The serverside in bed with microsoft ????

Discussions

TSS feedback: The serverside in bed with microsoft ????

  1. The serverside in bed with microsoft ???? (28 messages)

    In a previous thread someone mentioned that The Serverside is owned by Precise software solutions who are a strategic partner of Microsoft.

    After a little research I have dug up the following:

    http://www.tamirfishman.com/download/precise_18082002.pdf

    which details precises acquisition of The middleware company.

    Precise's strategic partnerships are laid bare in http://www.precise.com/Partners/Strategic/

    The middleware company runs www.theserverside.com.

    Have Bill Gates tendrils found another clever way to infiltrate the J2EE community?

    Or am I just being completely paranoid?

    Threaded Messages (28)

  2. Tim,

       You originally posted this as a news item, but I moved it to the feedback forum - a more appropriate place to discuss this.

       I've commented on your question a few times in the mega benchmark thread:
    http://www2.theserverside.com/home/thread.jsp?thread_id=16149#62731
    http://www2.theserverside.com/home/thread.jsp?thread_id=16149#63045
    http://www2.theserverside.com/home/thread.jsp?thread_id=16149#63144
    http://www2.theserverside.com/home/thread.jsp?thread_id=16149#63367
    http://www2.theserverside.com/home/thread.jsp?thread_id=16149#63407

      In a nutshell - TSS had no part in this benchmark outside of simply doing a news post - which I would have done regardless of who (TMC or not) did the benchmark.

      TSS is an independent business unit at TMC, which is an independent business unit at Precise. I assure you that you can expect TSS to continue doing its job just as it has these couple of years. Infact - after the Precise acquisition, we are more independent now than before!

    Floyd
  3. Don't worry Floyd, I think everyone understands that you just did what you had to do. I mean, surely noone would ask of you to check your sources, or make certain that the news that is posted is accurate! What an outrage that would be! It would mean that instead of only reposting what you're fed in your daily inbox queue, you'd have to actually think about what is posted.

    So, don't worry Floyd. You're in the clear here.

    /Rickard
  4. Rickard, as usual, your attempts to provoke me are obvious, petty and immature.

    Floyd
  5. You forgot "true" ;-)
  6. Hi

    But it would be no mistake to answer the hidden question of Rick. How does your quality assurance look like. If i have to detect myself if the news on your site are just some commercials, there is no added value to google myself for news.

    Tobias
  7. Tobias, I can respond to your polite and well-founded question.

    It is always our goal to summarize things in a way that saves us much time for our readers as possible, so that they don't even have to follow the links for more info. For over 2 years of news posting, we've regularly done things like call up companies/originators of a news piece to get the inside scoop, and/or extra information that would be releveant to our members.

    We always try to distill the essence of a piece of news in a 5 line summary you can see on the homepage. We often download the product in question to make sure its for real, go through and read changelogs and other items like that.

    In contrast to what Rickard thinks, the TSS news posting is an intricate, involved, and timeconsuming. Its a process that works because we love what we do and feel responsible to the community to do it well. Even for the news posts that are submitted by others, a TSS editor usually rewrites the summary to make it informative, hype-free, and complete.

    I think we are very good at doing this, and I hope that our members (yourself and Rickard included) will be reasonable enough to not dismiss 2.5 years worth of examples of our editorial abilities - due to one news post in which we had to let our members decide for themselves what to think about the benchmark.

    Floyd
  8. Tobias, I can respond to your polite and well-founded

    > question.

    And note that Floyd here without breaking a sweat avoided the issue of quality control. All he mentioned was reducing quantity. I.e. it's like a big funnel where you put stuff in, and things come out, but there's no real checking of the things that pass through.

    If ever there is a question of whether a news posting is true or not, Floyd has the perfect scapegoat: "we had to let our members decide for themselves what to think about [it]."
    Thus ensuring that Floyd doesn't have to think about it.

    It's quite brilliant, but oh so transparent.

