Discussions

News: BEA announces WebLogic Platform ISV Edition

  1. BEA announces WebLogic Platform ISV Edition (26 messages)

    BEA has announced that they are going to offer a new edition of the WebLogic Platform: ISV edition. WebLogic Platform ISV Edition features the application server, portal and integration software, JRockit JVM, and Workshop. It is priced at $17,000 per CPU for a limited use license.
    BEA said this tailored edition of its WebLogic platform provides a pre-integrated application server platform bundle that can help ISVS save both time and money by avoiding assembling piece-parts of the WebLogic platform, before embedding or bundling BEA's software with their own application.

    BEA, which has seen revenue from new licenses of WebLogic shrink during the last two fiscal quarters, believes WebLogic Platform ISV Edition will help to boost its business, as customers upgrade to the full WebLogic Platform further down the line.
    BEA targeting ISVs with latest WebLogic package

    Threaded Messages (26)

  2. Gartner advice[ Go to top ]

    http://att.com.com/BEA+earnings+rise+amid+internal+turmoil/2100-1012_3-5307890.html

    from above:
    "research firm Gartner issued a note recommending that customers "delay new large-scale strategic commitments to BEA until the company's long-term strategy becomes clearer."

    and http://news.netcraft.com/archives/2003/04/10/java_servlet_engines.html

    Tomcat works fine you know. (oh, yeah, the list is not as good support as the "paid" support)
    .V
  3. Gartner advice[ Go to top ]

    Vic,
    Tomcat works fine you know. (oh, yeah, the list is not as good support as the "paid" support)
    The last time I checked, Tomcat was just a servlet engine.

    That's like saying "Why should I use Tomcat, when I can build a servlet container in Java?" The answer is that sometimes you should just use Java, but that doesn't obviate the value of other things (like Tomcat) that have already been built in Java to provide additional functionality.

    Similarly, a good number of traditional J2EE web apps could have been relatively easily built/deployed in Tomcat, but that doesn't obviate the value of a product like WebLogic platform that provides a large number of capabilities.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters
  4. Gartner advice[ Go to top ]

    Cameron said:
    The last time I checked, Tomcat was a servlet engine
    Since then, Tomcat has added JSP support, it provides J2EE declerative security, it provides JNDI Data Source pools, JSTL, JSF, a manger app, class reloading, and evertying else that is considered a popular good practice. Even Eclipse plug in. Anyone who used propriatery featers of a container... we'll that's the MS approach. You can see the link I gave above, where it shows that BEA is *NOT* in the top 3 or 4 deployed.

    People use Tomcat with Spring and Struts all the time now a days, and mostly people don't code servlets other than the ones the framework has. A war devloped on Tomcat can run on Resin, IBM or one of the other 12 containers. So from ISV, you develop a solution and how the war gets deployed is up to the client.

    .V
  5. who listens to Gartner advice?[ Go to top ]

    |
    |Tomcat has added JSP support
    |
    ...which is still a servlet...


    |
    |it provides JNDI Data Source
    |
    But try and find anyone who knows how to set one up...
    Possibly the worst JNDI implementation around. Completely ignores the existance of a JNDI ENC. All so they dont have to have a tomcat DD... :-(

    |
    |class reloading
    |
    And yet if I redeploy an app more than a few times, on tomcat, I get an OutOfMemory error and have to restart it... :-(
  6. who listens to Gartner advice?[ Go to top ]

    if I redeploy an app more than a few times, on tomcat, I get an OutOfMemory error and have to restart it... :-(
    I use Tomcat just fine with reloads (and Eclipse), and it is the most popular platform according to above link, so I think your flame is just FUD.
    Tomcat is very reliable in production.

