Slashdot Mirror
Microsoft Threatens Oracle Over Benchmarks
n8willis writes "ZDnet is reporting that Microsoft is now threatening to sue Larry Ellison if he doesn't stop saying that Microsoft SQL Server only performs well on benchmarks, not real apps. The meaty part of this story is MS's claim that Ellison's comments violate the SQL Server license agreement, which requires permission before publishing any benchmarks. A dubious term to begin with, if you ask me, but in this case, Ellison is demonstrating the software head-to-head in his speeches, not publishing anything." Frankly, I don't care about it being Microsoft or not - any company deciding that any like this is proprietary is ridiculous. I mean, if I wanted to publish database results, does this mean I can't write a report about it?
> This is typical of the Slashdot mentality
What mentality exactly are you talking about? The idea that after I've read a book and people ask me how I liked it, I shouldn't be able to share that opinion? Or the idea that car magazine journalists test driving a new car shouldn't be allowed to publish its top speed or any possible shortcomings? Unless you've been living under a rock sheltered by warmth and humidity, you would know that it's far from only the Slashdot community that is dissatisfied with the state of EULAs today, and that Microsoft in particular takes all kinds of liberties with them. By saying that if you don't agree with the EULA you shouldn't use the software you are trivializing the matter beyond any reasonable discussion.
Do you think Microsoft would react this way if Oracle did not have a point? If their product was faster for real-life uses, they'd definitely be demonstrating in the same way. The benchmarks are in need of a serious overhaul, anyway. All anyone does now is tune and design the physical layout just for a particular benchmark -- in MS's case they created many, many small partitions of data on brand new 15,000 rpm compaq drives. Throw in 192 CPUs... It still bareley beat DB2 which ran on an older box. Very, very few applications are designed, or could be designed, to take advantage of how they tuned sql server for the tpc benchmark. DB2 and Oracle still pound sql server for performance and reliability in the real world.
For those that do not know, Consumer reports is a non profit organization that compares products. They have nothing to do with software specifically, but that is my point.
--- -- - -
Give me LIBERTY, or give me a check.
Note that Oracle has EXACTLY THE SAME CLAUSE in their EULA as MS.
The last thing you want is your competitor publishing benchmarks on your software.
The reason for this is obvious. It's just too easy to cook database benchmarks to favor whatever you want to favor.
So only the TCP (a neutral third party) is allowed to publish them.
I'll bet that this is completely enforceable. ESPCIALLY when the person you are trying to enforce it on is a competitor who has the same clause in their EULA.
tj
Isn't it REASONABLE for them to tell Larry not to publish benchmarks of their software?
It's not like they are sending Cease and Desist letters to a neutral 3rd party...
tj
INAL, but no, not all contracts are explicit and signed. I order a meal in a restaurant and it's mutually understood that I'll be responsible for the bill before I leave. My guess is that the software industry will assert that the same general knowledge applies, that professionals at least (and SQL Server isn't a consumer product) have been familiar with agreements like this for years.
There are parts of the whole thing which are a bit smelly, it's true, but some of those are simply defensive. Read the tiny print on a box of film, for instance -- Kodak is not responsible for the value of your pictures of Elvis smoking crack with Jimi Hendrix and two aliens, and if their film fails they will refund or replace the film and only the film, just as Oracle will replace your media but not your company if their database fails you. If you want to run your life support system on OS/2, I'm guessing that IBM will in no way accept responsibility for your demise, and I don't imagine that Dell or Compaq will either. Record companies don't warrant that the album will be good, only that it will play.
On the other side of the street, is Linux warranted for any particular purpose? Is Linus going to lose his shirt the first time somebody gets hurt? (I speak hypothetically here -- I'm sure somebody's already been hurt as a result of some sort of failure -- it's almost inevitable in an imperfect world.)
AFAIK you can sell the media if you'd like -- just obliterate the contents first.
This next song is very sad. Please clap along. -- Robin Zander
> to prevent unscrupulous people from skewing the data.
And what is to prevent the vendor from being the unscrupulous one?
-- Don't Tase me, bro!
It is interesting to note that the EULA for Windows says nothing of the sort.
If MS really is the great satan, AND they
have a durable monopoly, why don't they?
tj
hmm...what might those be?
