1 2 3 4 5 Previous Next 73 Replies Latest reply: Nov 26, 2011 10:21 AM by Arkadiusz Kaczyński Go to original post RSS
  • 60. Re: Seam.Next discussion
    Arbi Sookazian Master

    Spring in a Nutshell is due out


    Seam in a Nutshell will never be published.


    I was a die-hard Seam advocate during the early Seam 2.x days.  Now, the managers say how can I find a Seam developer for this Seam project?  Maintenance nightmare.


    It's all JVM/Spring/Hibernate/RDBMS for now and for a while (just check the job listings).  Done deal.


    I do wish JBoss core dev team the best of luck, but personally I'm moving on to Spring 3...

  • 61. Re: Seam.Next discussion
    Robert Morse Apprentice

    Someone should write a Seam 2.x to Spring 3 migration tutorial.
    I've given up on ever seeing the Seam 2.x to Seam 3.x migration document.

  • 62. Re: Seam.Next discussion
    Arbi Sookazian Master

    Robert Morse wrote on Nov 11, 2011 18:47:


    Someone should write a Seam 2.x to Spring 3 migration tutorial.
    I've given up on ever seeing the Seam 2.x to Seam 3.x migration document.


    Is it even possible to easily port a 2.x app to 3.x?  The core is drastically different (DI is Weld now and uses proxies with totally different design, etc.)


    How bout 2.x to 3.x and then migrate to Ceylon?


    By then AI will be real and ubiquitous and Siri will finally answer my personal questions about her!

  • 63. Re: Seam.Next discussion
    Robert Morse Apprentice

    The document that has been promised for months (years?) was a migration manual.  In other words,  if you have this kind of widget in Seam 2.x, here's what it would look like in Seam 3.x.  I'm pretty much with you on giving up.  I'm stunned that whoever is responsible for marketing Seam still has a job.


    I'd be very interested in carrying on this discussion off of the forum. If you're interested, send me a message:  rmorse at predictiveLogic dot com.


  • 64. Re: Seam.Next discussion
    Shane Bryzak Master

    We have the following wiki page which Jozef has been working on:


    http://www.seamframework.org/Seam3/Seam2ToSeam3MigrationNotes


    As Arbi pointed out it's not a trivial exercise to port from Seam 2.x to 3.x, as Seam 2 was based on a proprietary component model and Seam 3 is based on CDI.  The proper question that people should be asking though is 'How do you port from Seam 2.x to Java EE6?'  We probably haven't done enough to enlighten our users in this area though, least of all by giving our wiki page a title of 'Seam 2 to Seam 3 Migration Notes'.  However this is what our users search for when they want to know how to upgrade their Seam 2 application.

  • 65. Re: Seam.Next discussion
    Khalil El Mazouari Newbie

    Hi,


    Do you know what is the intent and/or strategy behind seam 3 in terms of productivity?.
    At this time, why should one uses seam 3 instead of  vanilla java ee6 or Spring3...


    It's clear that many seam2 advocate will not migrate to seam 3 as it is now. On the other hand, it's also clear that these people will not keep using seam 2 forever. Means that all these users are potential candidate for Spring3, vanilla javaEE6 ... So, may I ask the following:
    what is the seam team strategy to keep these seam 2 users and advocates in the seam ecosystem? Are these people just collateral damage for seam 3?


    Regards,


    Khalil

  • 66. Re: Seam.Next discussion
    Joshua Davis Expert

    Do you know what is the intent and/or strategy behind seam 3 in terms of productivity?.
    At this time, why should one uses seam 3 instead of  vanilla java ee6


    That is (IMO), one of the good points about Seam 3: It's basically just a bunch of CDI component libraries that you can use to enhance your application, or just to make working with JSF2 and CDI easier.   I'm not sure why, but it looks like a lot of folks are expecting it to be some big, monolithic framework.   Did this kind of thing happen to Hibernate when JPA came along?


    Anyway, in terms of productivity, I see Solder and Seam Faces as being essential.  I've found them very useful in the JEE6 projects I've done so far.

  • 67. Re: Seam.Next discussion
    Joshua Davis Expert

    Arbi Sookazian wrote on Nov 11, 2011 17:14:

    I was a die-hard Seam advocate during the early Seam 2.x days.  Now, the managers say how can I find a Seam developer for this Seam project?  Maintenance nightmare.

    It's all JVM/Spring/Hibernate/RDBMS for now and for a while (just check the job listings).  Done deal.


    This is exactly the kind of thing that makes me want to go build my next thing in some strange, but cool framework like Scala/Lift. ;)


    There is a definite tendency for managers to have the perception that having a big developer pool with direct experience in very specific technologies is a good thing.  However, IMO this is a bad way to select technologies, and a terrible way to build a team (just my opinion).  IMO good developers can learn anything, and fast.   The ones that want to learn TheMostPopularFramework are a commodity.  The ones that have interest in new things are more likely to be great engineers.  So, the big developer pool is a good thing if you want a team of mediocre developers.


