A spat has broken out between HP and BEA Systems over the future of HP's Netaction enterprise application server. BEA chief executive Alfred Chuang has reportedly told Wall Street analysts that, while Netaction would remain for vertical markets such as telecoms, HP has selected BEA's WebLogic for the mainstream market. But HP has strongly denied any such decision.
Read HP and BEA in Netaction spat
In contrast to BEA's comments, I have seen HP invest a lot into building their server and promoting it in and around the major sites (such as TheServerSide). Bob Bickel also mentioned some pretty impressive product plans to me at Java One. Thus, it seems doubtful that HP would simply begin selling Weblogic, as BEA mentions.
The marketing effort of HP's app server is impressive, I received already 2 CDs distributed with JDJ...I looked at it, but I have some kind of basic concern with it, since I don't get their model. So, they give away the product for free, still it is not open-source.
Probably having more clear plans or communication will do better, taking it beyond the fact that "it's free". The first thing one thinks about for a revenue model would be that they plan to sell services with the app server and this is how they make the money necessary to support further development. But then, is it well documented ? How is it about performance, usability, support etc.? Again, I think that just trying to get a user community with the "it's free" message is not enough.
I think their strategy is to make 'free' just enough to run a good standalone site with, but then if you want to go enterprise (clustering, etc) you need to buy some add-on modules.
Hmm, they don't simply say "its free".
They try to generate revenue with the so called "resilient edition" which has additional features like session failover, full-featured JMS, ...
Additionally they provide commercial products on top of it, the "HP IOE" like WebServices, mobile etc. infrastructures. And the services of course...
HP-AS is not open source because they cannot release the source due to licensing restrictions (I think this is mentioned in the HPAS review).
It is indeed well documented, usability is okay, support is great and it is quite fast.
Additionally you should expect a _lot_ within the next months from HPAS!
My suggestion: Try it, it is cool ;-)
IMHO, BEA must be really scared by this whole "commoditizing the app server" movement. Other vendors have other stuff (databases, applications, services, hardware) the app server is all BEA has.
Movements like HPs (possibly Suns) and the entire open source movement (especially JBoss 3.0) must send shivers down the spines of their accountants (and share holders).
What are the odds of BEA aquiring / merging / being aquired by a company providing either hardware, services, or both in the next two years?
"IMHO, BEA must be really scared by this whole "commoditizing the app server" movement. Other vendors have other stuff (databases, applications, services, hardware) the app server is all BEA has."
That's BEA's leverage in the market. Were they to take on services seriously, for example, their multiplier would drop like a rock, and then they couldn't buy companies that do things that they actually _need_.
and then they couldn't buy companies that do things that
> they actually _need_.
Hint, hint, Tangosol has such a nice caching stuff... ;-)
Andreas: "Hint, hint, Tangosol has such a nice caching stuff... ;-)"
Actually, we're saving up to buy BEA.
Actually, we're saving up to buy BEA.
WebLogic license or the company? ;-)
Cameron: "Actually, we're saving up to buy BEA."
Dimitri: "WebLogic license or the company? ;-)"
The company. We think they have some good products. With the current market conditions, they're a bargain.
BEA's still got a clear technical leadership in the J2EE app server space, IMHO. The optimizations and customizations available in WebLogic 7 are world class: they've embeddeded significant portions of TOPLink into their CMP engine, and their JMS implementation is finally looking peachy. They also have some great web services plumbing, and their clustering support has always been top notch.
And WebLogic Workshop really is a great app, even for experienced developers, imho. Hopefully this rumored "liquid data" thing coming down the pipeline is going to do for data integration what Workshop is doing with conversational web services. Probably is the next generation of their integration/transformation product with a slick interface (oh how we need more of those!)
Technical superiority counts for a lot of why BEA has traditionally been popular. But then again, IBM has always been a laggard, and they're catching up in share.
I think it would be an exaggeration to suggest that BEA is "shaking in its boots" over HP or jBoss. Both of those application servers have negligable marketshare, and it would require at best 2 or 3 years of compounded 100%+ growth for it to come near BEA's marketshare. Not even Linux has had that level of growth. And let's not also forget that it would eat into IBM WebSphere marketshare too. Open source has never been about marketshare, anyway.
HP probably has the potential to grow significantly, but they need to shed the poor reputation that Bluestone had in some camps. I've heard good things about HPAS 8 so far.
I think part of, "BEA's still got a clear technical leadership in the J2EE app server space," that needs emphasis is "J2EE". If requirements exist that can't be optimally satisfied in a J2EE container, you can't play in Weblogic's or WebSphere's playground. Their approach has always been that _everything_ must be a nail if I've got a hammer in my hand.
The architecture of HP-AS (and JBoss for that matter) is it's most appealing aspect to me. HP/Bluestone took a core services framework and implemented services on top of it that implement the J2EE APIs. Additionally, outsider's can build and deploy their own services into the service framework. This is a very powerful capability and seems to indicate a certain level of dignity on behalf of the app server. It's as if the app server is saying, "I've provided this standard approach to solving most problems, but I can't possibly have the answers to everyone's problems so I'll give you the ability to solve them as you see fit."
That being said, very few of these battles come down to technical superiority (e.g. Apple vs. MS). They come down to marketing and gradding marketshare first. As much as I like HP-AS, I'm skeptical about HP's ability to last in the app-server market.
Regarding Weblogic Workshop, I haven't looked at it, but I'm under the impression that it is only for building web services.
BTW, I was unfamiliar with Bluestone until HP bought them. What was wrong with their reputation?