (kits, packages,
templating solutions which Webware users find useful), and tips on using
Webware.

There was an effort last year to revamp the Webware website.  We had
contributions for a logo, and I think we voted/decided on one.  Does
anyone have a copy of the logo we chose?

Tom von Schwerdtner (tvon@...) began working on a new website for
the project.  I saw a preview of it at one point, but I haven't seen a
post from him in over 6 months.  I contacted him off-list 3 weeks ago,
asking if he still had the site and whether someone could continue work
on it.  No response though, so I guess we should just start a new one.

Is anyone interested in volunteering some time to design a new website
for the project?  I'd be happy to help coordinate this effort.

peace,
Jason



-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
Webware-discuss mailing list Webware-discuss@...
https://lists.sourceforge.net/lists/listinfo/webware-discuss

> This would be really great. I would like to extent this effort to
> include the "dogbowl" effect: The next Webware site IMO should be run
> by Webware. I know this is next to impossible to do on the Sourceforge
> servers, so maybe some external hosting of the pages would be needed. 

I can donate 2U or rack space at our POP (Scranton Pennsylvania, USA) for a server to host WebWare's site. I would suggest a dedicated server so that it could be optimized for WebWare use.

Eric Radman
 http://www.teisprint.net

On Fri, 2004-03-26 at 12:12, jose wrote:
> Well as far as the logos go, we did have a some contributions, but I
> don't remember there actually being any decision as to which would be
> official.  

You're right, I don't think we actually did decide on one.  Maybe we should
try to do that...  

>From the logos posted so far, people seem to be most excited about (some
variation) of David Eriksson's.  It might be useful to cut some of these out
and post them on the site, so that we can comment on them separately.  We
could also try changing the colour to see if we prefer something other than
blue.  

Anyways, I'd encourage people to submit new and/or derived logos to Jose
to post on the site, and maybe we can come to some sort of consensus.

peace,
Jason

On Fri, 2004-03-26 at 11:41, Frank Barknecht wrote:
> This would be really great. I would like to extent this effort to
> include the "dogbowl" effect: The next Webware site IMO should be run
> by Webware. I know this is next to impossible to do on the
> Sourceforge
> servers, so maybe some external hosting of the pages would be needed.

Yes, that would be great.  Eric Radman has offered to provide space for a
server, but I don't know if we'll be able to come up with a server.
Can anyone provide space on an existing server which is Webware-friendly?

> But anyway, having a Webware site that also serves as an example site
> of what could be done with Webware would make WW much more visible
> per
> se. And it also could lead to a very cool example application like
> Plone is for Zope. Well, not that big, but you get the idea, I hope.

> And maybe this even could breath some new live into some kind of
> UserKit? ;)

Let's not bite off more than we can chew.  ;) 

I agree that it would be a great sample app, but I'd propose not to build
a general "portal solution" (at least not right away ;) --  but to build a
site customized to the needs of the Webware project.  It should be simple
to start with, so that we actually get it up and running sometime soon.
We can add bells and whistles later.

peace,
Jason

Hello Jason,

Friday, March 26, 2004, 7:47:03 PM, you wrote:

JH> On Fri, 2004-03-26 at 11:41, Frank Barknecht wrote:
>> This would be really great. I would like to extent this effort to
>> include the "dogbowl" effect: The next Webware site IMO should be run
>> by Webware. I know this is next to impossible to do on the
>> Sourceforge
>> servers, so maybe some external hosting of the pages would be needed.

JH> Yes, that would be great.  Eric Radman has offered to provide space for a
JH> server, but I don't know if we'll be able to come up with a server.
JH> Can anyone provide space on an existing server which is Webware-friendly?

I would really suggest a virtual server. An own piece of hardware is i
think not a good choice as long as we don't get the popularity of
apache. If something happens to the hardware, how many days will it
take to find someone responsible for changing the hardware ?


-- 
Best regards,
 Lothar                            mailto:llothar@...

Hallo,
Jason Hildebrand hat gesagt: // Jason Hildebrand wrote:
> I agree that it would be a great sample app, but I'd propose not to build
> a general "portal solution" (at least not right away ;) --  but to build a
> site customized to the needs of the Webware project.  It should be simple
> to start with, so that we actually get it up and running sometime soon.
> We can add bells and whistles later.

I totally agree. This is something, that can grow over time. But a
simple, DB-driven content manager should be fast to build. I mean,
this is something, Webware is used for daily by us, so it's probably
on everyone's harddisk anyway. (I only prefer SQLObject, not
MiddleKit.)

Regarding graphics: Generally I think, that most coders don't make
good graphics people, that's why normally those are two different
departments. Maybe someone here has some graphics wizard at hand who
could put together a cool layout? (I don't care much about logos, a
good layout is more important for a future WW website.)

ciao
-- 
 Frank Barknecht                               _ ______footils.org__

> -----Original Message-----
> From: webware-discuss-admin@...
> [mailto:webware-discuss-admin@... Behalf Of Jason
> Hildebrand
> Sent: Friday, March 26, 2004 10:36 AM
>
> On Fri, 2004-03-26 at 12:12, jose wrote:
> > Well as far as the logos go, we did have a some contributions, but I
> > don't remember there actually being any decision as to which would be
> > official.
>
> You're right, I don't think we actually did decide on one.  Maybe
> we should
> try to do that...
>
> >From the logos posted so far, people seem to be most excited about (some
> variation) of David Eriksson's.  It might be useful to cut some
> of these out
> and post them on the site, so that we can comment on them separately.  We
> could also try changing the colour to see if we prefer something
> other than
> blue.
>
> Anyways, I'd encourage people to submit new and/or derived logos to Jose
> to post on the site, and maybe we can come to some sort of consensus.
>
> peace,
> Jason
>

Coming up with a quality logo can be a journey, but a worthwhile one.

One fun exercise to really get the process started is to ask everyone on
the list, or via a survey of some kind, to give one phrase, or idea, or a
feeling, or color, or actor, or movie, etc that Webware reminds them of.

e.g.:
If Webware were a feeling, what would it be?
If Webware were one idea, what would it be?
How would you summarize Webware in less than 5 words?

This survey could even be a page on the Wiki that gets extended with
answers.

So you end up with a whole bunch of little phrases and metaphors for
Webware, which become the inspiration for developing a logo that conveys
those ideas. This gives a logo designer a pool of ideas to pull from,
and also provides a kind of ideological baseline which in turn provides
consistency across different logo designers' efforts (if they read it!).

At some point the group can say, Ok we've got this list of ways to
describe Webware, now everybody start submitting logos. And once enough
submissions come in, the group can vote or the core team managing the
development can just pick from those options.

Regards,
Ben

On Fri, 2004-03-26 at 13:20, Lothar Scholz wrote:

> I would really suggest a virtual server. An own piece of hardware is
> i
> think not a good choice as long as we don't get the popularity of
> apache. If something happens to the hardware, how many days will it
> take to find someone responsible for changing the hardware ?

Good point.  This would be my preference, too.  

--
Jason D. Hildebrand
jason@...

Well if the group at large thinks that webware should invest in its own
server then isn't it time us users put our money where our mouth is?  I
would certainly be happy to donate $10 bucks to the cause now if only 30
other folks would feel the same way we could get something.  Source
forge now has this program where people can donate money to a project
maybe that should be added to the SF site.

Jose

-----Original Message-----
From: webware-discuss-admin@...
[mailto:webware-discuss-admin@...] On Behalf Of Jason
Hildebrand
Sent: Friday, March 26, 2004 11:56 AM
To: Lothar Scholz
Cc: webware-discuss
Subject: Re: Re[2]: [Webware-discuss] improving Webware visibility


On Fri, 2004-03-26 at 13:20, Lothar Scholz wrote:

> I would really suggest a virtual server. An own piece of hardware is i
> think not a good choice as long as we don't get the popularity of
> apache. If something happens to the hardware, how many days will it
> take to find someone responsible for changing the hardware ?

Good point.  This would be my preference, too.  

--
Jason D. Hildebrand
jason@...



-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
_______________________________________________
Webware-discuss mailing list Webware-discuss@...
https://lists.sourceforge.net/lists/listinfo/webware-discuss

On Fri, 2004-03-26 at 14:32, jose wrote:
> Well if the group at large thinks that webware should invest in its
> own
> server then isn't it time us users put our money where our mouth is?
> I
> would certainly be happy to donate $10 bucks to the cause now if only
> 30
> other folks would feel the same way we could get something.  Source
> forge now has this program where people can donate money to a project
> maybe that should be added to the SF site.

Let's wait and see if someone can offer hosting for free.  If not we can
pursue other options.  BTW - I'm keeping my eye on "webware.org"; it expires
in 6 days.

peace,
Jason

Count me in: Maybe someone can put up a paypal-donate page (I'm in Chile=
)

<cita quien=3D"jose">
> Well if the group at large thinks that webware should invest in its ow=
n
> server then isn't it time us users put our money where our mouth is?  =
I
> would certainly be happy to donate $10 bucks to the cause now if only =
30
> other folks would feel the same way we could get something.  Source
> forge now has this program where people can donate money to a project
> maybe that should be added to the SF site.
>
> Jose
>
> -----Original Message-----
> From: webware-discuss-admin@...
> [mailto:webware-discuss-admin@...] On Behalf Of Jaso=
n
> Hildebrand
> Sent: Friday, March 26, 2004 11:56 AM
> To: Lothar Scholz
> Cc: webware-discuss
> Subject: Re: Re[2]: [Webware-discuss] improving Webware visibility
>
>
> On Fri, 2004-03-26 at 13:20, Lothar Scholz wrote:
>
>> I would really suggest a virtual server. An own piece of hardware is =
i
>> think not a good choice as long as we don't get the popularity of
>> apache. If something happens to the hardware, how many days will it
>> take to find someone responsible for changing the hardware ?
>
> Good point.  This would be my preference, too.
>
> --
> Jason D. Hildebrand
> jason@...
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=3D1470&alloc_id=3D3638&op=3D=
click
> _______________________________________________
> Webware-discuss mailing list Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss
>
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=3D1470&alloc_id=3D3638&op=3D=
click
> _______________________________________________
> Webware-discuss mailing list
> Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss
>


-- 
Eduardo Elgueta
Senior Consultant
Navix

correo/email: eelgueta@...
tel=E9fono/phone: +56 (2) 209-8905
celular/mobile: +56 (9) 821-0033
web: http://www.navix.cl

Av. Pedro de Valdivia 555 Of. 216
Providencia 750-0913
Santiago, Chile

I might be convinced to host the Webware site on my dedicated box.  It's
one that I have leased through EV1 for almost 2 years now.  I have been
using Webware for work projects but I am planning to start using it for
personal ones too, so I will probably be installing Webware on the box
soon anyway.

