CRN has released survey results based on more than 9,000 phone interviews conducted between February and April 2001. Application servers were measured by percieved price/performance ratio, quality, scalability, tech support, etc. The final ranking placed the vendors (in order) at: Oracle, IBM, Macromedia, Borland and BEA.
Personally I find these results higly questionable. If the interviews were conducted from Feb. to April, then the interviee's must have been using Oracles old application server product (not the new Orion licensed server). If you take the comments by TheServerSide.com members in the Oracle Application Server Review forum
as an indicator of the status of the old Oracle App. server, then the survey results seem dubious. Infact they seem reasonable if in the context of Oracle Database, not Oracle Application Server.
Read the CRN Surveys application server category results
Read Oracle's press release: Oracle9i Application Server Ranks First in CRN's Channel Champion Survey
The document states : What's exciting about Oracle9i AS is that a lot of what you do in the app server is driven by the database," says Leslie. "What's backing up the app server is the Oracle database, which is proven and well established."
I always thought that an application should be driven by business requirements, which turns into uses cases and then business objects. The data should not drive the application.
Leslie forgets to say (just like Larry at JavaOne) that Oracle's app server is in fact Orion's application server which sells for +10 times cheaper. I guess the extra cost covers thechnical support.
Remember too that during the times of these results Oracle was not using Orion's app server for its java implementation but instead a well-outdated version of Apache JServ. Thus, as has been well-documented here in other threads, it is almost a fact that these individuals surveyed were using other pieces of OAS/IAS, as in the Oracle 8.1.7 database or Reports/Forms. Certainly, these are the only components that could even be remotely considered "reliable", something that CRN ranked IAS first in. Additionally, if you have gotten solid technical support from Oracle you've likely paid an arm-and-a-leg for it. It is a joke article, and a bigger joke that Oracle released it as a press statement.
The key to this whole article is Mathew's statement ""[Oracle's application server] is well-liked in the IT community because it's bundled well with Oracle applications..." Gee -- I'm running to the store -- what a vote of confidence, with such technical validation too. This article is about the database and its surrounding products, not about anything remotely J2EE based. How this article touts the features of the "app server" while ignoring a major components of it is beyond me. Is CRN owned by Oracle?
Yeah, there are a lot of odd things about these results. BEA ranks dead last for Quality? And dead last for Compatability/Scalability? Would anyone say that Macromedia scales better than BEA???? This is made even more complicated by the fact that they don't mention which Macromedia product they're talking about (JRun or ColdFusion)...
Oh well. It is CRN after all.
I question this as well. Like you said, it was conducted prior to the release of the new Orion-based app server. And - while my favorite app server does happen to be Orion, the developer's group on OTN are finding that, like the previous series on the corporate sales cycle, that its not as straight forward as perhaps the marketing folks might have ,well, 'implied'. Interesting, indeed.
I've never seen any clap trap worse than this in my whole life. Another stonking publicity stunt by Big Mouth Larry. Oh well, At least you folks all know the truth. God Save us from the spin doctors
This game of reporting surveys to show that your application server is better really stinks. It's all in the agency doing the survey and the questions they ask. It's pretty easy to get a survey that recommends your product. Just focus on a select group of developers and ask them leading questions. IBM and Microsoft have done this for years. Anyone who reads anything into these surveys should not be evaluating or recommending products. It's obvious though that they are targeting at non-technical decision makers who have influence in product selection.
Good thought. Why don't Oracle do a survey on a open community like serverside.com. I think we are pretty unbais(of course we have some biases against .NET, but I think Oracle will tolerate it)towards vendors,and on top of that we know little bit about J2EE application servers too :) Oracle loves to give challenges, do they know how to accept one.
The order should be reverse -
BEA - BORLAND - JRUN - IBM (EJB 1.0) - ORACLE ( 3 month old 9i)
Two interesting lines from the CRN article:
"[Oracle's application server] is well-liked in the IT community because it's bundled well with Oracle applications,"
"Oracle fared worse in price/performance, where its score of 62.02 put it in last place and eight points behind Borland."
From these two one can conclude either:
1) The IT community is made of lazy morons who prefer bundled app servers with maximize price/performance ratio
2) Oracle sales is doing an aweful lot of brainwashing and paying for plenty of golf trips. :-)
This just further verifies the "Buying into Trouble: Oracle's hard sell illustrates industrywide problems" article circulating on zdnet and cnet (see same title on theserverside.com).
The evaluation results of a product is getting highly confusing because the surveys shows one and the best product awards shows another...and if we read 10 magazines or articles all of them shows different results and different winners...one will show websphere..the other will show BEA...and Oracle 9i has also come int picture....
hmmm. I'll be honest - when the second best IT company at marketing (#1 being Bill's lot) come top in survey like this I wonder how much research has been done looking at www.<insert company>.com/products web pages.
I've used WLS quite a bit and whilst BEA charge like a wounded bull for it it is a quality product. I've never heard a good word spoken (or written until now) about OAS.
I don't lend much credence to any kind of survey which has Oracle Application Server listed as the top application server. This survey definitely reeks of spin and public relations.
Are we sure that the survey wasn't saying something like "9,000 phone interviews of Oracle employee"?
Hey, check this out: link on the homepage of BEA to another article in CRN, written by the SAME author Elizabeth Montalbano, except this one is suspiciously pro-Weblogic!