    /Rickard
  9. Rickard, I don't think you read my reply. I directly mentioned the quality control measures we take, but did not mention reducing quantity.

    Your reply says that I mentioned reducing quantity but not quality control. I am not sure what message you read to base your response on.

    Floyd
  10. Floyd,

    I'm not convinced that you posted the study in good faith. After all, one can find every sort of 'study', proffering all sorts of opinions on the web. The reason we as readers come to sites such as TSS, is to find "Quality" material, that has met certain standards.

    One could compare your role to that of a medical journal, presenting various studies on drugs and treatments. What an affront to the Journal's credibility it would be, if it turned out that a drug that got a very flattering review also happened to be manufactured by a close partner of the Journal.

    This is, I hope, a good analogy to your situation, and perhaps explains why I and many others are quite perturbed at your apparent lack of scruples and good judgment. At the very LEAST, you should have plainly stated, at the very beginning of the study, that TSS has deep rooted business relations with Microsoft. Only THEN would you have even the thinnest veil of good conscience behind which to cower.

    You have quite a bit of answering to do, and perhaps the damage is grave enough to be irreparable.

    Next time, you will do well to explain clearly the factors that may have severely compromised your judgment.

    Regards,
    Reuben Cleetus.
  11. Reuben, TSS does not have deep rooted business connections with Microsoft. The most amount of exposure me or my team has had with anyone at Microsoft was the Hard Core Tech Talk we did with a .NET manager earlier this year.

    Here is a metaphor for you to consider. New Line Cinema company is owned by AOL Time Warner. Saying that TSS has deep rooted connections with Microsoft is like saying that the New Line Cinema company has connections Real Networks, just because AOL does.

    "
    One could compare your role to that of a medical journal, presenting various studies on drugs and treatments. What an affront to the Journal's credibility it would be, if it turned out that a drug that got a very flattering review also happened to be manufactured by a close partner of the Journal.
    "

    I understand, but please pay close attention to the key point in your above statement: "...a drug that got a very flattering review also...". Yes, it would be unethical if the journal posted a flattering review, because the review is considered the opinion and official position of the journal.

    This example does not apply to TheServerSide, because we did not put in any opinion/position/editorial. All we did was carry a news item that stated that The Middleware Company did a review, had some related info and contained a link to middleware-company.com. Infact, the news post itself was not posted by me or my team but a TMC employee! So I don't think that our members should blame us for simply carrying the news, because we did not add any official TSS position/opinion.

    Infact, as Rickard pointed out in one of his anti-TSS weblogs, we rarely add opinion to our news posts. What we do is find the news, summarize it in a neutral and factual way, and link to the source of the news, and provide a place to discuss the news and bring people together.

    In this regard, TSS did not act any differently than it did with any other news post we have. I'm sorry people are blaming TSS simply because of our relationship with The Middleware Company.
  12. This example does not apply to TheServerSide, because we

    > did not put in any opinion/position/editorial. All we did
    > was carry a news item that stated that The Middleware
    > Company did a review, had some related info and contained
    > a link to middleware-company.com.

    I thought, until now, that if a news is put on TSS, that is a statement like "Hey, look, we found/heared/produced something good, its worth to spend some time reading it.".
    But how the benchmark was carried out, was at least arguable. The conclusion is arguable, even if the benchmarkresults would reflect reality. Even if you have only 5 minute time to check the news, you can found enough flaws in this report to reconsider publishing the news.

    Ok, you already said, you are not responsible for the news, somebody of TMC put this ( without your knowledge ? ) on TSS.
    But here we are at the beginning of this discussion. Is there no quality insurance ? Can everybody put some news on TSS, without somebody checking quality ? And if not, who is allowed to do so apart from TMC ?

