Snakes and Rubies?

D

Daniel Berger

James said:
This may be of interest, especially if you live in or near the Windy Ci= ty.
=20
http://snakesandrubies.com/event
=20
quote:
On December 3rd, Ruby and Python developers will gather at DePaul=20
University in Chicago to hear two of the leaders in the Web 2.0 movemen= t=20
debate the merits of each other=92s frameworks. Adrian Holovaty, the=20
creator of the Django framework for Python, and David Heinemeier=20
Hansson, the creator of Ruby on Rails framework will answer questions=20
about their work and the future of Web application development.
=20
Hand picked questions submitted below will be selected by a moderator=20
and presented to Adrian and David at the meeting.
=20
:etouq
=20
=20
James

I think it's time to start making the "My web framework can beat up your =
web=20
framework" t-shirts.

Dan
 
J

John W. Long

I think it's time to start making the "My web framework can beat up
your web framework" t-shirts.

Hopefully not. How about "the future is not Java" t-shirts. We would=20
like this to be a bi-partisan event. :)
 
R

Robert Klemme

Daniel said:
I think it's time to start making the "My web framework can beat up
your web framework" t-shirts.

<soapbox>Sometimes I get the impression that people are more busy
*creating* web frameworks than actually *using* them. One blink of the
eye and ten new web frameworks sprang into existence...</soapbox>

robert
 
M

Martin DeMello

Robert Klemme said:
<soapbox>Sometimes I get the impression that people are more busy
*creating* web frameworks than actually *using* them. One blink of the
eye and ten new web frameworks sprang into existence...</soapbox>

Maybe the web frameworks can use the dependency injection frameworks

martin

(well, someone has to!)
 
A

Adrian Holovaty

Robert said:
<soapbox>Sometimes I get the impression that people are more busy
*creating* web frameworks than actually *using* them. One blink of the
eye and ten new web frameworks sprang into existence...</soapbox>

Actually, Django was extracted from Real-World Web sites two years ago
(lawrence.com, ljworld.com, et al). In its development, it's
consistently been designed to solve immediate real-world problems.

That's what makes it so good. It's intensely focused on getting things
done.

Adrian
(Django developer)
 
J

James Britt

Adrian said:
Actually, Django was extracted from Real-World Web sites two years ago
(lawrence.com, ljworld.com, et al). In its development, it's
consistently been designed to solve immediate real-world problems.


Same goes for Nitro, I believe.

Yes, people often create alternative tools simply as an academic
exercise, but many folks have good, clear reasons why available choices
are inadequate and need to build their own. And it's great that they are
made available, and that people can then pick a tool best suited for them.



James

--

http://www.ruby-doc.org - Ruby Help & Documentation
http://www.artima.com/rubycs/ - Ruby Code & Style: Writers wanted
http://www.rubystuff.com - The Ruby Store for Ruby Stuff
http://www.jamesbritt.com - Playing with Better Toys
http://www.30secondrule.com - Building Better Tools
 
M

Michael Campbell

Martin DeMello said:
Maybe the web frameworks can use the dependency injection frameworks

This is happening in big ways in Java land, right now (WebWork is
using Spring, Hibernate is using Hivemind...)
 
J

J. Merrill

james_b said:
Adrian Holovaty wrote:
Yes, people often create alternative tools simply as an academic
exercise, but many folks have good, clear reasons why available choices
are inadequate and need to build their own. And it's great that they are
made available, and that people can then pick a tool best suited for
them.

Who's writing the hopefully-unbiased "how to choose a (Ruby | Python |
Java | .Net) web framework" document (wiki?), with a description of the
guiding principles, ease of getting started, licensing cost,
documentation quality and availability, (perceived) advantages and
disadvantages, extensibility info (without modifying source), speed of
development and deployment, speed of resulting site, etc etc?

I've seen a .Net-based system that has well over 1mb of source code,
most of which is generated automatically; the framework being used (by a
consulting firm) seems quite capable but very much non-trivial to grok
and not quick to build with etc. I'm not sure that Rails would achieve
the "less total source than the XML config used by other frameworks"
goal, but it sure would be smaller.

Ideally, someone would come up with an allegedly-typical site design and
encourage people who like the respective frameworks to provide their
sample implementations. The databases that sit behind might well be
different (e.g. a Rails developer might use the "table with multiple
related classes" notion but other frameworks would likely use multiple
tables for the same thing). Of course, that could give an advantage to
some particular framework -- maybe the "sample site" from each
framework's equivalent of the "Ruby on Rails" book could be implemented
using other frameworks. Or maybe the site I've watched get built with
that .Net framework could be built by the various open-source
frameworks...

(Would anyone pay to subscribe to an "Open Source [or not!] web
frameworks comparison" living document / site? Something has to bring
in the $$$...)
 
K

Kirk Haines

Same goes for Nitro, I believe.

Each of Rails, Nitro, and IOWA are definitely used in real world applications,
and at least with Rails and IOWA, the development is driven largely by the
practical needs. I'm less in touch with Nitro's development process, but was
told the other day that there are a number of production applications and 10
or so active users. Certainly not a large crowd, but definitely enough to
provide a good feedback cycle to shape the framework's practical development.

The same goes for IOWA. I have around 60 of sites and/or applications in
production with it myself, and have a similarly small but active group of
users providing me with regular feedback and ideas for areas that need
focused attention and development.

Some of this work, at least for Ruby, in the past, may have been theoretical,
and I know that at least part of the incentive behind Wee was simply to
explore other ways to do things, but now there are actual users behind Wee,
too, each of these frameworks continues to exist and be used because they
offer a practical solution to real world problems that people are faced with,
and in turn those real world problems help drive their continued development.


Kirk Haines
 
M

Michael Campbell

Michael Campbell said:
This is happening in big ways in Java land, right now (WebWork is
using Spring, Hibernate is using Hivemind...)

Great googly moogly, I meant TAPESTRY is using Hivemind! (Apologies
for replying to self...)
 
G

George Moschovitis

practical needs. I'm less in touch with Nitro's development process, but=
was
told the other day that there are a number of production applications and= 10
or so active users. Certainly not a large crowd, but definitely enough t= o
provide a good feedback cycle to shape the framework's practical developm=
ent.

Many ideas for Nitro were extracted while working on a number of
ruby-powered websites over the last 3 years. One of these sites,
joy.gr is quite probably one of the biggest real world Ruby
applications:

~ 20.000.0000 page views / month
~ 4.000 new registered users / month
~ 200.000 unique visitors (IPs) / month
~ 1000-3000 concurrent online users

Running from a single, old server, using an old, unoptimized ruby
powered web engine. All this before the first version of Rails was
even released.

We have experienced many problems from our old engine, and during the
last 3 years we have had the chance to carefully study the practical
issues and work out effective solutions. The fruit of our efforts is
Nitro. Restarted from scratch about one year ago, influenced by Rails
and other projects (Wee, CherryPy, etc..) it now provides everything
you need to build professional Web applications in the most natural
way.


regards,
George.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,744
Messages
2,569,484
Members
44,903
Latest member
orderPeak8CBDGummies

Latest Threads

Top