Jason, or anyone else, if you're willing to put in the work to get
things orginized then I think I can part with some system resources for
the good of the group.

Drop me a line (aswitzer@...), or find me on IRC if you want to
discuss it further.

- Aaron Switzer



On Fri, 2004-03-26 at 15:47, Jason Hildebrand wrote:
> On Fri, 2004-03-26 at 14:32, jose wrote:
> > Well if the group at large thinks that webware should invest in its
> > own
> > server then isn't it time us users put our money where our mouth is?
> > I
> > would certainly be happy to donate $10 bucks to the cause now if only
> > 30
> > other folks would feel the same way we could get something.  Source
> > forge now has this program where people can donate money to a project
> > maybe that should be added to the SF site.
> 
> Let's wait and see if someone can offer hosting for free.  If not we can
> pursue other options.  BTW - I'm keeping my eye on "webware.org"; it expires
> in 6 days.
> 
> peace,
> Jason
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> Webware-discuss mailing list
> Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss
-- 
Aaron Switzer
aaron@...
416-350-8376 ext. 235

* Ben Parker <ben@...> [2004-03-26 15:02]:> 
> So you end up with a whole bunch of little phrases and metaphors for
> Webware, which become the inspiration for developing a logo that conveys
> those ideas. This gives a logo designer a pool of ideas to pull from,
> and also provides a kind of ideological baseline which in turn provides
> consistency across different logo designers' efforts (if they read it!).

glad to see this campaign picking up again...

In addition to Ben's good points above, I'd like to confirm/clarify a
few details:

- is the full name 'Webware' or 'Webware for Python'?

- if the latter, is it really necessary to have 'for Python'?

- i think a more functional 'sub title' such as 'application server',
or 'web components', or 'framework', or something along those
lines would be better in both a marketing and informative
sense. 'appserver' of course is the most marketingese, and I'd guess
makes some people shudder. But I don't think that should discount it.

- are any color palettes/schemes ruled in or out? or are we still wide
open?

- does any of this effort require approval from any particular people?
from Chuck for example? or is this group consensus? this is an area
that often seems to trip up these types of efforts, so it'd be good to
know ahead of time...

Looking forward to trying to help out here, just wanting to get some
ground rules before starting...

-- 

___________________________
toddgrimason*todd@...

On Fri, 2004-03-26 at 16:12, Todd Grimason wrote:
> - is the full name 'Webware' or 'Webware for Python'?

AFAIK the official name is "Webware for Python", to disambiguate from other uses 
of the term "Webware" or "WebWare" on the internet.

> - if the latter, is it really necessary to have 'for Python'?
> - i think a more functional 'sub title' such as 'application server',
> or 'web components', or 'framework', or something along those
> lines would be better in both a marketing and informative
> sense. 'appserver' of course is the most marketingese, and I'd guess
> makes some people shudder. But I don't think that should discount it.

I think for a lot of people, the fact that it's Python is important -- so I
wouldn't necessarily want to throw that away.  But I agree that a more
functional subtitle could be beneficial.

> - are any color palettes/schemes ruled in or out? or are we still
> wide
> open?

I'd say it's wide open until someone starts working on a site.  And even
after that it's not too late to change, you just have to deal with inertia. :)

> - does any of this effort require approval from any particular
> people?
> from Chuck for example? or is this group consensus? this is an area
> that often seems to trip up these types of efforts, so it'd be good
> to know ahead of time...

Good question.  I don't know if Chuck considers himself to be the Webware
BDFL -- I'll let him speak to that.  :)  

My take on the situation is that most developers are very busy and don't
have huge amounts of time to invest in the project right now, but would
be very happy to see a new website.

--
Jason D. Hildebrand
jason@...

On Fri, 2004-03-26 at 15:32, jose wrote:
> Well if the group at large thinks that webware should invest in its own
> server then isn't it time us users put our money where our mouth is?  I
> would certainly be happy to donate $10 bucks to the cause now if only 30
> other folks would feel the same way we could get something.  Source
> forge now has this program where people can donate money to a project
> maybe that should be added to the SF site.

I wholeheartedly agree.  Jose should pay 10 bucks. :)

Hello Aaron,

Friday, March 26, 2004, 10:05:11 PM, you wrote:

AS> I might be convinced to host the Webware site on my dedicated box.  It's
AS> one that I have leased through EV1 for almost 2 years now.  I have been
AS> using Webware for work projects but I am planning to start using it for
AS> personal ones too, so I will probably be installing Webware on the box
AS> soon anyway.

I tried EV1 - when they called themself Rackshack. At least at this
time the transfer rate out of the USA was terrible slow for a server.
Getting 15 KB (instead of 90KB from sourceforge) via DSL from
Berlin/Germany.

-- 
Best regards,
 Lothar                            mailto:llothar@...

> I think for a lot of people, the fact that it's Python is important -- so I
> wouldn't necessarily want to throw that away.  But I agree that a more
> functional subtitle could be beneficial.

I agree. The fact that this is a Python framework is very important. This
framework would not have the same appeal if it was VB!

Eric

Lothar Scholz wrote:
> I tried EV1 - when they called themself Rackshack. At least at this
> time the transfer rate out of the USA was terrible slow for a server.
> Getting 15 KB (instead of 90KB from sourceforge) via DSL from
> Berlin/Germany.

Is this the same EV1 that bought a SCO license?

> Is this the same EV1 that bought a SCO license?

Yes.

On Mar 26, 2004, at 12:25 AM, Jason Hildebrand wrote:
> Although Webware allows you to reload individual modules, dependency
> problems can be a big issue, as dependencies do not get reloaded
> automatically.  You'll also run into problems if you reload modules
> containing modified class definitions while there are still objects
> instances created with the previous version of the class.  In short, 
> you
> don't want to go there.
>
> Luckily, restarting Webware doesn't mean you'll drop requests.  The
> appserver only shuts down after it was processed all pending requests.
> Requests which come in while the appserver is restarting will be 
> delayed
> somewhat (in apache), as mod_webkit attempts to reconnect to the 
> appserver.
> mod_webkit will retry though (up to 10 times I think), so as long as 
> the
> appserver comes up within a few seconds, all requests will be served.

I actually gave a lightning talk at PyCon describing how Webware deals 
with this, and why all other ways of dealing with it suck (since 
there's corner cases that will drive you crazy).  I still have to put 
the slides up online, but then there's nothing that will be new in it 
to the Webware crowd.

>> So back to my question.  Is Webware intended to be a webservices
>> (web/xmlrpc/soap) app server?
>
> Certainly.  The only reason Webware lacks soap support is that no one 
> has
> taken the initiative to work on it.  I'm sure if that functionality 
> existed,
> people would use it.  If you're excited about doing that, go for it!

And I bet it would be really easy, as the XMLRPC support for Webware is 
very simple.

> On a somewhat related note, I think Ian Bicking has done some work to
> allow the appserver to more easily support multiple protocols, 
> listening
> on several ports (i.e. not just http).  I don't know what the status 
> of this
> work is, though.

In Webware CVS it has code for this.  Actually, it's the code for 
supporting HTTP, so that you don't need Apache (previously only the 
mod_webkit protocol was supported).  (There's actually a third protocol 
-- the MONITOR stuff -- but I never even tested it, and I'm pretty sure 
it had been broken for a while preceding those changes)

I'm pretty sure it also allows for the possibility of request objects 
besides HTTPRequest.  But that won't actually be necessary for any of 
this -- XMLRPCServlet just accepts an HTTPRequest (which was how XMLRPC 
was designed), and SOAP is pretty much the same.  (Unless you want 
SOAP-over-SMTP or other such tricks -- which isn't impossible, but 
certainly isn't the place to start)

--
Ian Bicking | ianb@... | http://blog.ianbicking.org

On Mar 26, 2004, at 2:54 PM, Ben Parker wrote:
> e.g.:
> If Webware were a feeling, what would it be?
> If Webware were one idea, what would it be?
> How would you summarize Webware in less than 5 words?

Well, lately when I've been describing Webware to other Python 
programmers I often describe it as open pastures -- it doesn't provide 
much (compared to Zope), but it gives you a lot of freedom.

At least, that's what it is coming from Zope.  Coming from another 
Python framework or another language it's going to be something very 
different.

--
Ian Bicking | ianb@... | http://blog.ianbicking.org

On Fri, 26 Mar 2004 17:12:26 -0500, Todd Grimason wrote:
> In addition to Ben's good points above, I'd like to confirm/clarify
> a few details:
>
> - is the full name 'Webware' or 'Webware for Python'?

When I started Webware, there were already other Webwares (but none of them=
 were app servers). There was a column, webware.com, a non-hosted=
 webware.org, the generic term, etc.

I made the official, public name "Webware for Python" to avoid stepping on=
 toes. Plus Python rocks and Webware certainly feels pretty Pythonic since=
 it is straightforward and practical.

> - does any of this effort require approval from any particular
> people? from Chuck for example? or is this group consensus? this is
> an area that often seems to trip up these types of efforts, so it'd
> be good to know ahead of time...

I'm comfortable that if we want to revamp the web page or have a logo, that=
 we decide as a group either as vote by the developers or by the entire=
 community.

On Fri, 26 Mar 2004 16:43:41 -0600, Jason Hildebrand wrote:
> AFAIK the official name is "Webware for Python", to disambiguate
> from other uses of the term "Webware" or "WebWare" on the internet.

Bingo.

> Good question.  I don't know if Chuck considers himself to be the
> Webware BDFL -- I'll let him speak to that.  :)

Over time we've naturally evolved into an oligarchy of developers where I'm=
 no longer "laying it down" code- and policy-wise like I did in the first=
 year. And so far, so good!

I'm comfortable with people taking initiatives like these and following=
 through. If you're getting consensus from Webware developers like Jason,=
 Ian and Geoff, consider it blessed.

I think the oligarchy can work for most, maybe all, things. I won't be=
 flexing my BDFL title unless some conflict arises or you start embedding=
 Perl in the source base.  :-)

> My take on the situation is that most developers are very busy and
> don't have huge amounts of time to invest in the project right now,
> but would be very happy to see a new website.

Bingo.


Regarding money, we can almost certainly get hosting for free from somebody=
 and it looks like a somebody has already appeared. But we could also use=
 money to run ads on SF.net and other places to see if we can ramp up our=
 user base which inevitably leads to good things. Just a thought.