    P.S.: Could you gave me a link to the anti-TSS Weblog of Rick, i can find only his estimation of the benchmark.
  13. "I thought, until now, that if a news is put on TSS, that is a statement like "Hey, look, we found/heared/produced something good, its worth to spend some time reading it.". "

       Usually it is like that. However, sometimes it's also "look, this news may not be good news for you but it is something you should be aware of", and that is the spirit in which TSS allowed this news post to be posted on TheServerSide.

       Discussions of quality control aside, TSS is a J2EE site, and we need to report on things important to J2EE developers. It would not have been acceptable for Slashdot and theregister.co.uk to have reported on the benchmark (which they did), and not TheServerSide. That would have made our members call us incompetent, saying 'I thought this was the place I could go for J2EE news, how come TSS is not reporting this???'

       The job of a news portal is to keep you guys informed. Part of that job means linking ot things like this. To not link to it while other news sites do represents a poor level of service to our members.

    Floyd

    PS - I can't find the post on Rickards blog, its was an old one.
  14. Make a decision. Is TSS a news site, than your handling of this post was ok, but give me a reason why i should prefer TSS over hundreds of other j2ee/java news site.

    Or are you an j2ee site, than i expect you use your knowledge to check quality of a posting.

    I think nobody would have a problem, if the news about the benchmark would have include some statements about the arguable quality.
  15. but give me a reason why i should prefer TSS over hundreds of other j2ee/java news site


       Simply because if are developing enteprise java software, TheServerSide is basically the only place you can go to find out what the latest news is in that industry. We take hours every day to surf all the other news portals and network with the J2EE vendors and within Sun to get you all the news you might care about, all nicely summarized and easy to read.

       We also provide other unique, hard-core sources of knowledge such as the techtalks, the articles, etc.

       If you didn't use TSS but wanted to keep as uptodate on whats going on, you'd have to spend hours a day browsing tens of other sites, and even that would be incomplete, because I get lots of news directly from Sun/vendors/open source project people, that isn't usually announced anywhere else.

       So, Tobias, hats's why you should prefer TSS over hundreds of other J2EE/Java news sites. Speaking of which, if you know of any good ones, please let us know, so that we can check those daily for news as well...

    >Or are you an j2ee site, than i expect you use your knowledge to check quality of a posting.

    >I think nobody would have a problem, if the news about the benchmark would have include some statements about the arguable quality.

        Tobias, I got your comment. I wish we could have done that, but I hope you appreciate that given that we are talking about our parent company, we chose to simply be neutral and let our members decide for themselves.
  16. Floyd and company, and I guess everyone else,

    I think that most of your many members do not truly see you as bought or biased. I used to be a J2EE developer for a national daily newspaper, and I saw how hard it is to post the news without biased editorializing especially given that that newspaper was owned by someone else. Even then, publications both on and off-line tend to earn a reputation for being biased in one way or another (Boston Globe as liberal and Boston Herald as conservative come to mind quickly). I'm not even sure you shouldn't be able to show some bias. Members of the press do it every day.

    We don't really think you're (TSS as opposed to Floyd) biased. The problem is a multi-faceted one. There are always issues in a situation like this, and there are always disaffected people.

    First, while I deeply respect Rickard Oberg's technical ability, I find that his comments are often biased,a bit strident, and often quite aggressive and accusatory for my liking. I think that his rebuttle of the .NET thing certainly had some good points though.

    Second, and this is addressed TSS and its readers: It seems likely that TMC is really quite responsible for placing their child entity, TSS, in a difficult or even untenable situation. The TMC report *did* have a ridiculous number of flaws in it. However, even if it hadn't, J2EE people would find ways to cry foul. It would be the same with .NET developers if it went the other way. This is more about marketing than technology. It's a fact of life. Blame the TMC people for this. I think you guys are killing the messenger.

    The problem is that TSS is getting a ton of "guilt by association" compounded by the fact that TSS posted the article. I don't think TSS did anything wrong even in posting the article, if only because it compelled people like Rickard Oberg to find the flaws.