This kind of reminds me of the AD&D campaign world Planescape and the Blood War. Essentially, all the creatures of Baator (the Lawful Evil outer plane) and the Abyss (the Chaotic Evil outer plane) are in an eternal war with each other. No one from any other plane wants either party to win, because once one of them does, there's nothing to hold them in check anymore, and who knows who they'll go after next? I'm not saying either company is evil (I'm sure there's enough people here to give opinions on that), but do we want either one to win?
I think this headline really misses the point of what Microsoft is trying to do. They're just defending their product against unfair claims and trying to find a legal way to do it. I don't think it's the best way, but I do think it is pretty unfair for Oracle to insinuate that SQL Server is optimized for benchmarks and then to show off examples that are optimized for Oracle. Oracle is just slandering Microsoft because Ellison has an inferiority complex and Bill is too nerdy to know how to fight back.
--- Don't be a player hater: I meta-mod ALL negative mods as Unfair.
All "winning" combinations from MS are achieved in a farm of clustered servers; Oracle on Sun is on one BIG box.
These results are incomparable, since you can't always use a cluster, especially transparently for the application.
There are 2 primary types of database applications: OLTP and Data Warehousing.
OLTP can be used on cluster only if the application is explicitely written for it.
Tigers respect lions, elephants and hippos. Maggots respect no one. (C) S. Dovlatov
> Upon entering the premises, representatives of our company may help themselves...
>
I disagree.
Let's make Ricky Martin the referee who gets the crap beaten out of him, just like they do in WWF.
Bill Clinton: Pimp we can believe in. - The Shirt!!!
> Software that sells for 100k a copy doesn't mess around with wimpy licenses like the one at issue.
The company I work for sells it's software for 100-500k per installation, and we don't have any of that obscene crap in our legal agreements. You have to have hardass marketdroids and 3leet managers to end up with shit like that.
First, I have to say, this situation is ridiculous. Thankfully, MS doesn't rule the world enough to force us to use MS SQL (although they'd like to think they do), and as I don't use MS SQL, I have no reason to complain about the license. This benchmark-NDA would be a real thorn if I had to use MS SQL (or Oracle), though.
This being said, the whole thing is a chess game of billion-dollar egos. Larry Ellison is, in a roundabout way, launching FUD at MS. The best way for MS to refute Ellison's claim is by having him disclose the info he's gathered or conjured; if MS doesn't demand that, it could make it seem like MS is scared of the results. Instead of refuting Ellison's claim, MS is whining about technicalities. I think Larry's winning this round of ego-chess, unless he's bluffing.
Kelledin Tane, the Dreaming Minstrel
http://kelledin.tripod.com/scovsms.jpg
Larry Ellison Exercises Jerk Options
---
Bush is about as pro free market as Reagan was. Talk the free trade up, while increasing government interference in the marketplace to "level the playing field" for American corporations v. the rest of the world on the one hand and corporations v. the workers on the other. The free market they speak of is one where corporations are free to operate without incurring any of the traditional risks and liabilities associated with doing business.
As the parent to your post wrote, if American leaders were as concerned about maintaining the free market as they like to appear in front of gullible TV audiences, we wouldn't have the government bailing out LTCM with tax payer money, nor would perversions such as the current EULAs be legal.
--
Violence is necessary, it is as American as cherry pie.
H. Rap Brown
The publicity would certainly be favorable to Oracle...
sulli
RTFJ.
The conflict between Oracle and Microsoft on the topic of database performance has been extremely nasty of late. Essentially, Oracle was taken entirely by surprise when some respected tests revealed that SQL2000 on Win2k beats the pants of Sun/Oracle peformance.
:-)
You can see the benchmarks here: TPC tests
The margin by which Oracle is beaten must be pretty humiliating when they are competing for the same account. Oracle and Sun have attempted all sorts of dirty tricks to disqualify the results, and TPC even removed them at one point, but MS has always fixed the technicality and pumped out even higher results at even lower prices.
Hence Oracle's recent market promise that if they can't triple your speed they will give you a million dollars. Some of us inside have joked about setting up a really badass SQL service and getting that million as a stunt
At any rate, I am not surprised that the two companies eventually are coming to slights and legal maneuvering. Oracle knows quite well that they are one of the companies Microsoft has marked as "Make irrelevant in five years" and they really don't want to go the way of Lotus, Corel, etc.
-konstant
Yes! We are all individuals! I'm not!
-konstant
Yes! We are all individuals! I'm not!
If they push the point of absurd and unconstitutional EULAs that don't even permit speech cover in 'fair use' reviews, they may lose ground on hte only thing they've successfully innovated: license agreements.