-Chuck
-- 
http://ChuckEsterbrook.com/

* Chuck Esterbrook <ChuckEsterbrook@...> [2004-03-27 18:12]:
> On Fri, 26 Mar 2004 17:12:26 -0500, Todd Grimason wrote:
> > In addition to Ben's good points above, I'd like to confirm/clarify
> > a few details:
> >
> > - is the full name 'Webware' or 'Webware for Python'?
> 
> When I started Webware, there were already other Webwares (but none of them were app servers). There was a column, webware.com, a non-hosted webware.org, the generic term, etc.
> 
> I made the official, public name "Webware for Python" to avoid stepping on toes. Plus Python rocks and Webware certainly feels pretty Pythonic since it is straightforward and practical.

Ah - I didn't realize there were(is?) other projects out there with
the same name. I certainly didn't mean to attack Python (obviously),
just thought it kinda sounds like 'Struts for Java' or 'Zope for
Python', etc.

Going to 'Webware', with maybe a tagline kinda like 'Python Appserver'
or whatever might be something to consider. But based on my other
forays into this topic on other projects many people seem to have
strong personal attachments to things like names, so I won't be
pushing any further... 

(though I'd say there's a bit of precedent set from other projects
such as Zope, and Java, Ruby, etc.. equivalents. but I suppose some
might see that as more reason not to go that way.)

-- 

___________________________
toddgrimason*todd@...

On Sat, 2004-03-27 at 19:59, Todd Grimason wrote:
> Ah - I didn't realize there were(is?) other projects out there with
> the same name. I certainly didn't mean to attack Python (obviously),
> just thought it kinda sounds like 'Struts for Java' or 'Zope for
> Python', etc.

> Going to 'Webware', with maybe a tagline kinda like 'Python
> Appserver'
> or whatever might be something to consider. But based on my other
> forays into this topic on other projects many people seem to have
> strong personal attachments to things like names, so I won't be
> pushing any further...

I think it might be worth pursuing this a bit further -- it certainly can't
hurt to brainstorm a little bit.

I like the idea of a tagline which gives a better clue as to what the
project is about; perhaps using a word like "toolkit" or "framework" in the
tagline ("Appserver" doesn't do it for me, <shrug>).  

I guess the image of "tools" appeals to me because that's what I feel
Webware gives me: lots of tools which I can use how I want.  It also
conveys the idea that Webware is made up of several components (kits),
not all of which must be used.  You pick the tools you want to use.

Using this idea, I find "Webware" with tagline "Python Web Application Toolkit" 
quite appealing.  

--
Jason D. Hildebrand
jason@...

On 20:59 Sat 27 Mar     , Todd Grimason wrote:
> Ah - I didn't realize there were(is?) other projects out there with
> the same name. I certainly didn't mean to attack Python (obviously),
> just thought it kinda sounds like 'Struts for Java' or 'Zope for
> Python', etc.
> 
> Going to 'Webware', with maybe a tagline kinda like 'Python Appserver'
> or whatever might be something to consider. But based on my other
> forays into this topic on other projects many people seem to have
> strong personal attachments to things like names, so I won't be
> pushing any further... 
> 
> (though I'd say there's a bit of precedent set from other projects
> such as Zope, and Java, Ruby, etc.. equivalents. but I suppose some
> might see that as more reason not to go that way.)

"Cold Fusion" is an example of a great name--it's a word picture that's unique.
The word "Zope" also carries a very clear identity that "WebWare for Python"
does not.

I recommend a new name!

Armature
  Python Kinetics for the Web

Has anyone else thought of what they would name it? This topic is important
because it changes the way that this framework is presented. In the idea above,
a visual theme could now incorporate graphics terminology that borrows from the
analogy of an electric motor.

Eric

Eric Radman wrote:
> "Cold Fusion" is an example of a great name--it's a word picture that's unique.
> The word "Zope" also carries a very clear identity that "WebWare for Python"
> does not.

Well, there's one thing I think several of us agree on about the name: 
"Webware" **not** "WebWare".

With-constant-vigilence-against-mixed-case'ly y'rs,
   Ian

MAy I humbly suggest sticking with webware.

The names out there. The branding sticks. Some projects disapear after the
established brand name changes. Ever remember what happened to
Killustrator?

Also if one MUST change it, theres something microsoft worked out:: The
name should indicate what it is.

word = word processor
windows = windowing environment
office = office suite.

So if you MUST, then it should indicate that its at core a servlet and
psp engine with 'middleware bits'.

Seriously tho. Keep the name. I found it via my nifty little Dietel Python
book and if the name had changed , I'd of assumed it was a dead project
and whatever it became was 'yet another python cgi jibbajabba'

--
Shayne O'Neill
http://perth.indymedia.org
I know how hard it is for you to put food on your family."
 ----George W. Bush

On Mon, 29 Mar 2004, Eric Radman wrote:

> On 20:59 Sat 27 Mar     , Todd Grimason wrote:
> > Ah - I didn't realize there were(is?) other projects out there with
> > the same name. I certainly didn't mean to attack Python (obviously),
> > just thought it kinda sounds like 'Struts for Java' or 'Zope for
> > Python', etc.
> >
> > Going to 'Webware', with maybe a tagline kinda like 'Python Appserver'
> > or whatever might be something to consider. But based on my other
> > forays into this topic on other projects many people seem to have
> > strong personal attachments to things like names, so I won't be
> > pushing any further...
> >
> > (though I'd say there's a bit of precedent set from other projects
> > such as Zope, and Java, Ruby, etc.. equivalents. but I suppose some
> > might see that as more reason not to go that way.)
>
> "Cold Fusion" is an example of a great name--it's a word picture that's unique.
> The word "Zope" also carries a very clear identity that "WebWare for Python"
> does not.
>
> I recommend a new name!
>
> Armature
>   Python Kinetics for the Web
>
> Has anyone else thought of what they would name it? This topic is important
> because it changes the way that this framework is presented. In the idea above,
> a visual theme could now incorporate graphics terminology that borrows from the
> analogy of an electric motor.
>
> Eric
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> Webware-discuss mailing list
> Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss
>

Hello everyone,

I know it's really hard to get everyone to agree on a logo and a name,
but one thing is for sure--Webware needs a nice logo. In the interest
of progress, I made a logo this morning:

http://www.alphamotion.com/logo1_02.gif

I've always liked pygame's logo and cherrypy's. I guess I dig the
cartoon-look. Anyway, how does it grab you?

Jacob
   ...
       _/ _/  _/   Jacob Hanson
      _/ _/_/_/   mailto:jacdx@...
 _/_/_/ _/  _/   http://www.jacobhanson.com


Monday, March 29, 2004, 11:25:52 AM, you wrote:
> MAy I humbly suggest sticking with webware.

> The names out there. The branding sticks. Some projects disapear after the
> established brand name changes. Ever remember what happened to
> Killustrator?

On Mon, Mar 29, 2004 at 11:59:04AM -0700, Jacob Hanson wrote:
> Hello everyone,
> 
> I know it's really hard to get everyone to agree on a logo and a name,
> but one thing is for sure--Webware needs a nice logo. In the interest
> of progress, I made a logo this morning:
> 
> http://www.alphamotion.com/logo1_02.gif
> 

nice :))) 

regards,
kender

I really like this logo... I like it better than the other ones that
were mentioned earlier.  How would an icon size version of it look
like?  The big one is nice for the head of a web page, but when I think
of a logo, I usually think of something extremely simple and
recognizible, like the Nike swoosh or Red Hat's, well, red hat.  Maybe
it could be something like a simpler version of the python with the
wrench in it's mouth and a blue W behind it... just a thought.
 
Secondly, for those wondering about my hosting offer.  Yes, the server
is through EV1 (formerly rackshack.net) and yes it's the same EV1 that
gave money to SCO.  I have been considering moving to a new company for
a while, but it would be a huge hassle getting the various websites a
run setup again.  But if anyone in the Webware community has a good
suggestion for a company to move to I may just take the plunge.  Since
I've had the machine for a couple of years, even just leasing a new
server from EV1 and canceling my current one would get me a better
system for cheaper.

- Aaron Switzer



On Mon, 2004-03-29 at 13:59, Jacob Hanson wrote:
> Hello everyone,
> 
> I know it's really hard to get everyone to agree on a logo and a name,
> but one thing is for sure--Webware needs a nice logo. In the interest
> of progress, I made a logo this morning:
> 
> http://www.alphamotion.com/logo1_02.gif
> 
> I've always liked pygame's logo and cherrypy's. I guess I dig the
> cartoon-look. Anyway, how does it grab you?
> 
> Jacob
>    ...
>        _/ _/  _/   Jacob Hanson
>       _/ _/_/_/   mailto:jacdx@...
>  _/_/_/ _/  _/   http://www.jacobhanson.com
> 
> 
> Monday, March 29, 2004, 11:25:52 AM, you wrote:
> > MAy I humbly suggest sticking with webware.
> 
> > The names out there. The branding sticks. Some projects disapear after the
> > established brand name changes. Ever remember what happened to
> > Killustrator?
> 
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> Webware-discuss mailing list
> Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss
-- 
Aaron Switzer
aaron@...
416-350-8376 ext. 235

Jacob Hanson wrote:
> Hello everyone,
> 
> I know it's really hard to get everyone to agree on a logo and a name,
> but one thing is for sure--Webware needs a nice logo. In the interest
> of progress, I made a logo this morning:
> 
> http://www.alphamotion.com/logo1_02.gif
> 
> I've always liked pygame's logo and cherrypy's. I guess I dig the
> cartoon-look. Anyway, how does it grab you?

I quite like it.  It's not too corporate or boring, but still kind of 
professional feeling.  In an irreverent sort of way.

   Ian

This was discussed in regards to a logo for Python overall, to the
point that heads exploded, but I'll mention it anyway: 

You generally need to decide whom your logo is meant to appeal to, and
what values it should relay.

[Here I would cut and paste 987906 emails from the marketing list]

In short - if the goal is to be taken more seriously, and
"professional", I'd vote against cartoon-ish approaches. Yes, even to
the point of "boring", though hopefully not.

If the point is just a catchy logo other Python programmers will
appreciate, then it's less vital - though if their usage is dependent
on departmental approval, it still matters.

Just thought I'd shortcut things here so we don't repeat that long
long thread.

-- 

___________________________
toddgrimason*todd@...

Todd Grimason wrote:
> This was discussed in regards to a logo for Python overall, to the
> point that heads exploded, but I'll mention it anyway: 
> 
> You generally need to decide whom your logo is meant to appeal to, and
> what values it should relay.
> 
> [Here I would cut and paste 987906 emails from the marketing list]
> 
> In short - if the goal is to be taken more seriously, and
> "professional", I'd vote against cartoon-ish approaches. Yes, even to
> the point of "boring", though hopefully not.