    .V
  7. who listens to Gartner advice?[ Go to top ]

    |
    |so I think your flame is just FUD.
    |
    oh yeah, you are right. I am just making it up...
    I restart our tomcat 5.0 for fun...
  8. Congratulations[ Go to top ]

    Congratulations to the BEA marketing department. LiquidData article was great. Lots of words and absolutely no substance. Now the 'WebLogic Platform ISV Edition'. Woohoo the innovation keeps coming.
  9. Congratulations[ Go to top ]

    !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
    Chuang on Thursday announced that Wai Wong, who was general manager of Computer Associates' Unicenter management software, will become the head of product development at BEA
    !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

    Hey I have a business suggestion for Chuang

    He has to take BEA back to China
    Make all product with Made in China label
    it will sell good in US market. It is a proven business model for 100s of years.
    Also Chauang don't have to pay anything more or invest anything else for new release. Chineese can convert SUN,IBM or anyone else innovation to made in china label and ship to US.

    Man, BEA sucks which I know 3 years ago.
  10. next version[ Go to top ]

    with all the techies leaving BEA, I think BEA will be only left repackaging their existing produces in the near future... and ofcourse changing the price point to enter new markets... *yawn*

    Looks like BEA has lost its edge and it is going to be very difficult for it to get back those good old days syndrome...
  11. Congratulations[ Go to top ]

    I can certainly understand the perspective of some members here being critical of BEA as they were the market leader in application servers, which made them an instant target. But, the recent tone against them is somewhat harsh, so I would like to offer a more clinical approach.

    BEA is making some interesting moves. In their last earnings call, Alfred talked about creating value in the application server by creating industry-specific versions, in particular telecommunications which has more stringent SLA requirements, soft real time expectations, and other management necessities that could require bidirectional access in the JVM. BEA's IP in application server and the JVM makes them qualified to create this type of packaging for the market. The average deal size for telecommunications middleware is probably $1-10MM, so they don't need to have many transactions to be wildly successful in the high end.

    As for this announcement, I find it intriguing. For example, one of the things I've been eager to see is a Platform-equivalent packaging for the open source community. I know it'll happen eventually, but it could take years because of the sophistication and complexity of the challenge. BEA's WebLogic Platform contains an application server, portal server, integration (EAI + BPM + B2B), enterprise security, common installation, (eventually) common management, common IDE. While most projects don't need the entire platform, having it packaged together with consistent APIs, methodology, development approach and horizontal services that can be plug-and-play, is appealing and could simplify an organization's view on the middleware choices and evaluation that they have to go through. I believe that only BEA and Oracle offer the same degree of packaging, so there isn't an equivalent in the open source space and enterprises need to go to a larger software vendor to get this type of offering.

    In terms of this announcement, $7K / CPU for ISVs is very appealing. The engineering cost to build such a platform for BEA was well over a billion dollars, so a price point much higher than this could be justified. Howeverbusiness application companies or other larger ISVs can create a very compelling offering of their system with the entire BEA stack embedded at a price point lower than IBM or on-par with Oracle. This is certainly a move to appeal to the bottom line of BEA's partners.

    On the financial side, BEA is in a nice position. They have about 30% of their market capitalization in cash, so their book value is great and the downside in their stock price is limited. The ratio of services to licenses has slightly moved towards favoring services over the past two years, but BEA's number of license deals has remained flat and the number of large deals has increased.

    Overall, while BEA isn't necessarily the choice of the vocal majority here, they are still performing reasonably well in the market and seem to be crafting products that have a market interest among ISVs, large enterprises with mission critical systems, and companies with diverse middleware needs that want a more simplified approach to purchasing / managing a middleware stack.

    Tyler
    Former BEA Guy
  12. Congratulations[ Go to top ]

    "In terms of this announcement, $7K / CPU for ISVs is very appealing. The engineering cost to build such a platform for BEA was well over a billion dollars, so a price point much higher than this could be justified."

    This is either a typo or grossly misinformed. $1B? Not on your life.
  13. Congratulations[ Go to top ]

    The engineering cost to build such a platform for BEA was well over a billion dollars, so a price point much higher than this could be justified. "This is either a typo or grossly misinformed. $1B? Not on your life.
    Yeah, I can see how that statement could be interpreted as shock value, but it's actually not. There is a bit of foundation to that statement. In the trailing three years, BEA spent $390MM on R&D costs per their income statement. Given the trends that are there, BEA invested about $100M for the previous four years on WebLogic and Tuxedo derivatives. This gets us close to $800MM in post-WebLogic acquisition R&D spend. R&D costs don't include the support organization and any efforts that they may perform in bug tracking, bug resolutions, case management, etc. It also doesn't include the costs of partners and investments that BEA has made in an M&A basis in other organizations that are building technology BEA gets to license and embed. This also doesn't include costs for acquisitions of companies that have technology BEA has embedded.