    Lastly, we could all do with a little less of the religious wars. Our jobs as developers or....whatever....is to develop and protect the information and information architecture of the people we work for or sell software to. In light of that, our first ethical obligation is to our respective companies. Our second, or even third or fourth obligation, is to promote the technologies that we see as good, better, or best.

    I'm almost embarrassed by much of the behaviour of some of the J2EE people that I've seen since this published report.

    If Floyd says that there was nothing untoward in his actions, I believe him. I have no reason or cause to think otherwise. I certainly have plenty of reason to think that Floyd and company had decent motives: this community, the publications they've written, and the respect that they've garnered in our particular field.

    Hell, I'm not even sure I think TMC is in bed after Microsoft. They certainly know who butters their bread better than I. However, I expect that this excersize was more a matter of misguided manegement, bad leadership, or a confluence of events that Microsoft did a better job of controlling. I doubt that anyone at TMC is intentionally malicious, just maybe a bit misguided. If anything, this should damage their reputations, because after reading the reports, I think their J2EE specialists who did the tuning must really suck (sorry for the harsh word) if you were that person and you're reading this, but you should go read some books or something, maybe try a new line of work. I'm no world class developer, but even I could find some real problems.

    Anyway, I think Floyd and company are doing a good job. These accusations will only make their jobs more untenable. They will, in the future, be more careful about posting things like this. Despite what you think, these things are probably good for the community. The adversarial and competitive processes are and have been two of the most critical sucess factors in making relatively open economies like America's so successful, as has an independant and free press.

    Jason McKerr
    Northwest Alliance for Computational Science and Engineering
  17. I have to agree with Jason here. I think the point about the .NET report's shortcomings has well and truely been made and made well, not least due to major contributions by Rickard.

    The other point worth remembering, whilst assaulting the TSS editorial abilities, is that TSS is not just a news portal. If you don't like the news here don't read it. TSS has a lot more to offer, such as a large community of experienced and bright members who are, for the most part, quite willing to help each other out. I think that is worth a lot more than the quality of the news and in itself answeres the question 'Why not go somehwere else?' Because the community is here. Go elsewhere for your news if you like, but you may have trouble finding the same sort of member base elsewhere.

    I guess that in turn reflects on the news as well. Every news item has associated a thread of discussion. I would argue that even if the news is shabby, and I'm not necessarily saying it is, at least here you will find out why.
  18. While I do agree with you Floyd about reporting on items in a fair, unbiased, and timely fashion I do not believe you see yourself or TSS responsible for the information discriminated here.

    When a major newspaper posts and news article that contains an error no matter how small or insignificant you could stake you’re reputation on the fact that the next issue would mention the error or relevant correction. And while you did append several messages to the original news posting explaining TSS's involvement, they came off as pretentious.

    I appreciate the effort put forth by you and others to make this an excellent source for J2EE happenings. And while most articles are obviously non-controversial, you do come across the occasionally piece of news like this so call performance test debacle. Items like these should raise a red flag, and a more vigorous quality control process should be enacted to validity the report no matter who the source is. You must understand the responsibility you wield acting as what you refer to as basically the only good j2ee news site.
  19. While the points of others in this thread have merit (i.e. Jason) the simple fact is the TSS will become a victim of the actions of TMC, regardless of intention.

    As it states on TMC's home page - "The Server Side - A Service of The Middleware Company". You can characterize your relationship with TMC as independant as you like, but we (the community) don't have honest visibility into your relationship and in this situation must assume the worst until we have definitive information to the contrary.

    A direct quote from the article: "Based on this feedback from the Java community, the Middleware Company, as experts with J2EE Web-based application servers and based on feedback from enterprise developer postings on TheServerSide, re-built the Java Pet Store using J2EE..."

    In summation, this test has resulted in a complete loss of credibility in TMC, and as an affiliated organization to TSS, I remain highly skeptical of your motivations as well.