Realistically, Webware isn't going to appeal to the people that a 
corporate logo would appeal to.  Webware has the potential to bring in 
developers who are enthusiastic about Python.  Those developers may, in 
turn, convince boring corporate types.  But the boring corporate types 
just don't hang out in the same circles as Webware.  They probably don't 
even appreciate where Webware sits in the stack of tools a developer 
would use.

> If the point is just a catchy logo other Python programmers will
> appreciate, then it's less vital - though if their usage is dependent
> on departmental approval, it still matters.

We don't have any corporate backing or much of that appeal, but we can 
still appeal to management types that have an appreciation for open 
source software and the environment that entails.  That's my evaluation 
of Webware's appeal strictly on a technical level.

In that light, Webware needs not to show that it is corporate, but that 
it is alive and actively supported.  And even if there's not a huge 
amount of active development of the Webware core, it *is* consistently 
supported by the community on this list.  And an enlightened manager 
*will* still be concerned about the viability of the project.

 From that perspective an irreverent logo doesn't really hurt -- what 
you really need to show is that there's some care and effort.  A quite 
valid criticism is that currently the Webware site doesn't show that. 
So I think it's more important to do *something* rather than the "right" 
thing.  Personally, this logo is the one I like best so far.  But I 
wouldn't veto another logo.


> Just thought I'd shortcut things here so we don't repeat that long
> long thread.

I agree with everything Ian said here.

-Chuck
-- 
http://ChuckEsterbrook.com/



On Mon, 29 Mar 2004 16:06:29 -0600, Ian Bicking wrote:
> Todd Grimason wrote:
>> This was discussed in regards to a logo for Python overall, to
>> the point that heads exploded, but I'll mention it anyway:
>>
>> You generally need to decide whom your logo is meant to appeal
>> to, and what values it should relay.
>>
>> [Here I would cut and paste 987906 emails from the marketing list]
>>
>>
>> In short - if the goal is to be taken more seriously, and
>> "professional", I'd vote against cartoon-ish approaches. Yes,
>> even to the point of "boring", though hopefully not.
>>
>
> Realistically, Webware isn't going to appeal to the people that a
> corporate logo would appeal to.  Webware has the potential to bring
> in developers who are enthusiastic about Python.  Those developers
> may, in turn, convince boring corporate types.  But the boring
> corporate types just don't hang out in the same circles as Webware.
>  They probably don't even appreciate where Webware sits in the
> stack of tools a developer would use.
>
>> If the point is just a catchy logo other Python programmers will
>> appreciate, then it's less vital - though if their usage is
>> dependent on departmental approval, it still matters.
>>
>
> We don't have any corporate backing or much of that appeal, but we
> can still appeal to management types that have an appreciation for
> open source software and the environment that entails.  That's my
> evaluation of Webware's appeal strictly on a technical level.
>
> In that light, Webware needs not to show that it is corporate, but
> that it is alive and actively supported.  And even if there's not a
> huge amount of active development of the Webware core, it *is*
> consistently supported by the community on this list.  And an
> enlightened manager *will* still be concerned about the viability
> of the project.
>
> From that perspective an irreverent logo doesn't really hurt --
> what you really need to show is that there's some care and effort.
> A quite valid criticism is that currently the Webware site doesn't
> show that. So I think it's more important to do *something* rather
> than the "right" thing.  Personally, this logo is the one I like
> best so far.  But I wouldn't veto another logo.
>
>
>> Just thought I'd shortcut things here so we don't repeat that
>> long long thread.

Hello everyone,

I've received quite a bit of feedback today and I'm happy to say that
it's generally been quite positive. A few of you have felt that the
feel should lean a little bit more towards corporate and I agree. So,
I made a couple of adjustments that add some of that without giving
up the overall jovial nature. And I made some little 'powered by' images.

 http://www.alphamotion.com/jacob_hanson-logo_03.gif
 http://www.alphamotion.com/jacob_hanson-powered_01.gif
 http://www.alphamotion.com/jacob_hanson-powered_02.gif

I wasn't aware that there have been other attempts made at this before,
(thanks for the link Jose), but it looks like we've been deciding for
almost a year. :] Barring any objections to the current version, it
looks like we're pretty close to a *gasp* general consensus. I'm not
real picky--I'm mostly interested in seeing a decision made soon.

Jacob
   ...
       _/ _/  _/   Jacob Hanson
      _/ _/_/_/   mailto:jacdx@...
 _/_/_/ _/  _/   http://www.jacobhanson.com


Monday, March 29, 2004, 3:14:38 PM, you wrote:
> I agree with everything Ian said here.

> -Chuck
> --
> http://ChuckEsterbrook.com/

*snip*

Hallo,
Chuck Esterbrook hat gesagt: // Chuck Esterbrook wrote:

> I agree with everything Ian said here.

Me, too. ;)

In my opinion a logo is not important at all, or at least a decision
could be made much later. What *is* important, though, is a website
relaunch. And this is *much* more important than what logo later on
should be at the top of the page. But a website that has almost the
same content as last year really doesn't give that good an impression
for a project like Webware. 

Just my .05 cents.

ciao
-- 
 Frank Barknecht                               _ ______footils.org__

On Mar 29, 2004, at 3:40 PM, Jacob Hanson wrote:

> I wasn't aware that there have been other attempts made at this before,
> (thanks for the link Jose), but it looks like we've been deciding for
> almost a year. :] Barring any objections to the current version, it
> looks like we're pretty close to a *gasp* general consensus. I'm not
> real picky--I'm mostly interested in seeing a decision made soon.

Very new to Webware (or Python for that matter) so a little hesitant to 
offer suggestions as yet...  But while I like the design, I wonder if 
it really says anything about what Webware for Python does. I mean the 
suggestion could apply to any Python framework for any purpose.

Would something other than a wrench fit better? For example:

	<http://www.douglasburchard.com/demo-logo_01.jpg>

I know, there are several problems with this specific example, but 
perhaps someone could offer a better metaphor...

As for the concern that a 'mascot' is too cartoonish, there are plenty 
of examples to refute this. For example: Linux's Tux, BSD's Deamon, 
Darwin's Hexly, etc.  Personally, as a consultant I'd have no problems 
selling Webware to a client because of a cartoonish logo, as long as 
it's professionally presented (which Jason's design is I believe).


--
Douglas Burchard, President
DouglasBurchard.com, Web Applications
15024 NE 66th Street
Redmond, WA  98052, USA

direct: (206) 227-8161
solutions@...
http://www.DouglasBurchard.com

I like the design, but the small versions don't read well. They'll 
probably need to be adapted so that the reversed text is either axed or 
legible. And maybe a crop-in on the snake's face for the small one, for 
the same reason.

This is pretty SOP for folks who do identity design (as you probably 
know). Although sameness is very important, in reality, to *look the 
same* is more important than *being the same*. Adaptation is often 
required for small sizes on low-resolution media (like displays) ... to 
typography in particular.

I like the concept, I'm comfortable with the "vibe" of it (am not 
concerned about corporate-ness) but I agree that the wrench isn't 
perhaps the most "web" thing. Of course, I don't have a better 
suggestion. :-)

Kudos, though, this is nice!

Jacob Hanson wrote:

>I made a couple of adjustments that add some of that without giving
>up the overall jovial nature. And I made some little 'powered by' images.
>  
>
>  
>

On Mon, 2004-03-29 at 11:49, Ian Bicking wrote:

> With-constant-vigilence-against-mixed-case'ly y'rs,
> Ian

Been spending much time with Tim Peters?  :)

--
Jason D. Hildebrand
jason@...

On Mon, 2004-03-29 at 16:06, Ian Bicking wrote:
> In that light, Webware needs not to show that it is corporate, but
> that
> it is alive and actively supported.  And even if there's not a huge
> amount of active development of the Webware core, it *is*
> consistently
> supported by the community on this list.  And an enlightened manager
> *will* still be concerned about the viability of the project.

I agree very much with this sentiment.  Well-articulated, Ian.

I really like the logo, too, and would be happy to go with it (good 
work, Jacob!), although I think Matt Feifarek makes some good suggestions
for improving the small versions.

--
Jason D. Hildebrand
jason@...

Hallo,
Matt Feifarek hat gesagt: // Matt Feifarek wrote:

> I like the concept, I'm comfortable with the "vibe" of it (am not 
> concerned about corporate-ness) but I agree that the wrench isn't 
> perhaps the most "web" thing. Of course, I don't have a better 
> suggestion. :-)

Hammer and sickle? ;)

ciao
-- 
 Frank Barknecht                               _ ______footils.org__

* Ian Bicking <ianb@...> [2004-03-29 17:07]:
> 
> We don't have any corporate backing or much of that appeal, but we can 
> still appeal to management types that have an appreciation for open 
> source software and the environment that entails.  That's my evaluation 
> of Webware's appeal strictly on a technical level.
> 
> In that light, Webware needs not to show that it is corporate, but that 
> it is alive and actively supported.  And even if there's not a huge 
> amount of active development of the Webware core, it *is* consistently 
> supported by the community on this list.  And an enlightened manager 
> *will* still be concerned about the viability of the project.

OK, well I agree more with Adam Shimali's longer recent post, in that
it is a quality collection of software and should aim higher than a
small group that wanders by, but that's actually secondary for now. 

***
I think everyone agrees the #1 need is an updated site - that line at
the top: "Last updated August 1, 2003" kind of makes all this
discussion near-irrelevant.
***

I know everyone's super-busy, but maybe some of the more seasoned
Webware users could prune some of the Wiki, or make a new page on
there pointing out what docs are still valid/applicable, or close to
it with some modifications? Like:

url1: this tutorial is accurate but the session info is old
url2: this transaction desc needs updating from .6 to .8
etc...

I've found many are frozen in different points in time and it's often
hard as a newcomer to figure out which docs are currently
applicable. This would hopefully enable others to help organize
content for an updated site(?)

Or if someone already has an attack plan for updating content, seems
like a good time to announce it!


-- 

___________________________
toddgrimason*todd@...

Over the next few weeks I actually have to write developer docs for a
set of applications that I have built on top of Webware.  I will, of
coarse, have to cover how Webware itself works and how one goes about
developing with it.  During this process I will undoubtably go through
the majority of existing Webware documentation and pair it down to
something pretty focused.

Once I have finished the documentation I will make the Webware specific
stuff available to the community and would be interested in using it as
a base for re-orginization of the docs.

I'll mention it again when I actually have something to show.