Once they lose that, they lose the primary tool of holding their marketshare.
Two companies with the most disgusting EULAs on the planet attacking each other. No matter which one comes out on top, humanity wins. Kind of like a fight to the death between the Spice Girls and Hanson. I like it.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
86.v.(2): The user of this software Shall, upon acceptance of this license (i.e., through usage of this software), be heretofore bound to release the parentship and legal guardianhood of their first born child;
86.v.(3): In the event that the user of this software has no progeny, through adoption or sanguine procreation, the user shall complete their responsibility outlined in this agreement upon such time as they are able to procure the aforementioned child;
[...]
In the overwhelming spew of anti-MS hate that is sure to issue forth over this, it will be simply ignored that Oracle has exactly the same claus in it's agreements.
But that won't matter - this is a good anti-ms sound byte.
Yes, but on the other hand, in certain countries (eg : Belgium), it's not legal to advertize using benchmarks to begin with. I think it'd be okay for Larry in a keynote speech, but he definitely couldn't do an ad on TV or radio or whatever saying "x% faster and y% cheaper than [competitor]"
A EULA with this sort of verbage in it would stand no chance in court of being upheld. It is very similar to the sale of a book with the provision that it not be placed in a library. Coming at a time when the entire M$ anti-trust thing is coming up for review, it smacks of stupidity/ego on the part of someone at M$, rather than merit.
One of the conditions for a healthy free market is that the consumers can freely choose between competing products. How is that possible if corporations are able to prevent the publication of less flattering reviews of their products or test the products themselves (which could be made impossible by the UCITA backed shrink-wraps)?
MS SQL Server: connects fast, goes downhill from then onwards.
;8^)
.|` Clouds cross the black moonlight,
Oracle: doesn't connect, stays that good from then onwards.
So sue me, M$loth - I've let the cat out of the bag now...
~Tim
--
~Tim
--
Rushing on down to the circle of the turn
I make a car. The fuel tank explodes at 46 miles per hour. However, there is a licence in the glove compartment (which you agree to by buying the car) that says I am not liable for any damage caused by driving at 46mph, nor may you tell anyone that your car exploded. If someone else sees your car exploding, and writes about it, that's a violation of your licence, too.
It is your job to reassemble the car, but I can recommend some extremely expensive mechanics who just happen to work for me.
</metaphor>
Yes, the licence is odious. But the product is also inferior. I'm not voting for Nader, but this falls under the "keep it legal to embarass big companies" umbrella.
--
--
E_NOSIG
..at least for me. I use SQL Server 7.0 for a ton of database-driven projects, and have had nothing but success with it. Good performance, too, and a few of the projects are pretty high traffic.
:)
Of course, the fact it's running on a dual P3-600 with a gig of RAM might have something to do with that. YMMV.
Just my two cents..
Mike
The games developer Apogee (a.k.a. 3DRealms) even added a clause to its license agreement earlier this year that prevented negative reviews of its product!
This got me thinking. What if I never bought the software or even use it, but entered the house of someone who does use the software. You know, people have friends and like to show stuff off. What if I think that game Apogee made sucked, or if I saw my friend getting half the performance off Oracle than mysql? Could I start shooting off my mouth, or would I be gagged just by entering his private residence?
I do ask a lot of questions when seeing a demo of something I have never seen before. So would just being in the presense of a friend's computer bind me to top security? Do these software licenses require ISO procedures for security guards and access to the perimeters? Could they do any of this legally?
I think its real neat software companies have this sense of power. Too bad I don't share their views. My experience has shown not too many other people respect their values either. Enter the world of "piracy" as they call it. No, not following the letter of a license is not raping and pillaging their employees. Its this kind of arrogance that has turned me away completely from commercial software, supporting their business model, and has me backing freely licensed software.
Which is meaningless in court when someone makes a "public policy" arguement. I'm not saying Ellison isn't a wanker and hypocrite, I'm saying that *I* don't think that a EULA clause that gags basic performance reporting can be enforced.
The last thing you want is your competitor publishing benchmarks on your software.
Only an idiot would TRUST a competitor's benchmarks (or Ellison's dog-and-pony show). The REASON such a EULA clause is likely against public policy is that it gags any independent parties that don't toe a manufacturer's line (and lets them optimize for a chosen set of benchmarks, rather than the benchmarks of everybody).