In that Article OAS is not even mentioned as a compitetor for BEA :-)
Perhaps CRN's eStar is an award given out to the best of all app servers that begins with the letter 'W', while the CRN's survey ranks best app servers that begin with a vowel.
Don't you just love journalistic integrity? :-)
I must admit I was really wound up by the silly results of that rigged survey and challenged the authors. below is the reply and my original message. As you can see, we are shooting the messenger. There is always a Machiavelli somewhere in the background with a vested interest.
michael, i have absolutely nothing to do with the results of the survey. all the
reporters at crn are given categories to cover and the results, and we are
basically told to write a story about the results. to be perfectly honest, i was
extremely frustrated that oracle came out on top because as someone who covers
this field, i know that is certainly not the case in this particular category. i
was only doing my job and frankly, i protested writing the article but was told
to do so by my superiors. i will pass your letter on to the people who make
decisions about the survey--perhaps this will help them change the way they
collect information so it is more accurate next time.
| | "Michael |
| | Fasosin" |
| | |
| | |
| | |
| | 07/02/2001 |
| | 02:28 PM |
| | Please |
| | respond to |
| | mfasosin |
| | |
| To: Elizabeth Montalbano/CHB/CMPNotes@CMPNotes |
| cc: |
| Subject: RE Oracle Tops App Server Survey |
I don't normally respond to what I consider fatuous web surveys but this
survey leaves not only a sour taste in the mouth and it also makes a mockery
of our intelligence. If you think anyone believes your survey my mother gave
birth to Larry Ellison. Don't get me wrong this email is not to insult you
or your intelligence. More a frustrated response to the final straw
Resulting to reading articles such as yours. Normally I would ignore it but
you do my business a disservice by spreading spin without substance. What
Businesses want is quality information to make decision. What Annoys me is
folks like you with the power of the word miss -inform customers they then
end up spending millions of dollars on failed solutions.
By the way, you stated the following :-
What's exciting about Oracle9i AS is that a lot of what you do in the app
server is driven by the database," says Leslie. "What's backing up the app
server is the Oracle database, which is proven and well established."
Application development should always be driven by business requirements,
which turns into uses cases and then business objects. The data should not
drive the application. I doubt if you've ever heard of Model View Controller
Your article seems to be about Databases alone and not about Application
Servers or eBusiness Platforms. I hope and wish you will clarify this as it
is not right for anyone to insult our intelligence. By the way you might
want to review what people that use Application Servers really think about
Oracle Application Server at www.theserverside.com
I apologise if this email sounds aggressive. or angry. I love this industry
I work in unfortunately, there is so much childish antics in it that it
prevents us from looking credible or ever likely to be taken seriously.
My God! So much hatred in this world. I've been looking at the App Server market myself. I can see that BEA has a long following on this site. True, Oracle is a newer player in the app server field. I thought they did a good job addressing their J2EE solution quickly by licensing orion. Their app server also comes with a web cache among other components - probably why it's priced higher than the vanilla orion but (according to Oracle press releases which I haven't verified) cheaper than BEA and IBM plus the additional licenses needed to make them equivalent in feature components. Of course, the idea being the less consulting/integration work for businesses, the better. Has anybody had a bad experience in their new version that was announced at JavaOne? I would like to learn from you.
I remember Larry announcing that the new 9iAS is 3x faster than BEA and IBM or something like that. Is this true?
Larry also said that we need Video-On-Demand and that NC's would replace PCs so I guess Larry's right on this one too ;-).
I think you're missing the point here. Most people are ticked off at CRN for results that are obviously not credible. The real culprit here is CRN not Oracle.
But in this case, the messenger deserves to be shot! If I were a journalist of a reputable publication and was given the results of a survey that is clearly specious, I would strenuously object to writing it!
Note I am not pro or anti Oracle or BEA, I just want jounalistic integrity and consistency. As I already mentioned, Elizabeth writes a pro-Oracle article linked on Oracle site, and a pro-BEA article linked on BEA site. If I were to draw any conclusions, I would say she is on the take, and I cannot take anything she writes seriously!
I've just visited the BEA site and found following message:
"BEA Wins CRN E-Star Award Second Year Running for Successful Partner Program, Market Leadership and Superior Technology". Is it the same CRN?
(full text here: http://www.beasys.com/press/releases/2001/0628_BEA_Wins_EStar.shtml
The new war is coming (after browser and OS war)- the application server war. The BEA published their answer to Oracle and IBM benchmarks.
The big problem with this article is that IT professionals will look at this survey and start to consider a product that is both unproven and untested, I realize that Oracle is using Orion as their Application Server (which two thumbs up for that move) but since now they have gotten their hands on it you never know what they did to it. I actually never had any problems with Jserv, but anyone who used the old version of Oracles Application Server based on Jserv, have had a ton of problems. I have tried almost all of the leading Application Servers and they all have their problems and quirks, but to take a clearly low ranking (especially IN THE IT community) application server and rank it first, and then take one of the leading application servers and rank it last, is just plain idiotic.
I will say this, one thing this article has taught me is anything that comes from CRN is questionable, and you might as well disregard it as not facts, clearly someone paid quite a bit of money to have this "supposed" survey come out like this.