- Aaron Switzer



On Tue, 2004-03-30 at 10:30, Todd Grimason wrote:
> * Ian Bicking <ianb@...> [2004-03-29 17:07]:
> > 
> > We don't have any corporate backing or much of that appeal, but we can 
> > still appeal to management types that have an appreciation for open 
> > source software and the environment that entails.  That's my evaluation 
> > of Webware's appeal strictly on a technical level.
> > 
> > In that light, Webware needs not to show that it is corporate, but that 
> > it is alive and actively supported.  And even if there's not a huge 
> > amount of active development of the Webware core, it *is* consistently 
> > supported by the community on this list.  And an enlightened manager 
> > *will* still be concerned about the viability of the project.
> 
> OK, well I agree more with Adam Shimali's longer recent post, in that
> it is a quality collection of software and should aim higher than a
> small group that wanders by, but that's actually secondary for now. 
> 
> ***
> I think everyone agrees the #1 need is an updated site - that line at
> the top: "Last updated August 1, 2003" kind of makes all this
> discussion near-irrelevant.
> ***
> 
> I know everyone's super-busy, but maybe some of the more seasoned
> Webware users could prune some of the Wiki, or make a new page on
> there pointing out what docs are still valid/applicable, or close to
> it with some modifications? Like:
> 
> url1: this tutorial is accurate but the session info is old
> url2: this transaction desc needs updating from .6 to .8
> etc...
> 
> I've found many are frozen in different points in time and it's often
> hard as a newcomer to figure out which docs are currently
> applicable. This would hopefully enable others to help organize
> content for an updated site(?)
> 
> Or if someone already has an attack plan for updating content, seems
> like a good time to announce it!
-- 
Aaron Switzer
aaron@...
416-350-8376 ext. 235

Douglas Burchard wrote:

> Would something other than a wrench fit better? For example:

I like the wrench a lot, I think "tools" when I think of Webware.

>     <http://www.douglasburchard.com/demo-logo_01.jpg>

the feather ties it to Apache too much. while most of us probably use it 
with Apache, it is NOT an Apache extension.

> I know, there are several problems with this specific example, but 
> perhaps someone could offer a better metaphor...

Maybe a way to use the wrench, and something that says: "web". A globe, 
perhaps a wrench adjusting the globe? I'm NOT a designer, so I'm not 
going ot even try to make an image out of that idea!

More than a good website, more than a logo, what Webware needs to get 
some real exposure and use is a Book. I had a colleague who had worked 
with me on a Python based project, then set out to build a web app. 
After looking through the books on developing for the web with Python, 
and buying "Web Programming in Python", he was thoroughly confused about 
what combinations of tools to use (CGI, mod_python, mod_snake, jonpy, 
quiote, webware, etc.). If there was a book: "Building web applications 
with Python and Webware", that told you what tools to use, and how to 
use them, he probably would have used it. Instead, he was faced with 
Zope, which he didn't really like. So, he bought a book that told him 
how to use PHP and MySQL to do what he wanted. He didn't like PHP 
better, what he liked was that there weren't so many options he had to 
choose from for every part of the project. A self contained book about 
Webware would really make the difference!

That being said, I have neither the talent nor the time to write such a 
book, but I think someone who did might even be able to earn a decent 
profit from it.

-Chris





-- 
Christopher Barker, Ph.D.
Oceanographer
                                     		
NOAA/OR&R/HAZMAT         (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker@...

Hmm. A book.

I'm wondering if I should/could put my hands up on that one.

Being that I'm a journo by trade, I'm thinking I might be able to pull it
off. Now that said, I'd need some serious ears when I had a question to
bounce off. That said, I'm a reasonable enough programmer, or I'd like to
think I am :)

Perhaps the trick is I write a sample chapter and throw it at a few
publishers (orielly etc) and see if theres any interest. If theres not, I
doubt I could justify the expenditure of time, but if there is interest
It'd be a damn great way to spend this currently-between-jobs period I'm
in.

Whats the vibes here?

--
Shayne O'Neill
http://perth.indymedia.org
I know how hard it is for you to put food on your family."
 ----George W. Bush

On Sun, 28 Mar 2004, Chris Barker wrote:

>
>
> Douglas Burchard wrote:
>
> > Would something other than a wrench fit better? For example:
>
> I like the wrench a lot, I think "tools" when I think of Webware.
>
> >     <http://www.douglasburchard.com/demo-logo_01.jpg>
>
> the feather ties it to Apache too much. while most of us probably use it
> with Apache, it is NOT an Apache extension.
>
> > I know, there are several problems with this specific example, but
> > perhaps someone could offer a better metaphor...
>
> Maybe a way to use the wrench, and something that says: "web". A globe,
> perhaps a wrench adjusting the globe? I'm NOT a designer, so I'm not
> going ot even try to make an image out of that idea!
>
> More than a good website, more than a logo, what Webware needs to get
> some real exposure and use is a Book. I had a colleague who had worked
> with me on a Python based project, then set out to build a web app.
> After looking through the books on developing for the web with Python,
> and buying "Web Programming in Python", he was thoroughly confused about
> what combinations of tools to use (CGI, mod_python, mod_snake, jonpy,
> quiote, webware, etc.). If there was a book: "Building web applications
> with Python and Webware", that told you what tools to use, and how to
> use them, he probably would have used it. Instead, he was faced with
> Zope, which he didn't really like. So, he bought a book that told him
> how to use PHP and MySQL to do what he wanted. He didn't like PHP
> better, what he liked was that there weren't so many options he had to
> choose from for every part of the project. A self contained book about
> Webware would really make the difference!
>
> That being said, I have neither the talent nor the time to write such a
> book, but I think someone who did might even be able to earn a decent
> profit from it.
>
> -Chris
>
>
>
>
>
> --
> Christopher Barker, Ph.D.
> Oceanographer
>
> NOAA/OR&R/HAZMAT         (206) 526-6959   voice
> 7600 Sand Point Way NE   (206) 526-6329   fax
> Seattle, WA  98115       (206) 526-6317   main reception
>
> Chris.Barker@...
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> _______________________________________________
> Webware-discuss mailing list
> Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss
>

* Shayne ONeill <shayne@...> [2004-03-30 13:25]:
> 
> 
> Perhaps the trick is I write a sample chapter and throw it at a few
> publishers (orielly etc) and see if theres any interest. If theres not, I
> doubt I could justify the expenditure of time, but if there is interest
> It'd be a damn great way to spend this currently-between-jobs period I'm
> in.
> 
> Whats the vibes here?

My 2 cents would be that any publisher would be a bit hesitant until
the project's [Webware's] site and docs was more fleshed out and
organized. 

A useful comparison might be Mason, a perl toolkit (generally)
built on top of mod_perl. 

http://masonhq.com
(specifically: http://masonhq.com/docs/manual/ )

They had pretty decent documentation and guides before their book was
published by O'Reilly. (http://masonbook.com) 

But I'm certainly not a publisher and wouldn't stop ya!

-- 

___________________________
toddgrimason*todd@...

I would love to see a book - I've taught webware and written some 
tutorials and guides - but nothing recent.

I think a pure webware book would fail, and there have been a few less 
then interesting Web Development in Python books as well.

Something like Enterprise Integration in Python would be interesing - 
focus on webware but do python things like NT Service integration, 
Manage MS Exchange, XMLRPC, long running threads etc...

There are lots of things Python can do easily that are hard in Java, and 
there are more things that Webware can do well that PHP can't. (I like 
to position it as the elegance of J2EE w/ the development speed of PHP)

I think that a webware book should focus on tasks done well in Python 
and webware.  (Otherwise it  seen as a less capable Java server or 
overly complex PHP clone).

Thanks,
-Aaron Held

Shayne ONeill wrote:

>Hmm. A book.
>
>I'm wondering if I should/could put my hands up on that one.
>
>Being that I'm a journo by trade, I'm thinking I might be able to pull it
>off. Now that said, I'd need some serious ears when I had a question to
>bounce off. That said, I'm a reasonable enough programmer, or I'd like to
>think I am :)
>
>Perhaps the trick is I write a sample chapter and throw it at a few
>publishers (orielly etc) and see if theres any interest. If theres not, I
>doubt I could justify the expenditure of time, but if there is interest
>It'd be a damn great way to spend this currently-between-jobs period I'm
>in.
>
>Whats the vibes here?
>
>--
>Shayne O'Neill
>http://perth.indymedia.org
>I know how hard it is for you to put food on your family."
> ----George W. Bush
>
>On Sun, 28 Mar 2004, Chris Barker wrote:
>
>  
>
>>Douglas Burchard wrote:
>>
>>    
>>
>>>Would something other than a wrench fit better? For example:
>>>      
>>>
>>I like the wrench a lot, I think "tools" when I think of Webware.
>>
>>    
>>
>>>    <http://www.douglasburchard.com/demo-logo_01.jpg>
>>>      
>>>
>>the feather ties it to Apache too much. while most of us probably use it
>>with Apache, it is NOT an Apache extension.
>>
>>    
>>
>>>I know, there are several problems with this specific example, but
>>>perhaps someone could offer a better metaphor...
>>>      
>>>
>>Maybe a way to use the wrench, and something that says: "web". A globe,
>>perhaps a wrench adjusting the globe? I'm NOT a designer, so I'm not
>>going ot even try to make an image out of that idea!
>>
>>More than a good website, more than a logo, what Webware needs to get
>>some real exposure and use is a Book. I had a colleague who had worked
>>with me on a Python based project, then set out to build a web app.
>>After looking through the books on developing for the web with Python,
>>and buying "Web Programming in Python", he was thoroughly confused about
>>what combinations of tools to use (CGI, mod_python, mod_snake, jonpy,
>>quiote, webware, etc.). If there was a book: "Building web applications
>>with Python and Webware", that told you what tools to use, and how to
>>use them, he probably would have used it. Instead, he was faced with
>>Zope, which he didn't really like. So, he bought a book that told him
>>how to use PHP and MySQL to do what he wanted. He didn't like PHP
>>better, what he liked was that there weren't so many options he had to
>>choose from for every part of the project. A self contained book about
>>Webware would really make the difference!
>>
>>That being said, I have neither the talent nor the time to write such a
>>book, but I think someone who did might even be able to earn a decent
>>profit from it.
>>
>>-Chris
>>
>>
>>
>>
>>
>>--
>>Christopher Barker, Ph.D.
>>Oceanographer
>>
>>NOAA/OR&R/HAZMAT         (206) 526-6959   voice
>>7600 Sand Point Way NE   (206) 526-6329   fax
>>Seattle, WA  98115       (206) 526-6317   main reception
>>
>>Chris.Barker@...
>>
>>
>>
>>-------------------------------------------------------
>>This SF.Net email is sponsored by: IBM Linux Tutorials
>>Free Linux tutorial presented by Daniel Robbins, President and CEO of
>>GenToo technologies. Learn everything from fundamentals to system
>>administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
>>_______________________________________________
>>Webware-discuss mailing list
>>Webware-discuss@...
>>https://lists.sourceforge.net/lists/listinfo/webware-discuss
>>
>>    
>>
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by: IBM Linux Tutorials
>Free Linux tutorial presented by Daniel Robbins, President and CEO of
>GenToo technologies. Learn everything from fundamentals to system
>administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
>_______________________________________________
>Webware-discuss mailing list
>Webware-discuss@...
>https://lists.sourceforge.net/lists/listinfo/webware-discuss
>  
>


-- 
-Aaron
http://www.MetroNY.com/
If the car industry behaved like the computer industry over the last 30 years,
a Rolls-Royce would cost $5, get 300 miles per gallon, and blow up once a year 
killing all passengers inside.