    Arbi - Have fun with your non-JEE6, Spring Kool Aid, buddy! ;)

  • 68. Re: Seam.Next discussion
    Esteve Aviles Newbie

    Hi,
    Is anybody at Seam dev team going to put some comments and conclusions anywhere about this topic they started?


    What is going on at JBoss about Seam? People just want to know.


    I will give support to Seam 3, because it is a very good strategy, but we need to know how it will be executed and what we can do to help.


    Thanks

  • 69. Re: Seam.Next discussion
    Jason Porter Master

    We're almost at the point we can discuss what we have planned for Seam. We're making sure all of our i's are dotted and t's are crossed before we send out the message. Sorry for the delay.

  • 70. Re: Seam.Next discussion
    Brian Abston Newbie

    I don't think Seam has the support and resources from JBoss. This was confirmed for me when I took the Seam training class offered by JBoss. It was a joke. The documentation had so many errors and most of the examples didn't even work. It was so bad the instructor spent most of his time apologizing. The only saving grace for me and my guys was the Dan Allen book. We have a very large project coming up and we are trying to decide if we are going to stick with Seam. We will be watching closely to see how things go. I truly hope the team pulls it off and Seam 3 is a huge success.

  • 71. Re: Seam.Next discussion
    Tony Herstell Master

    Seam 2 was truly Awesome!


    Download Seam 2 and jump into code in the first few pages.
    hello world tutorial! and keep going...
    All of a sudden you were away; throwing in conversations and learning why they are so important (not going back to the database!).
    Want to do something smart go to the right chapter, drop in the relevant jars and wow.. it just works.


    Seam 2 was such a good eco-system, and so easy, that a complete idiot, like me, could use Seam 2 and not have ANY idea where the dividing lines were between JPA, JSF and all the other myriad of technologies were; and more importantly not have to care! (for a while)... I have to admit, in a few cases I still am not entirely clear! ;)


    Now we move onto Seam 3 or is it Weld or is it CDI or is it Solder... ALL the layers are back!
    Humm... Ok, well lets download Seam 3; perhaps its like Seam 2 was... OH NO! now we have to contend with Maven and the documentation looks like its written by framework writers for framework writers. Perhaps the examples... OH NO... Annotation HELL (mostly not commented)!


    Ok, too hard.... EXIT!


    Please bring back ECO-SYSTEM.
    Please bring back just drop in these jars to your application and add this nugget to get world-domination to work OR use this maven stuff... if you know what you are doing in Maven...
    PLEASE review the first few pages of Seam 2 doco and replicate!

  • 72. Re: Seam.Next discussion
    Trevor Sweeney Newbie
    Wow great gripe list... this thread is more about how to make Seam 3 better, and i don't think loose ranting will do that.

    We have converted our app which handles ~3 million hits per day from seam 2 to seam 3(CDI) and while it was painful i'm very glad we did it.

    Seam 2 was great and easy to use, however IMHO it was sloppy.  For example the idea of contexts were very loose and often a bean would live in the wrong context and get promoted as the framework saw fit. In the wonderful world of CDI its up to the developer to understand what they are doing and how to achieve it.

    Make no mistake, seam 3 is an infant, however if you want your web app to grow with the framework I suggest making the move and being on an active upgrade path.

    As for seam.next features I would love to see navigation handled a bit more gracefully.  If you are using pretty faces as we are your url and navigation live in two seperate locations, while they are very closly coupled and should remain coupled.

    Definatly a focus on excel exporting, pdf generation are all great.

    I would love to see a caching module.  We all know how greedy jsf is and each call of the same bean method in the same request with the same parameters may result in a performance hit.  We have a workaround for it to watch whats being called and cache the results to prevent these hits.  However the community at large may enjoy some of the same benefits.

    Thats my 2 cents

  • 73. Re: Seam.Next discussion
    Arkadiusz Kaczyński Newbie

    Yes, this thread contains a lot of useful tips for seam team, from developers (feedback) - especially it describes why seam2 was soo big success, and in which way seam3 should go.


    My summarize:


    - we really need good documentation and examples (knowledge and style)! - it is 'high priority'
    - seam 3 should be build on top of core part and additional modules (very important for me is fact, that core should contains not only cdi/weld and one or two modules - it should be one, consistent framework - full stack of futures and components, that are being mostly choosen for building JEE app)
    - please, do not close your minds only for RichFaces - Primefaces or IceFaces are big part of JEE world too
    - modules could have their own release cycle, but in that case 'stack' seam releases should be published more often than one or two per year
    - seam community (not developers) is not active as much as it should be



    I am not experienced developer, especially in java web apps, but i was thinking about seam 3 as 'future'; now I am a bit confused and lost. For me, seam3 in current state is not a better choice than Spring or 'plain' JEE6


    Regards,


    Arek


    P.S Seam3 can the bright star of Java web apps, and I appreciate your effort, but it need some additional work to be better and more usable (developer-friendly)

1 2 3 4 5 Previous Next