The reason for this is obvious. It's just too easy to cook database benchmarks to favor whatever you want to favor.
You're right, it's obvious. That's why no one will trust an Oracle bench of MS or an MS bench of Oracle, or a GM review of Ford or a Ford review of GM, or Pepsi survey on Coke. It's obvious. What the EULA clause does is allow a company to (try to) control the press they get and limit it to tame third parties. If the auto industry could EULA out Consumer Reports, they would. They can't and MS and Oracle shouldn't be able to gag everybody either, IMHO.
I'll bet that this is completely enforceable. ESPCIALLY when the person you are trying to enforce it on is a competitor who has the same clause in their EULA.
Again, in a "public policy" argument, that won't hold water, 'cause
- everybody gets treated the same in those arguments,
- it's taken as a given that a competitor who can benefit from winning such an argument is going to carry it forward in a lawsuit
"Public policy" doesn't change depending on the party to the lawsuit, and if a clause is against "public policy" it doesn't matter who's paying for (or benefiting from) the legal work that produces such a decision, in fact, it's a positive that somebody with a vested interest is making the fight, rather having the government decide what's worth fighting for or waiting for some small-fry to risk everything "on principle".Hahahahaa... you believe anything that comes out of Shrub's mouth? He will never stand up to any corporation, and will never fight against any laws like UCITA that benefit mainly large corporations. HAhahahaha... I laughed so hard my sides hurt. (This is not to say that Al Gore will either, they both suck... the USA is a two-party dictatorship.)
---
Seriously, you don't even have to know they tell you this, they can publish it on a website that you're supposed to check (yeah, right) and it's enforceable.
Aren't you glad you're not voting? See what happens when you decide to not hold your legislators' feet to the fire? [note - I'm not talking about the State Rep from Redmond, Laura Ruderman, she actually is part of why UCITA didn't pass in Washington State, plus she's cute]
--- Will in Seattle - What are you doing to fight the War?
"When they don't he whines about how it's real-world experience which show them as being better."
I've never known Ellison to push benchmark results. When did you hear him doing this? Certainly the topic of this story is a situation where he specifically does a demo INSTEAD of citing benchmarks....
"This fucked up company is also purposefully crippling their database as a server platform with WinNT/2k clients"
Um... as I understand it, they drew the line at using MS' API. They refused to do what everyone else does: code to undocumented calls so that MS can break their software every service-pack. I have a hard time calling this "crippling".
I was wondering if this is a way to get around the license: Say we have Mr X and Mr Y. Mr Y installs the software and agrees to the license. He does a benchmark and tells Mr X about it in private. Mr X publishes what he is told, and since he never agreed to the license, hes ok. Wouldn't this be an easy work around?
The article says he used a copy of mssql to do the benchmarks, but I don't know if it means Larry himself. Would the CEO of Oracle be doing benchmarks?
Yeah, then Oracle gets power of discovery to rifle through MS internal documents... and then the benchmarks become part of the public court record.
Shrewd move, Bill.
Read the details. You have to buy the hardware and software recommended by Oracle. In addition, you have to let Oracle Consultants tune it for up to 90 days. The cost of just that runs easily into the millions of dollars.
In addition, if Oracle were to fail and your website didn't run 3 times as fast, they would pay you the million dollars but you wouldn't be able to tell anyone about it. That's right, the only stories you will ever hear are about sites which get the improvement. The contract specifically requires that any site which wins the million dollars can't talk about it.
Mmmm.. Donuts
"Do you expect me to talk?" "No, Mr. Bond. I expect you to die!"
taco was laughing about you, specifically, (regarding the truncation of your user name) when he gave the talk at MIT recently.
Your wallet stays open. Our source remains closed. We are MSFT
If benchmarks are intellectual property, does this mean that anything the software is even vaugely related to auctomatically becomes Microsoft intellectual property as well? is this reply MS-IP because i mention SQL server? How about the system that the Database is running on, does that become intellectual property too? To paraphrase fletch "It's all intellectual property these days"
--
There are some people that if they don't know, you can't tell 'em.
Perhaps you could explain how the pricing is weird? I found it's very logical and cheap! Unlimited users if you pay per Mhz! Two year licences are dirt cheap weather you pay per user or per Mhz. If you went with an MS license you would have to pay per user AND per CPU for internet connections AND a per user for the NT license. Combine this with the fact that MS stops supporting older products thereby forcing you to upgrade every two years anyway oracle comes out much cheaper.