Again. Its one of those things I could probably do well. I guess the
problem for me personally is I'd have to find a publisher for me to
justify the rather large time investment book writing takes.

I'd probably also have to get someone else to lookysee in on the NT stuff
as I'm usually a linux guy these days. That said, I do have a reasonable
degree in experience on Com type programming, alas not thru python.

--
Shayne O'Neill
http://perth.indymedia.org
I know how hard it is for you to put food on your family."
 ----George W. Bush

On Thu, 22 Apr 2004, Aaron Held wrote:

> I would love to see a book - I've taught webware and written some
> tutorials and guides - but nothing recent.
>
> I think a pure webware book would fail, and there have been a few less
> then interesting Web Development in Python books as well.
>
> Something like Enterprise Integration in Python would be interesing -
> focus on webware but do python things like NT Service integration,
> Manage MS Exchange, XMLRPC, long running threads etc...
>
> There are lots of things Python can do easily that are hard in Java, and
> there are more things that Webware can do well that PHP can't. (I like
> to position it as the elegance of J2EE w/ the development speed of PHP)
>
> I think that a webware book should focus on tasks done well in Python
> and webware.  (Otherwise it  seen as a less capable Java server or
> overly complex PHP clone).
>
> Thanks,
> -Aaron Held
>
> Shayne ONeill wrote:
>
> >Hmm. A book.
> >
> >I'm wondering if I should/could put my hands up on that one.
> >
> >Being that I'm a journo by trade, I'm thinking I might be able to pull it
> >off. Now that said, I'd need some serious ears when I had a question to
> >bounce off. That said, I'm a reasonable enough programmer, or I'd like to
> >think I am :)
> >
> >Perhaps the trick is I write a sample chapter and throw it at a few
> >publishers (orielly etc) and see if theres any interest. If theres not, I
> >doubt I could justify the expenditure of time, but if there is interest
> >It'd be a damn great way to spend this currently-between-jobs period I'm
> >in.
> >
> >Whats the vibes here?
> >
> >--
> >Shayne O'Neill
> >http://perth.indymedia.org
> >I know how hard it is for you to put food on your family."
> > ----George W. Bush
> >
> >On Sun, 28 Mar 2004, Chris Barker wrote:
> >
> >
> >
> >>Douglas Burchard wrote:
> >>
> >>
> >>
> >>>Would something other than a wrench fit better? For example:
> >>>
> >>>
> >>I like the wrench a lot, I think "tools" when I think of Webware.
> >>
> >>
> >>
> >>>    <http://www.douglasburchard.com/demo-logo_01.jpg>
> >>>
> >>>
> >>the feather ties it to Apache too much. while most of us probably use it
> >>with Apache, it is NOT an Apache extension.
> >>
> >>
> >>
> >>>I know, there are several problems with this specific example, but
> >>>perhaps someone could offer a better metaphor...
> >>>
> >>>
> >>Maybe a way to use the wrench, and something that says: "web". A globe,
> >>perhaps a wrench adjusting the globe? I'm NOT a designer, so I'm not
> >>going ot even try to make an image out of that idea!
> >>
> >>More than a good website, more than a logo, what Webware needs to get
> >>some real exposure and use is a Book. I had a colleague who had worked
> >>with me on a Python based project, then set out to build a web app.
> >>After looking through the books on developing for the web with Python,
> >>and buying "Web Programming in Python", he was thoroughly confused about
> >>what combinations of tools to use (CGI, mod_python, mod_snake, jonpy,
> >>quiote, webware, etc.). If there was a book: "Building web applications
> >>with Python and Webware", that told you what tools to use, and how to
> >>use them, he probably would have used it. Instead, he was faced with
> >>Zope, which he didn't really like. So, he bought a book that told him
> >>how to use PHP and MySQL to do what he wanted. He didn't like PHP
> >>better, what he liked was that there weren't so many options he had to
> >>choose from for every part of the project. A self contained book about
> >>Webware would really make the difference!
> >>
> >>That being said, I have neither the talent nor the time to write such a
> >>book, but I think someone who did might even be able to earn a decent
> >>profit from it.
> >>
> >>-Chris
> >>
> >>
> >>
> >>
> >>
> >>--
> >>Christopher Barker, Ph.D.
> >>Oceanographer
> >>
> >>NOAA/OR&R/HAZMAT         (206) 526-6959   voice
> >>7600 Sand Point Way NE   (206) 526-6329   fax
> >>Seattle, WA  98115       (206) 526-6317   main reception
> >>
> >>Chris.Barker@...
> >>
> >>
> >>
> >>-------------------------------------------------------
> >>This SF.Net email is sponsored by: IBM Linux Tutorials
> >>Free Linux tutorial presented by Daniel Robbins, President and CEO of
> >>GenToo technologies. Learn everything from fundamentals to system
> >>administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> >>_______________________________________________
> >>Webware-discuss mailing list
> >>Webware-discuss@...
> >>https://lists.sourceforge.net/lists/listinfo/webware-discuss
> >>
> >>
> >>
> >
> >
> >-------------------------------------------------------
> >This SF.Net email is sponsored by: IBM Linux Tutorials
> >Free Linux tutorial presented by Daniel Robbins, President and CEO of
> >GenToo technologies. Learn everything from fundamentals to system
> >administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> >_______________________________________________
> >Webware-discuss mailing list
> >Webware-discuss@...
> >https://lists.sourceforge.net/lists/listinfo/webware-discuss
> >
> >
>
>
> --
> -Aaron
> http://www.MetroNY.com/
> If the car industry behaved like the computer industry over the last 30 years,
> a Rolls-Royce would cost $5, get 300 miles per gallon, and blow up once a year
> killing all passengers inside.
>
>

Perhaps I could however whack together a short tutorial on the basics of
getting a simple cgi-like servlet happening and then a few ideas how to
use the more stateful/servlet-y properties of webware.

Erm. Should I use one of the wiki's to do this? Is the new one stable
enuff to trust it with a bunch of work?

--
Shayne O'Neill
http://perth.indymedia.org
I know how hard it is for you to put food on your family."
 ----George W. Bush

On Fri, 23 Apr 2004, Shayne ONeill wrote:

>
> Again. Its one of those things I could probably do well. I guess the
> problem for me personally is I'd have to find a publisher for me to
> justify the rather large time investment book writing takes.
>
> I'd probably also have to get someone else to lookysee in on the NT stuff
> as I'm usually a linux guy these days. That said, I do have a reasonable
> degree in experience on Com type programming, alas not thru python.
>
> --
> Shayne O'Neill
> http://perth.indymedia.org
> I know how hard it is for you to put food on your family."
>  ----George W. Bush
>
> On Thu, 22 Apr 2004, Aaron Held wrote:
>
> > I would love to see a book - I've taught webware and written some
> > tutorials and guides - but nothing recent.
> >
> > I think a pure webware book would fail, and there have been a few less
> > then interesting Web Development in Python books as well.
> >
> > Something like Enterprise Integration in Python would be interesing -
> > focus on webware but do python things like NT Service integration,
> > Manage MS Exchange, XMLRPC, long running threads etc...
> >
> > There are lots of things Python can do easily that are hard in Java, and
> > there are more things that Webware can do well that PHP can't. (I like
> > to position it as the elegance of J2EE w/ the development speed of PHP)
> >
> > I think that a webware book should focus on tasks done well in Python
> > and webware.  (Otherwise it  seen as a less capable Java server or
> > overly complex PHP clone).
> >
> > Thanks,
> > -Aaron Held
> >
> > Shayne ONeill wrote:
> >
> > >Hmm. A book.
> > >
> > >I'm wondering if I should/could put my hands up on that one.
> > >
> > >Being that I'm a journo by trade, I'm thinking I might be able to pull it
> > >off. Now that said, I'd need some serious ears when I had a question to
> > >bounce off. That said, I'm a reasonable enough programmer, or I'd like to
> > >think I am :)
> > >
> > >Perhaps the trick is I write a sample chapter and throw it at a few
> > >publishers (orielly etc) and see if theres any interest. If theres not, I
> > >doubt I could justify the expenditure of time, but if there is interest
> > >It'd be a damn great way to spend this currently-between-jobs period I'm
> > >in.
> > >
> > >Whats the vibes here?
> > >
> > >--
> > >Shayne O'Neill
> > >http://perth.indymedia.org
> > >I know how hard it is for you to put food on your family."
> > > ----George W. Bush
> > >
> > >On Sun, 28 Mar 2004, Chris Barker wrote:
> > >
> > >
> > >
> > >>Douglas Burchard wrote:
> > >>
> > >>
> > >>
> > >>>Would something other than a wrench fit better? For example:
> > >>>
> > >>>
> > >>I like the wrench a lot, I think "tools" when I think of Webware.
> > >>
> > >>
> > >>
> > >>>    <http://www.douglasburchard.com/demo-logo_01.jpg>
> > >>>
> > >>>
> > >>the feather ties it to Apache too much. while most of us probably use it
> > >>with Apache, it is NOT an Apache extension.
> > >>
> > >>
> > >>
> > >>>I know, there are several problems with this specific example, but
> > >>>perhaps someone could offer a better metaphor...
> > >>>
> > >>>
> > >>Maybe a way to use the wrench, and something that says: "web". A globe,
> > >>perhaps a wrench adjusting the globe? I'm NOT a designer, so I'm not
> > >>going ot even try to make an image out of that idea!
> > >>
> > >>More than a good website, more than a logo, what Webware needs to get
> > >>some real exposure and use is a Book. I had a colleague who had worked
> > >>with me on a Python based project, then set out to build a web app.
> > >>After looking through the books on developing for the web with Python,
> > >>and buying "Web Programming in Python", he was thoroughly confused about
> > >>what combinations of tools to use (CGI, mod_python, mod_snake, jonpy,
> > >>quiote, webware, etc.). If there was a book: "Building web applications
> > >>with Python and Webware", that told you what tools to use, and how to
> > >>use them, he probably would have used it. Instead, he was faced with
> > >>Zope, which he didn't really like. So, he bought a book that told him
> > >>how to use PHP and MySQL to do what he wanted. He didn't like PHP
> > >>better, what he liked was that there weren't so many options he had to
> > >>choose from for every part of the project. A self contained book about
> > >>Webware would really make the difference!
> > >>
> > >>That being said, I have neither the talent nor the time to write such a
> > >>book, but I think someone who did might even be able to earn a decent
> > >>profit from it.
> > >>
> > >>-Chris
> > >>
> > >>
> > >>
> > >>
> > >>
> > >>--
> > >>Christopher Barker, Ph.D.
> > >>Oceanographer
> > >>
> > >>NOAA/OR&R/HAZMAT         (206) 526-6959   voice
> > >>7600 Sand Point Way NE   (206) 526-6329   fax
> > >>Seattle, WA  98115       (206) 526-6317   main reception
> > >>
> > >>Chris.Barker@...
> > >>
> > >>
> > >>
> > >>-------------------------------------------------------
> > >>This SF.Net email is sponsored by: IBM Linux Tutorials
> > >>Free Linux tutorial presented by Daniel Robbins, President and CEO of
> > >>GenToo technologies. Learn everything from fundamentals to system
> > >>administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> > >>_______________________________________________
> > >>Webware-discuss mailing list
> > >>Webware-discuss@...
> > >>https://lists.sourceforge.net/lists/listinfo/webware-discuss
> > >>
> > >>
> > >>
> > >
> > >
> > >-------------------------------------------------------
> > >This SF.Net email is sponsored by: IBM Linux Tutorials
> > >Free Linux tutorial presented by Daniel Robbins, President and CEO of
> > >GenToo technologies. Learn everything from fundamentals to system
> > >administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click
> > >_______________________________________________
> > >Webware-discuss mailing list
> > >Webware-discuss@...
> > >https://lists.sourceforge.net/lists/listinfo/webware-discuss
> > >
> > >
> >
> >
> > --
> > -Aaron
> > http://www.MetroNY.com/
> > If the car industry behaved like the computer industry over the last 30 years,
> > a Rolls-Royce would cost $5, get 300 miles per gallon, and blow up once a year
> > killing all passengers inside.
> >
> >
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: The Robotic Monkeys at ThinkGeek
> For a limited time only, get FREE Ground shipping on all orders of $35
> or more. Hurry up and shop folks, this offer expires April 30th!
> http://www.thinkgeek.com/freeshipping/?cpg=12297
> _______________________________________________
> Webware-discuss mailing list
> Webware-discuss@...
> https://lists.sourceforge.net/lists/listinfo/webware-discuss
>

