Jeah .. have to admit, that the number of acronyms might have been great, but for me it is more important how hard it is to fully understand the API :-)
________________________________________ Von: Derek Hohls [dho...@csir.co.za] Gesendet: Freitag, 24. April 2009 15:52 An: users@cocoon.apache.org Betreff: AW: Re: XSLT is Dead ?! Chris I had to smile - here you say: "the Maven build-process, the maven-plugins needed and the changes in the project-structures gave me the impression of relearning the entire thing from scratch." while in the next breath you talk about the "simplicity of a Spring+(Flex+BlazeDS)/CXF+Aegis application " - which has more API (acronyms - per - inch) than I could shake a stick at! Guess I'm just not keeping up anymore [shakes head] and I'm not (yet) convinced that its time to either move to Cocoon 2.2/3 or leave Cocoon altogether... Derek >>> On 2009/04/24 at 03:27, in message <202a1a1e28af7c42be78cc28cf94a27501f6a92...@cwareserver1.intern.c-ware.de>, Christofer Dutz <christofer.d...@c-ware.de> wrote: Hi, I have to confirm the first part of your post. I have been using Cocoon since 1.0.8 (I think) ... those were the times with processing-instructions, plain-xsp and no sitemap whatsoever and have used it in a number of projects very successfully. Currently I am considering using other frameworks for my newer projects though. 1. Starting a large and complex Cocoon project without a local Cocoon-Guru is very dangerous. 2. There are so many components involved in serving a response, that it's sometimes impossible to track errors back to the cause. 3. Even if I may be able to find these problems, it turns out that in larger projects I'm the only one permanently sorting out the problems, because I'm the only one able to find them (There is a really big difference between being able to solve a task using Cocoon and to understanding how Cocoon thinks) ... so I end up solving problems while all the others do the cool and fun stuff. 4. It is almost impossible to find employees that are willing and able to withstand the increasingly steep learning curve. I end up assisting them more time than they save me. So from a financial point of view this is not very good. 5. The step from Cocoon 2.x to Cocoon 2.2 was at least as challenging as the step from 1.x to 2.x (from my point of view). Even if you have a full understanding of Cocoons internals, the Maven build-process, the maven-plugins needed and the changes in the project-structures gave me the impression of relearning the entire thing from scratch. I had to try setting up a Cocoon 2.2 project 4-5 times until I finally managed to understand what I was doing (Most of the examples available only tell you what to do and not why). I have to admit that I have done Projects with Cocoon on my own, which I probably wouldn't have ben able to achieve using 4-5 Employees in the same time. I have done Projects with 4-5 Employees and ended up in supporting and coaching them with almost the same result als if I would have done it alone. On the bottom line, for me Cocoon is great for doing middle-sized projects on my own, because I am very familiar with the whole thing. As soon as the Project size exceeds my own working capacity and I need to do the Project with several people, I won't use Cocoon any more. Perhaps someday there will be a good Book on Cocoon 2.2 as the one done by Stephan Niedermeier which would enable me to tell my employees "Here .... Read this and let me work" ;-) ... I'd come back to Cocoon. But I have to admit that the simplicity of a Spring+(Flex+BlazeDS)/CXF+Aegis application is far more tempting for me at the moment because of its simplicity. Chris ________________________________________ Von: Merico Raffaele [raffaele.mer...@less.ch] Gesendet: Freitag, 24. April 2009 14:17 An: users@cocoon.apache.org Betreff: AW: Re: XSLT is Dead ?! Dear Cocoon Community First of all I have to say that I am a Cocoon lover. We are working with this framework since 2005 and we have developed many different types of applications that include: - integration of a legacy system with WebServices - complex e-commerce solutions - graphical rendering of data with fins - or just simple web sites Our experience is that Cocoon is a stable and scalable framework. Furthermore it does a great job concerning the separation of concerns. Now I come the point. What I did not like on Cocoon was the way XSLT was used to render the final output. Therefore I developed XSLTg (an XML template engine) that centres the XML template by supporting full standard XSLT/XPATH 2.0. I took me some time to publish a documentation that is up to date. Now you can find it under http://www.xsltg.com. I am convinced that XSLTg solves many of the problems addressed in the article. Please have a look and let me know what you think about and if http://www.xsltg.com may be able to become a part of Cocoon. Many thanks in advance Raffaele PS: Be warned, my English is not really the best. Improvements are welcome, I will be happy to clarify any obscurities. -----Ursprüngliche Nachricht----- Von: Derek Hohls [mailto:dho...@csir.co.za] Gesendet: Freitag, 24. April 2009 09:14 An: users@cocoon.apache.org Betreff: [!! SPAM] Re: XSLT is Dead ?! Oh, and on the flip side, there's this article: http://www.onenaught.com/posts/8/xslt-in-server-side-web-frameworks (but I would guess that is "preaching to the converted" on this mailing list!) >>> On 2009/04/24 at 09:03, in message <49f1805f.5ce9.00d...@csir.co.za>, "Derek Hohls" <dho...@csir.co.za> wrote: Carsten I had hoped comments like these would be added to the blog :) One other point, you say: "The attraction of Cocoon as a separate framework has decreased, but that's definitely not due to XSLT." Why do you say Cocoon's attractiveness is decreasing... should we all be looking around for a new framework to hop onto? (I'm genuinely curious here, not "trolling", because the older frameworks like JSP and Struts still seem to be going strong and so I'm wondering what it is about Cocoon that is making it :go out of fashion: ) Thanks Derek >>> On 2009/04/24 at 08:40, in message <49f15ece.5000...@apache.org>, Carsten Ziegeler <cziege...@apache.org> wrote: Derek Hohls wrote: > At least, according to this article: > > http://java.dzone.com/news/death-xslt-web-frameworks > > Maybe some of the developers, or other power users here, > would like to comment at this blog - I see Cocoon also gets > a "dig in the ribs" ... > Without commenting on this specific article, my only general comment is that you'll find articles for specific technologies/projects and you'll find as many articles against these (I guess the most famous topic in our area is Maven). Who's is wrong and who's right? Or more important: is there such an easy answer? I definitly doubt this. There isn't such a thing as the one programming language that rules the world or the one framework that makes everyone happy and is the golden hammer. Everyone is free to use what he thinks works best for him. Ok, coming back to the original topic :) Looking at the past 9 years where I've been using Cocoon and done a lot of projects with Cocoon and XSLT, I think it was a great tool by the time. And XSLT helped a lot in getting up to speed (once you managed the high entrance barrier to Cocoon itself). There are a lot of use cases still today for XSLT when it comes to create web sites. It really helps to separate the content from the layout. But in the end that's a matter how you design your application. I see a lot of people using other frameworks than Cocoon and pass the output from that framework to XSLT after the framework has rendered the content. So I don't think that XSLT itself is dead. The attraction of Cocoon as a separate framework has decreased, but that's definitly not due to XSLT. Carsten -- Carsten Ziegeler cziege...@apache.org -- This message is subject to the CSIR's copyright terms and conditions, e-mail legal notice, and implemented Open Document Format (ODF) standard. The full disclaimer details can be found at http://www.csir.co.za/disclaimer.html. This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. MailScanner thanks Transtec Computers for their support. --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@cocoon.apache.org For additional commands, e-mail: users-h...@cocoon.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@cocoon.apache.org For additional commands, e-mail: users-h...@cocoon.apache.org