    So, applying a reasonable factor, we can say that their costs to develop the entire platform over the past decade, is probably over $1B.

    Tyler
  14. follow-up[ Go to top ]

    Thanks Tyler for the inside scoop on BEA. We appreciate that inside story given you have been with BEA. It is going to take me a little time to consume this 1B$ thing you talk about. In my opinion BEA is loosing grip. Their sales numbers are dropping steadily. They are loosing steam not just in US, in Europe too. Way too many industry, analyst reports have also been very skeptical on where they are heading with all this. That is just my take...

    In any case it is good to have a view of both sides of the coin, so that other readers can make their judgements unbiased.

    Also, how popular are these price point entry editions news posts (on tss)? Are they really read or are they newsworthy to J2EE developers? Does any one care any more? It is natural that every company will price their product according to the market segment they want to funnel their product. How does it affect the core j2ee developer? Finally the decision to buy etc is made at a much higher level.
  15. bea starts to look like the Titanic[ Go to top ]

    I'm really disapointed to see such a good product (at least compared to websphere) being slowly killed by poor management and poor marketing teams.
    The technical guys are all leaving; it looks very bad when an once technically leading company is now only driven by marketing and manager departments : a driver is useless if the engine is gone.
    Really pityfull.
    Fortunately the OS containers (jboss, jonas) are now almost mature enough to really take off.

    jos.
  16. |
    |The technical guys are all leaving
    |

    That may be more to do with the fact that Google has a lot of cash and are hiring other good people.

    |
    | Fortunately the OS containers (jboss, jonas) are now almost mature
    | enough to really take off
    |

    Hmmm, not where I sit...
  17. bea starts to look like the Titanic[ Go to top ]

    A tech company that consistently makes money should never be compared to the Titanic.

    BEA is a large company. Some names well-known to TSS have left, but for every good BEA engineer who has decided to pursue Google millions, there are a dozen more who have been with BEA since the WebLogic days and are still innovating. Cedric would be the first to acknowledge this. I've been in the industry nearly 20 years now, starting in the heady days of Unix at Bell Labs, and BEA's current talent stacks up to any organization I have been with.

    Google is in a unique position in the industry right now, and I wish Cedric the best of luck there. But don't think a few good people out of thousands leaving BEA is tantamount to a cruiseliner hitting an iceberg.

    Thanks,
    Craig
  18. Congratulations[ Go to top ]

    Not to mention that it stated $17k/cpu and not $7k/cpu. Unbelievable. Have they not yet figured out that it really is not worth that much.
  19. innovation[ Go to top ]

    Hey Tyler,
      You might want to check out the 'innovation' rolling out of Microsoft. They are integrating all kinds of stuff for you like the os, browser, database, appserver. If you like that kind of 'innovative one stop shop', I suggest using Microsoft products.
  20. innovation[ Go to top ]

    Hey Tyler,  You might want to check out the 'innovation' rolling out of Microsoft. They are integrating all kinds of stuff for you like the os, browser, database, appserver. If you like that kind of 'innovative one stop shop', I suggest using Microsoft products.
    I love all of the technology that is coming out in the middleware space. I'm just a huge fan of middleware. I'm motivated by the movements I've seen at all of the vendors: Oracle, Microsoft, IBM, BEA, Sun -- and those from the open source arena. My comments above weren't an indication that BEA's solution is the perfect one, just that they have ear-marked a distinct portion of the market and have a pretty clear roadmap on the product they are delivering. Their technology has a unique design and meets the needs of many of their customers, but I don't think they intend for it to be a market dominator.