On Apr 22, 2004, at 1:03 PM, Aaron Held wrote:
>  I would love to see a book - I've taught webware and written some 
> tutorials and guides - but nothing recent.
>
>  I think a pure webware book would fail, and there have been a few 
> less then interesting Web Development in Python books as well.

I agree -- I don't think a Webware book would work.  In fact, I don't 
think Python is ready for any web programming books (except Zope -- but 
even those suck tremendously, so maybe Zope isn't ready either ;).

The books that have existed seem to be scattered and incomplete, or 
just cover one framework under the guise of being general.  There was 
one that covered Slither, I think... ever heard of Slither?  I hadn't.  
There's another book with a chapter on Webware/PSP -- I haven't had the 
heart to even read the final version.  I reviewed it, and the structure 
was horrible, with none of the depth where it was necessary, and 
pedantic coverage of useless details elsewhere -- when they returned a 
second draft with none of the changes I suggested, I gave up on them.

The reality is that Python web programming is a fucking mess.  It's 
pathetic.  There's no way to write a book, because there's no material 
that's appropriate book material.  I like Webware, but it's not going 
to take over the world, not even the modest Python world.  Neither will 
Zope, not even Zope 3... and if I thought it would, maybe I'd switch 
over happily.  Well, to Zope 3, Zope 2 is a fucking mess, and fixing it 
only makes it worse, and I know I'll never be happy there.  All the 
other frameworks are in the same shape as Webware -- they have some 
users, occasional spurts of activity.  Some have solid enough 
communities to have a viable future, but many do not.  None of them is 
compellingly great.  Yes, even Webware.

BTW, if anyone wants to talk about these larger issues, Web-SIG is 
probably an appropriate venue: http://www.python.org/sigs/web-sig -- 
feel free to post anything that comes to your mind, it's too damn quiet 
there.

--
Ian Bicking | ianb@... | http://blog.ianbicking.org

Actually the webware chapter in dietel was probably the only part of the
book that really failed imho. Only because it missed the good stuff and
concentrated on the psp. That said, theres probably rhyme and reason there
in that it was intending to show that such a beast does exist, although
I'm inclined to thing webwares psp support is less important now that
mod_python does the same thing (I think). The Dietel book was fantastic
otherwise. Lots and lots and lots of good stuff. On general python it was
well written, pretty much everything I've needed to find on python is in
there. And the examples are great.

Essentially I use it as a reference.

I wonder if its worth writing a "good" chapter on webware, and hawking it
out to the book venders for inclusion in whatever python tome they come
out with next?

--
Shayne O'Neill
http://perth.indymedia.org
I know how hard it is for you to put food on your family."
 ----George W. Bush

On Thu, 22 Apr 2004, Ian Bicking wrote:

> On Apr 22, 2004, at 1:03 PM, Aaron Held wrote:
> >  I would love to see a book - I've taught webware and written some
> > tutorials and guides - but nothing recent.
> >
> >  I think a pure webware book would fail, and there have been a few
> > less then interesting Web Development in Python books as well.
>
> I agree -- I don't think a Webware book would work.  In fact, I don't
> think Python is ready for any web programming books (except Zope -- but
> even those suck tremendously, so maybe Zope isn't ready either ;).
>
> The books that have existed seem to be scattered and incomplete, or
> just cover one framework under the guise of being general.  There was
> one that covered Slither, I think... ever heard of Slither?  I hadn't.
> There's another book with a chapter on Webware/PSP -- I haven't had the
> heart to even read the final version.  I reviewed it, and the structure
> was horrible, with none of the depth where it was necessary, and
> pedantic coverage of useless details elsewhere -- when they returned a
> second draft with none of the changes I suggested, I gave up on them.
>
> The reality is that Python web programming is a fucking mess.  It's
> pathetic.  There's no way to write a book, because there's no material
> that's appropriate book material.  I like Webware, but it's not going
> to take over the world, not even the modest Python world.  Neither will
> Zope, not even Zope 3... and if I thought it would, maybe I'd switch
> over happily.  Well, to Zope 3, Zope 2 is a fucking mess, and fixing it
> only makes it worse, and I know I'll never be happy there.  All the
> other frameworks are in the same shape as Webware -- they have some
> users, occasional spurts of activity.  Some have solid enough
> communities to have a viable future, but many do not.  None of them is
> compellingly great.  Yes, even Webware.
>
> BTW, if anyone wants to talk about these larger issues, Web-SIG is
> probably an appropriate venue: http://www.python.org/sigs/web-sig --
> feel free to post anything that comes to your mind, it's too damn quiet
> there.
>
> --
> Ian Bicking | ianb@... | http://blog.ianbicking.org
>
>

* Ian Bicking <ianb@...> [2004-04-22 23:26]:
> On Apr 22, 2004, at 1:03 PM, Aaron Held wrote:
> >  I would love to see a book - I've taught webware and written some 
> > tutorials and guides - but nothing recent.
> >
> >  I think a pure webware book would fail, and there have been a few 
> > less then interesting Web Development in Python books as well.
> 
> I agree -- I don't think a Webware book would work.  In fact, I don't 
> think Python is ready for any web programming books (except Zope -- but 
> even those suck tremendously, so maybe Zope isn't ready either ;).

I'd agree and say just a decent [up to date with CVS or 0.8] tutorial
would be the best balance and use of time. Perhaps implementing
whatever the common example is for Java servlets these days - the
Petstore? I think the petstore is actually the whole J2EE fruit salad.

Anyways, I think the most alluring and useful tutorial would be
targeted at servlet/jsp developers, as the servlet paradigm is
similar, as is the ability to run an appserver on its own, which often
the masses using PHP don't have. And with Jython, BSF, and noise
around Groovy the javafolks are likely to come across Python.

> The reality is that Python web programming is a fucking mess.  It's 
> pathetic.  There's no way to write a book, because there's no material 

It does seem so. Perhaps a chicken/egg situation. Or lack of a niche -
those who can will run servlet (or full J2EE) containers, others
mod_php. (And the whole MS world too.) How much middle ground is
there? The most visible opportunities to ever get much of an
improvement to me would be having it run near-transparently on top of
java servlet containers (or .NET? I don't know anything about it) or
have mod_python fork and grow the 'safe_mode' features mod_php does so
that ISPs could deploy it as easily and widely as PHP.

> BTW, if anyone wants to talk about these larger issues, Web-SIG is 
> probably an appropriate venue: http://www.python.org/sigs/web-sig -- 
> feel free to post anything that comes to your mind, it's too damn quiet 
> there.

I think exasperation hit all the participants in web-sig. Hands thrown
up. Without a top-down, BSD- or BDFL-like process driving this sort of
project, people seem to talk for a bit then just go and choose the one
of the 349870987 frameworks they like the best. Seems one group just
needs to make a kick-ass framework, push it hard, and build support,
Darwin (the guy not OS) style. Not an easy thing to do... but not
impossible!


-- 

___________________________
toddgrimason*todd@...

On Apr 22, 2004, at 9:50 PM, Shayne ONeill wrote:
> Erm. Should I use one of the wiki's to do this? Is the new one stable
> enuff to trust it with a bunch of work?

Sure -- it saves everything right away to plain text files, and the 
only exceptions I've seen lately are from bots making invalid requests 
(dumb bots).  I'd advise using external editor -- it's a better writing 
experience, and it's pretty careful to avoid losing work (e.g., if you 
save a file and quite the editor, and zopeedit.py gets an exception 
when you try to save the file, then it'll ask you if you want to reopen 
the file with your changes).

--
Ian Bicking | ianb@... | http://blog.ianbicking.org

Have you guys seen 'epoz' (now renamed to 'kupu')? I've been trying to
integrate it with the code i'm doing. Its quite an astonishing html editor
thats all javascript and looks and works great.

Caveat: ie5.5 + or moz. and not ie5.5 on mac.

However it really is the same user experience as working within word or
its ilk. Infact I've been told you can copy from word and paste into kupu
and it retains formatting via some sort of wierd deep magic.

I've integrated into my little tinpot wiki, tho I'm having awful problems
getting the ie layout right. Partly cos I dont have ie to play with! But
it makes things sure nice.