    In mu humble opinion I believe that we have seen the end of TSS as an active and valuable resource to the J2EE community.
  20. I think everyone for their words of support in this thread. And to address Jonathan's (and other people with the same opinion's) point:

    >In mu humble opinion I believe that we have seen the end
    >of TSS as an active and valuable resource to the J2EE
    >community.

       Your statement implies that TSS will somehow change as a result of this benchmark (which we had nothing to do with), as though all of a sudden the TSS team will do a poorer job building J2EE community content, etc.

       This is infact the opposite of what is happening. Our team has grown over the years, we're getting better at what we do and we are actively implementing new features, getting new articles, filming new tech talks, and improving the site in many ways that may not be readily visible to the end user.

       I really am clueless as to why people seem to think that TSS is all of a sudden (or will soon be) less active and valuable. Nothing has changed man! And since the Precise acquisition (TSS now has larger budgets to re-invest into the site) - its only gotten better!

    Floyd
  21. Tobias:

    > I think nobody would have a problem, if the news about the benchmark would have include some statements about the arguable quality.

    Floyd:

    > Tobias, I got your comment. I wish we could have done that, but I hope you appreciate that given that we are talking about our parent company, we chose to simply be neutral and let our members decide for themselves.

    Floyd, I understand the difficulty the "parent" problem creates. But you shouldn't apologize for being "neutral". TSS provides a very useful service for the J2EE community precisely by doing what it did in this case: bringing the benchmark study to the attention of the TSS community.

    It's unfortunate that the benchmark controversy is obscuring that. TSS did the right thing by putting it out for review by the TSS community. The result is the megathread, which has made TMC realize very quickly that it screwed up. That's good, because a fast (but smart) rematch is essential to counter the use Microsoft is making of the TMC blunder.

    It's unfortunate that the megathread is so cluttered with conspiracy comments. I don't believe for a moment that TMC had any desire to "be in bed with" Microsoft. TMC makes its money by selling J2EE expertise. It's not in TMC's interest to blacken either J2EE or its own image of J2EE competence.

    I realize it's hard for you to avoid being defensive when so many people are hitting TSS for publishing the benchmark results. But please don't fall into the opposite trap of being too defensive of the TMC implementation.

    Rickard's comments on the TMC implementation (http://dreambean.com/petstore.html) are valid criticisms. Ignore his comments re TMC's motivations. The best way of dealing with them is to focus on his substantive comments.

    The TMC implementation is outdated (maybe because Petstore 1.2.1 is outdated) and does not use the major improvements introduced in EJB 2.0. It doesn't even follow the good advice in your EJB Design Patterns book or in Ed Roman's second edition (like to use CMP instead of BMP or to cache wherever possible). Rickard's substantive comments really boil down to "TMC, read your own books!"

    I hope for its own sake that TMC will press ahead with a "rematch" that is based on a better J2EE design that takes advantage of the power of modern J2EE 1.3+ servers (1.4 for web services) like WebLogic 7.0 or Websphere 5.0. If TMC doesn't do this, maybe TSS can organize and make available vendor responses that do.

    If TMC does the rematch, it is important that it gets BEA and/or IBM (and/or other server providers, like Rickard and other open-source groups) involved. TMC made a major blunder in not involving them earlier.

    John
  22. After Floyd explained the relationship between TSS and TMC, I trust TSS's intention was purely to disseminate a piece of news that is of high interest to the J2EE community, not endorsing its content.

    Now where is Ed Roman in this whole debacle? We haven't heard a single word from him. If he doesn't come out to do damage control (it is better late than never), his company may lose any chance of being a J2EE consulting shop, although it can morph into a .NET shop and survives the whole thing.

