Sebastien Arbogast wrote:
Why is Orbeon "the first one"? FYI, Cocoon started in 1999. Orbeon
claims to be standards-compliant but it's not much more than Cocoon is,
and Cocoon's architecture brings much more potential to *integrate*
implementation of standards.
"The first one" was more a rhetoric formula : you always present the
alternative solutions first, and then the one you've chosen because
you think there is the best. It's a way to put things in perspective
to explain the motivation of this project : it's not only because I
worship Cocoon like others worship Firefox ;o), but because I've
compared it to other solutions on the market. And in every comparison
there are strengths and weaknesses on both sides.
You missed the rhetoric formula, and the text is (was, now) as it talks about opensource XML frameworks, and then "The first one is Orbeon", implying some precedence either in age or value.
Yeah, they claim XForms compliance. But how much of XForms? Not that
much more than what XMLForm was providing in Cocoon 2 years ago when we
decided to abandon it because a server-side XForm implementation is
either overly complex or too much limited.
I don't quite agree about that : I find their approach of XForms quite
natural in fact, and much more "developer-friendly" and structured
than Cocoon Forms one, even if I agree it is less powerful. And the
thing is the current trend seems to tend to a massive adoption of
XForms on client-side. So implementing XForms is an asset for them
because what you learn for OPS, you will be able to use it out of it
in future applications. Whereas for CForms...
Whereas for CForms you can do today things that are totally omitted by XForms because it's a client spec, such as connection and interaction with data outside the form. How can you e.g. validate in an XForm that an input is an existing purchase order ID in your database?
Now if XForms is successful in the client (Firefox support is a nice thing, but what about IE?) you can be sure that Cocoon will provide a powerful combo using the CForms infrastructure server-side connected to client-side XForms.
Also FWIW, you can use CForms without even knowing that you do. I have a few stylesheets that produce form definitions and templates from raw HTML pages containing a few additional attributes...
Also being a W3C member means nothing regarding the quality of a
company: you just have to pay [1] to be a W3C member.
</rant>
Yes but it's also a proof that you're deeply involved in the
standardization process and work to maximize the use of standards. Of
course you can do that without paying the fee to enter W3C, but if I'm
not wrong, you can't participate in the process, you can't make the
standards yourself, you can just "undergo".
FYI, at my company we had some contacts with W3C people and we recently received a mail from their PR team saying (tranlsated) : "This would be the appropriate time for your company to join the consortium, participate in the numerous W3C activities and share your experiences in the interested working groups. W3C also offers its members a lot of opportunities to promote their activities internationally".
What about respecting standards there? Nothing. Join the W3C because its good for your business (and good for W3C's also).
Don't get me wrong, I don't make propaganda for competition there. I
just try to be objective, to present advantages and drawbacks of both
solutions in a few lines, to explain why I think (as a user) that
Cocoon is a better solution, and why it deserves a few improvements
(and especially documentation improvements) to try to make it the
ideal solution. Have you ever seen OPS documentation : it's not very
rich because as I said OPS has less features than Cocoon, but it's
clear, coherent, well-designed and structured, and updated. That's
precisely why I came to OPS first even if I had heard of Cocoon, and
I'm sure this happens to others.
I totally agree with you on this point. OPS docs are clean, even I they left me frustrated because they leave a number of things at the tutorial level. But what they have is clearly better presented that what Cocoon has.
I don't think that we should be afraid of competition. That's what is
interesting with Open Source : you can learn from competition, compare
to it, and even take good parts out from competition to use them for
yourself. And let's not forget the "point-of-view" issue : you and I
are convinced, we know that Cocoon is the best Open Source solution on
the market for XML presentation and publication servers, but newcomers
don't. Look at this page :
http://www.orbeon.com/community/cocoon
I don't know exactly if all this information is verified, but I think
this type of comparison is a good point.
We looked at it, and some points in this comparison are clearly wrong.
Anyway, this was not the purpose of my article because it's not really
for newcomers. I just tried to put things in perspective, inside my
own user and developer experience, in order to explain my own initial
motivation on this project. BTW I replaced "the first one"... ;o)
Thanks. The new version makes it clear that you explain your own path through XML frameworks and that _you_ met Orbeon first. That wasn't what was implied by the previous version.
BTW, I met one of Orbeon's founders and he explained me they started OPS because they were frustrated by Cocoon 1. Now Cocoon people also were frustrated, and this led to Cocoon 2.
Sylvain
-- Sylvain Wallez Anyware Technologies http://apache.org/~sylvain http://anyware-tech.com Apache Software Foundation Member Research & Technology Director