--
Shayne O'Neill
http://perth.indymedia.org
I know how hard it is for you to put food on your family."
 ----George W. Bush

On Fri, 23 Apr 2004, Ian Bicking wrote:

> On Apr 22, 2004, at 9:50 PM, Shayne ONeill wrote:
> > Erm. Should I use one of the wiki's to do this? Is the new one stable
> > enuff to trust it with a bunch of work?
>
> Sure -- it saves everything right away to plain text files, and the
> only exceptions I've seen lately are from bots making invalid requests
> (dumb bots).  I'd advise using external editor -- it's a better writing
> experience, and it's pretty careful to avoid losing work (e.g., if you
> save a file and quite the editor, and zopeedit.py gets an exception
> when you try to save the file, then it'll ask you if you want to reopen
> the file with your changes).
>
> --
> Ian Bicking | ianb@... | http://blog.ianbicking.org
>
>

Also of interest is htmlarea: 
http://www.interactivetools.com/products/htmlarea/

I think if I was going to do WYSIWYG on a new project, I'd use that.

On Apr 23, 2004, at 12:05 AM, Shayne ONeill wrote:
> Have you guys seen 'epoz' (now renamed to 'kupu')? I've been trying to
> integrate it with the code i'm doing. Its quite an astonishing html 
> editor
> thats all javascript and looks and works great.
>
> Caveat: ie5.5 + or moz. and not ie5.5 on mac.
>
> However it really is the same user experience as working within word or
> its ilk. Infact I've been told you can copy from word and paste into 
> kupu
> and it retains formatting via some sort of wierd deep magic.
>
> I've integrated into my little tinpot wiki, tho I'm having awful 
> problems
> getting the ie layout right. Partly cos I dont have ie to play with! 
> But
> it makes things sure nice.
>
> --
> Shayne O'Neill
> http://perth.indymedia.org
> I know how hard it is for you to put food on your family."
>  ----George W. Bush
>
> On Fri, 23 Apr 2004, Ian Bicking wrote:
>
>> On Apr 22, 2004, at 9:50 PM, Shayne ONeill wrote:
>>> Erm. Should I use one of the wiki's to do this? Is the new one stable
>>> enuff to trust it with a bunch of work?
>>
>> Sure -- it saves everything right away to plain text files, and the
>> only exceptions I've seen lately are from bots making invalid requests
>> (dumb bots).  I'd advise using external editor -- it's a better 
>> writing
>> experience, and it's pretty careful to avoid losing work (e.g., if you
>> save a file and quite the editor, and zopeedit.py gets an exception
>> when you try to save the file, then it'll ask you if you want to 
>> reopen
>> the file with your changes).

On Thu, 2004-04-22 at 23:56 -0400, Todd Grimason wrote:
> * Ian Bicking <ianb@...> [2004-04-22 23:26]:
> I'd agree and say just a decent [up to date with CVS or 0.8] tutorial
> would be the best balance and use of time. Perhaps implementing
> whatever the common example is for Java servlets these days - the
> Petstore? I think the petstore is actually the whole J2EE fruit salad.
> 
> Anyways, I think the most alluring and useful tutorial would be
> targeted at servlet/jsp developers, as the servlet paradigm is
> similar, as is the ability to run an appserver on its own, which often
> the masses using PHP don't have. And with Jython, BSF, and noise
> around Groovy the javafolks are likely to come across Python.

For me it was incredibly painful trying to figure out the "proper" way
to deliver a Webware based site.  A few nice examples of reasonably
complete sites would go a long way.  I remember trying to use the vorbis
site as an example and quickly abandoning it.  It almost seems because
Webware caters to a crowd that appreciates a simple and reasonably
low-level platform, everyone that uses it is content/capable of building
everything they need from scratch.  The only reason I stuck with it was
I was trying to make a "write it in anything want" job more interesting.

A glance at the Wiki gives you:

* A high level description of a transaction.
* General pointers on MVC style design.
* Setting up Apache.
* Tips on common setup errors.
* Ideas for applications.

Not terribly useful to uninitiated.  At best there is tentative
suggestions instead of best practices.  Since the extent of my web-devel
work was PHP and I was also new to Python, I would have killed for:

* A complete example of a non template (pure servlets) site.
* A complete example of a Cheetah and/or PSP and/or ZTPL site.
* When/Where/How to use Actions.
* Best practices for managing DB connections.
* Easier to grok SecurePage (maybe that's just me).
* Best practices for integrating FunFormKit or FormKit.

Perhaps people have apprehensions about advocating their method as *the
method*, but I think a forum where multiple examples are available could
alleviate that.

-- 
Thomas E Jenkins <mort@...>

On Friday 23 Apr 2004 07:06, Thomas E Jenkins wrote:

> Not terribly useful to uninitiated.  At best there is tentative
> suggestions instead of best practices.  Since the extent of my web-devel
> work was PHP and I was also new to Python, I would have killed for:
>
> * A complete example of a non template (pure servlets) site.
> * A complete example of a Cheetah and/or PSP and/or ZTPL site.
> * When/Where/How to use Actions.
> * Best practices for managing DB connections.
> * Easier to grok SecurePage (maybe that's just me).
> * Best practices for integrating FunFormKit or FormKit.
>
> Perhaps people have apprehensions about advocating their method as *the
> method*, but I think a forum where multiple examples are available could
> alleviate that.

Yes please. 

The w4py wiki code is really helpful in getting ideas & understanding how to 
do things.

The more the better.

David

Ian Bicking wrote:
> I agree -- I don't think a Webware book would work.  In fact, I don't 
> think Python is ready for any web programming books 

It's very ready..but what should that book look like? Personally, I 
really like Steve Holden's book, but it's at a low level. It has the 
stuff I need to understand to be able to someday work with the internals 
of a package like Webware, but if you want to just put a site together, 
the publishing world makes it look like ZOPE or straight CGI is your 
only answer.

> or just 
> cover one framework under the guise of being general.  There was one 
> that covered Slither, I think... ever heard of Slither?

I think Slither was pretty much written for the book, as was Holden's 
AWeFUL, although Holden was quite clear that AWeFUL was meant to be a 
teaching example, not a robust framework.

> The reality is that Python web programming is a fucking mess.  It's 
> pathetic.  There's no way to write a book, because there's no material 
> that's appropriate book material.

I think this is a chicken-egg thing. Books are used by the general 
public as a way to gage the viability of something. The fact that there 
are a lot of Python books helps a lot of folks see it as viable.

Also, I don't think one could just write a Webware book with Webware as 
it stands. Writing the book would help highlight a LOT of places that 
Webware needs to be cleaned-up, improved, added-to, etc. It doesn't have 
to a be a whole book, any one chapter in the Wiki will help.

>  I like Webware, but it's not going to 
> take over the world, not even the modest Python world.  

It doesn't need to take over the world. Frankly, I had a very hard time 
choosing between Quixote, CherryPy, Albatross, Webware, etc, etc. If Any 
of these had really comprehensive documentation (i.e., a book), I'd have 
gone with it.

> All the 
> other frameworks are in the same shape as Webware -- they have some 
> users, occasional spurts of activity.  Some have solid enough 
> communities to have a viable future, but many do not.  None of them is 
> compellingly great.  Yes, even Webware.

Right, it's like the GUI frameworks, only worse. And it hurts Python 
adoption a lot. Frankly, it seems to me that there are a Huge number of 
frameworks that were written with a "more lightweight than Zope" goal, 
and I didn't see a clear distinction. Unfortunately, I think there is 
too small a barrier to entry, so a lot of people decide to roll their own.

> BTW, if anyone wants to talk about these larger issues, Web-SIG is 
> probably an appropriate venue: http://www.python.org/sigs/web-sig 

That is a good idea. Personally, I have far to little to offer. I just 
want something simple, complete and documented that works.

I think I started this thread because I had a colleague who was 
considering using Python to build a web app. If I could have pointed him 
to a book (or a web site, I suppose, but a book gives added legitimacy) 
that he could read, and it would tell him how to do everything he needed 
to do, he would have used it. As it was, he found himself wading through 
all the various web app frameworks out there, and then looking at a PHP 
book, that told him just how do so what he needed (A fairly simple 
database-backed web app.)

I have a level of commitment to Python that got me this far, but I'd 
really like that book too!

Shayne ONeill wrote:

> I wonder if its worth writing a "good" chapter on webware, and hawking it
> out to the book venders for inclusion in whatever python tome they come
> out with next?

Well, that could help, but it might just contribute to the "There's more 
than one way to do it" state of web development with Python.

Todd Grimason wrote:
> Seems one group just
> needs to make a kick-ass framework, push it hard, and build support,
> Darwin (the guy not OS) style. Not an easy thing to do... but not
> impossible!

Exactly. It's hard without the profit motive...Open source projects tend 
to get to the point where they are useful to their developers 
easily...getting beyond that is tough. What's the motivation?

Anyway, the recent work on updated the web site, and getting new stuff 
in a nice Wiki is a great move in the right direction. If we really keep 
building the wiki, it could turn into a good basis for a book some day.

-Chris

-- 
Christopher Barker, Ph.D.
Oceanographer
                                     		
NOAA/OR&R/HAZMAT         (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker@...

Thomas E Jenkins <mort@...> writes:

> For me it was incredibly painful trying to figure out the "proper" way
> to deliver a Webware based site.  A few nice examples of reasonably
> complete sites would go a long way.

Agreed. Oracle have recently incorporated a thing called HTML DB into
their database - essentially it's a web-based development platform for
producing web applications which do database display/update
transactions. They are marketing it as a competitor to MS Access. I've
used it to develop a simple web-based application, and it's very nice,
but the developer UI is lousy, and minor cosmetic changes are far too
hard. I'd love to redevelop in Python, but can't work out how to
start.

What I'd like is one or more sample applications with:

* A decent look & feel to them (I'm lousy at design, and I much prefer
  to nick other people's :-)
* Tabular displays based on a database query, including paging where
  there are a lot of rows
* A "create record" page
* An update page, accessible from the table display by clicking on a
  row
* Some sort of search page
* Support for multiple tables (either a tabbed display, with one table
  per tab, or a sidebar menu with entries for each table - something
  like that)
* Username/password login

In my experience, *lots* of applications have basically this database
update structure. The success of MS Access and similar tools leads me
to think that I'm not alone in this experience. Having a Python/
Webware sample application which offers this structure would be a big
win.

Given that a significant advantage of Python/Webware is the way that
you can separate logic from look, and build apps up from components,
maybe better would be having sample components - tab-based UI,
sidebar-menu UI, database display, database update, login, etc, with
examples of hooking them together.

Actually, the look & feel side of things is probably the main point
here. It's not really that hard to hack together a table display page,
but it looks hacked together. The real selling point with HTML DB was
that we hacked together an application, but the default look & feel
*looked* slick.

Paul.
-- 
This signature intentionally left blank