I also agree. The learning curve is WAY too steep. I purchased the docs and still had a 5 week implementation.I don't consider what I was doing to be a complicated setup either. Also... I am a UNIX guy that has been configuring software (with success) and found this hard hard hard.
I actually write RPC servers and can understand this technology. I would hate to be someone without any development experience trying to do this.
I purchased docs are worth 10 dollars and that is about it. It appears that the person writing the documentation developed the code. Not everyone has his/her understanding and it is written that way. Way too complicated but... If you can somehow do it is it worth it (have to say something good)
I agree. I found the documentation more geared towards people who want to contribute to the codebase, rather than people who want to just learn its features and how it will help them solve their business problems. Who would recommend software for production use that is poorly documented?
Yup, this is really frustrating. I wonder from the thousands of downloads weekly that are claimed, how many can get any thing to work?
This is the my nth download of jboss; trying again to use it. I've used Weblogic 6.1, it has a nice admin console.
Here is my history: Downloaded the examples zip, set everything up, then compiled the jms examples using 'ant jms-compile'. Now what? No mention in the on-line docs what to do next. The jboss Admin and dev book (which covers only jboss 2.4.3 btw, doesn't help too much in this regard either).
I would suggest that the JBoss team include in the binary downloads simple "hello world" type j2ee applications. In fact, in addition to unit tests, it should include functional tests, perhaps using JFunc or Cactus.
Then the thousands of downloads will one day lead to thousands of actual deployments.
To everyone here in this thread who is complaining about the install procedure and test examples, I would invite you to take it apon yourself to create better intall and test examples and submit to JBoss. That's why its called Open Source. Everyone is invited and expected to contribute, not sit back and complain and expect someone else to do it.
How can they contribute examples if they can't even get anything running themselves? If you can't even get to the point where you get something useful yourself, how can you help others and how will you ever get to that point without help from those who are already at a more advanced level?
If you spend enough time on something, you can probably get to that point. But you'll loose a lot of people who would be willing to contribute but never will because they can never get traction with the product.
Well, I've gotten a lot of traction from JBoss and I started out with exactly what you are starting out with, namely what is here on this website and the downloads. Your options are: 1. continue working until you get traction, 2. pay someone to help you, 3. pay really big bucks to one of the "other" app servers. I can absolutely guarantee you that complaining is a huge waste of your energy and will get you no where fast. That's it I'm done with this tread. Time to go study the Jboss source code to figure it out. ;-)
That's a pretty naive point of view you have. Actually, there's a lot of utility in complaining. As a developer (not on this project) I have to say the feedback is a precious thing. Often negative feedback is the most useful.
"I can figure it out, why can't you" is a pretty crappy attitude that doesn't do much to promote wide acceptance of this software. Perhaps you had better luck because you fall within the target audience of the documentation - which many do not.
Just to add another voice to this thread. I'm a developer who's company's currently quite happy with using/selling another appserver. I'm keen to try & have them use/offer JBoss as an alternative but until I can get a basic system to compare with our current offering, I've only got my limited spare time to investigate & attempt to get to that stage.
If it was accepted, then I'd expect we'd be taking up some of the JBoss support options, so hopefully a gain to the JBoss Group too. On the other hand, some of the responses could be taken to suggest that JBoss is only suited to people who have large amounts of spare time to work out what it's doing!
I suspect that's not a particular 'target', but the senario mentioned in this thread might be worth considering in terms of a return-on-investment when looking at the available docs, etc...
The example was developed for JBoss2.4.* not for 3.0. That's why there are some small problems. If you are new to JBoss and J2EE, the best way to get things to work quickly is to hire a JBoss consultant for a few hours. This won't cost you much but help you get on the right track immediately. Remember, you time is also expensive to your employer.
my 2 cents.