    Microsoft is making some really interesting moves, but their strategy hasn't fully gelled yet. However, when you mix together their client-side Longhorn strategy, with the capabilities of .NET and their stated desire to move towards annotation and repository-based application / middleware, Microsoft is also producing a potentially visionary product.

    I believe (and am hopeful) that there is a lot more innovation left for us to witness in this industry. Companies such as BEA, IBM, Oracle and Sun have done an excellent job of introducing the world to what is possible and now we are witnessing the transformation of the industry into its next stage.

    Tyler
  21. <navEndya>
    <title>Yes,I agree with craig.</title>
    <body>Bea is a large Company.One or two people leaving doesn't make any difference. I dont know why people take it so seriously.For Bea guys,I would like to say "Edi emaina Kaani manam munduku vellali"(Whatever Happens we should move ahead).BEA should move positively.I am seriously looking forward for Apache Beehive Project.Regarding Pricing $17,000\CPU.BEA customers are large organizations.Large Organization can afford these.
    I hope Large Organization dont jump to conclusion.:-)
    ->whats the vision towards GRID Computing
    ->support of JSF
    </body>
    </navEndya>
  22. Congratulations[ Go to top ]

    Could you explain what you mean by "bidirectional access in the JVM"?
  23. Congratulations[ Go to top ]

    Could you explain what you mean by "bidirectional access in the JVM"?
    Sure. I might not have picked the right set of words here. But, what I was trying to convey was that telecommunications companies that want to use application servers as a backbone for VOIP or other communications networks have soft real time requirements. These soft real time requirements can be throughput or response time requirements of packet transmission. Point to Point transmissions on a computer will happen socket-socket, but the next layer is JVM-JVM. So, in order to monitor, manage and meet SLAs for soft real time requirements, an application server may need interrupt access at the JVM layer and / or the JVM layer may need to interrupt operations within the application server.

    Now, I'm explaining this from a half-baked understanding from the description that I gathered from BEA's most recent earnings call and from the publicly non-NDA information available. I'm sure there are some inaccuracies in my description, but this is the basic gyst of the type of engineering modifications that might have to be made to build a telco application server.

    Tyler
  24. Congratulations[ Go to top ]

    So, in order to monitor, manage and meet SLAs for soft real time requirements, an application server may need interrupt access at the JVM layer and / or the JVM layer may need to interrupt operations within the application server.
    You seem to be describing prioritized preemptive schedulling, which I agree is missing in Sun's JVM specification.
  25. Voodoo economics[ Go to top ]

    In terms of this announcement, $7K / CPU for ISVs is very appealing. The engineering cost to build such a platform for BEA was well over a billion dollars, so a price point much higher than this could be justified.
    That's quite a funny logic I must say (apart from the fact that it's really $17K not $7K). What price is justified, depends on the supply and demand situation in the market, not on a comparison of overall development cost and unit price.
  26. The news release says $17,000 (not $7,000) for a limited use license. What is a "limited use license"? Anyway the BPM (Business Process Management) part of the offering alone is worth the license fee.

    Many, probably most, IT shops don't have the bandwidth to mess with moving targets in the open source community, especially for high risk bet-the-business applications. The focus is on their businesses, not on what's cool about the latest J2EE craze.

    Without BEA and Websphere the market for J2EE developers would be dramatically smaller because M$ would have it all. The open source community creates a richer gene pool of competing next generation software. Everyone benefits, including the commerical software firms. So it is a symbiotic relationship where the open source J2EE community depends on the vitality of the commerical software sector and vice versa. But the open source community is a destructive force when it takes on a gang mentality that tries to stifle competing products, whether open source or commercial software.
  27. Spot on Paul. I've been working with Weblogic since it was Tengah 3.x. This knowledge helped me gain employment at places where I later pioneered many open source products, including JBoss. In my current organization, we run both JBoss (for departmental applications) and Weblogic (for the mission critical application). IMO, there is plenty of room for both. In addition to the bandwidth constraint you describe, there is also a risk factor. For mission critical applications, organizations want guarantees and dedicated support. More importantly for BEA, these organizations are willing to pay for it (although $17k per CPU is a little shocking).