    Also how much internal training does TMC' consultants/trainers receive? How much industrial experience do they have? How do they keep up with the latest J2EE technologies? They don't seem to know the merits of the latest version (1.3.1) of the Petstore, as far as I can tell.
  23. The serverside in bed with microsoft ????[ Go to top ]

    I pose a question to everyone here who has questioned the motives of TSS and TMC. If the results had shown the reverse, would this thread even exist?
    Regardless of the relationship with M$, TMC performed, what I thought, was a reasonable comparison. The fact of the matter is that, a well designed, platform specific technology such as .NET will always outperform a technology such as Java.
    I definetly agree that, as stated in other messages, performance could be improved, but what does it matter? There are only so many users out there and there are very few (if any) sites who require the performance levels .NET can provide and java cannot.
    I know, as with most java advocates, we are all a little dissapointed with the results, but there are other factors which contribute to the overall success of a system and I think it would be found that in a well designed system and environment, java would equal, if not surpass any other technology available.
  24. I pose a question to everyone here who has questioned the

    >motives of TSS and TMC. If the results had shown the reverse,
    > would this thread even exist?

    You're forgetting that there are no results. There is only a fabricated report.

    >Regardless of the relationship with M$, TMC performed, what I
    > thought, was a reasonable comparison.

    Except that it was flawed, contained a massive amount of errors, and even lies.

    >The fact of the matter is that, a well designed, platform
    >specific technology such as .NET will always outperform a
    >technology such as Java.

    That is not a fact since there are to date no reports or benchmarks to support this.

    >I definetly agree that, as stated in other messages,
    >performance could be improved, but what does it matter?

    The idea behind the benchmark report is that performance matters.

    >There are only so many users out there and there are very
    >few (if any) sites who require the performance levels .NET
    >can provide and java cannot.

    Famous last words. Besides, noone has shown to date that .NET can achieve performance levels that Java cannot.

    >I know, as with most java advocates, we are all a little >dissapointed with the results,

    Again, there are no results. Only a fabricated report.

    >but there are other factors which contribute to the overall
    > success of a system and I think it would be found that in >a well designed system and environment, java would equal, >if not surpass any other technology available.

    Absolutely, so why is everyone so fascinated with this performance thing?
  25. Todd: "The fact of the matter is that, a well designed, platform specific technology such as .NET will always outperform a technology such as Java."

    Rickard: "That is not a fact since there are to date no reports or benchmarks to support this."

    Rickard is absolutely correct, except he should use "objective" before the term "benchmarks." Since .NET developers have compared the Sun Hotspot server JVM (for example) against the MS CLR 1.0sp2 and found that their integer operations test run 4x as fast on Hotspot as on the CLR (posted publicly on Microsoft's own news groups,) it is anything but a matter of fact that the CLR is faster, even on Windows.

    The CLR is immature and relatively buggy (compared to Java -- it is much better quality than most tools 1.0 releases from Microsoft.) It is also proprietary and non-portable. And in many tests, it is also slower than Java.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Easily share live data across a cluster!
  26. Don't you people have work to do?
  27. No. I live in Norway, where we're paid to be lazy. My brother lives in America where he's paid to look busy.
  28. Speed comparisons[ Go to top ]

    Cameron:

    > Rickard is absolutely correct, except he should use "objective" before the term "benchmarks." Since .NET developers have compared the Sun Hotspot server JVM (for example) against the MS CLR 1.0sp2 and found that their integer operations test run 4x as fast on Hotspot as on the CLR (posted publicly on Microsoft's own news groups,) it is anything but a matter of fact that the CLR is faster, even on Windows.

    From what I've seen, JRockit (BEA's JVM) is even faster.

    Let's also not forget that J2EE servers have been optimized for a long time: CORBA roots for more than seven years, J2EE servers through many update cycles. There is a lot of performance optimization built into them. And this shows in benchmarks of Websphere vs .NET done by IBM and similar benchmarks done by other server vendors. Admittedly, they're done by J2EE server vendors and so are not "objective". But they're at least as meaningful as benchmarks done by Microsoft. And they show well-designed J2EE apps not only being easier to maintain but also outrunning .NET.

    John
  29. Before you can call yourself a partner of MS you have to sign a contract with MS. Read such a contract and think again.