Oh yea DB/2 is cheaper still.
Postgres, interbase, SAPdb are all robust, proven, feature rich, and are FREE! All of these databases will meet the needs of the 90% of businesses in the world and they don't cost jack squat. According to SAP 800 companies are running R/3 on SAPDB. That's a mission critical, enterprise level, 24X7 app we are talking about here.
In two years all databases will be free including oracle and mssql mark my words.
A Dick and a Bush .. You know somebody's gonna get screwed.
War is necrophilia.
Judges have some ability to strike sections from license agreements that are unfair, illegal or not in the public interest. You can't just put arbitrary things in a contract or license and expect the courts to enforce them.
The problem is that companies will knowingly put unenforceable terms in licenses and contracts for their intimidation value.
Mea navis aericumbens anguillis abundat
If you're in a position to convince your management that Microsoft is not the best solution out there. Here's a sample line of FUD to get you started;
"As demonstrated by this latest action, Microsoft clearly demonstrates that they hafve something to hide. If you go with Microsoft products and at some point in the future would like to point out shortcomings or your dissatisfaction with some product, don't be surprised to learn that you cannot share this information with anyone outside your organization."
Fist Prost
"We're talking about a planet of helpdesks."
Fist Prost
"We're talking about a planet of helpdesks."
-Jaron Lanier
Whenever I suggest a article with a bad Linux or Transmeta benchmark, Slashdot doesn't post it.
Not allow benchmarks to be published without permission seems to be the standard for the database companies. I know Microsoft, Oracle, Sybase, and Informix all do this.
While we all agree to hate Microsoft collectively, Ellison is nothing but a Gates wanna-be. If anything, he's a much bigger egomaniac than Gates ever had the personality to be, topped maybe only by Scott McNealy or Steve Jobs. So maybe we should just include him on our list of daily hate prayers.
that was recently recalled, although in europe this is still seen as 'Not done'
//rdj
No one can understand the truth until he drinks of coffee's frothy goodness.
--Sheikh Abd-Al-Kadir, 1587
If Ellison violates the legal agreement he made by using the software, then he should be prepared to suffer the consequences.
Nobody forced him to accept the terms of Microsoft's licensing agreement.
--
--
You are a fucking moron.
Ellison pushes benchmarks when they show Oracle is faster.
When they don't he whines about how it's real-world experience which show them as being better. How does this sound when ebay has been having severe database problems for the last month?
This fucked up company is also purposefully crippling their database as a server platform with WinNT/2k clients because of their hatred of Microsoft.
Then in light of the weird new pricing strategy, what are customers supposed to do? It's a lot easier to switch out a backend database than it is clients, especially if new database software costs a tenth as much.
Oh yeah, isn't it about time for the yearly "We will bury Seibel" announcement?
Larry Ellison needs a boot to the head.
Oracle, Informix are this way as well. Microsoft didn't start this stupid trend, but I dont know why the continue it. One thing im sure everyone knows is people pay a lot of money on a per user basis for a good fast database. over 1,000 per user per server easy. Not only that but Oracle loves some of its new pricing schemes, not only is it per processor, but they calc a rating on the machines cpu power total, and charge a price based on that. Anyways the only way these companies will get more money, is if people keep believing in some mystical way they are the fastest with out any hard proof. Unfortantly there is no real way to see which one is fastest unless you buy each database yourself and run some benchmarks. We still dont have an open source database that can even get close to the speed of Oracle, Informix or MS SQL Server. It will be very cool when that is no longer true.
Fortunately, Id isn't as protective of their "intellectual property" as Microsoft is. They did provide the source to qagame so that modmakers could exploit the entire code. Meanwhile, even attempting to start a Windows programming project requires you to plunk down hundreds, and even thousands of dollars to Microsoft just to fund their cause.
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
When you buy software, you are actually purchasing a license to use to the software,
Legal grounds for that statement? IANAL, but the Software Copyright Act of 1980 is, IIRC, still the controlling act for software purchases.
Now, businesses generally purchase licenses for software, true. But when you go into CompUSA and buy a box with the media, you are purchasing one copy of the software, as per the Software Copyright Act, and you have all the rights you recieve under the Act from the moment of purchase.
Now, after you open the box, you may be presented with a license agreement. You are only bound by a contract under common law if you recieve something of value in exchange, and most licenses do not grant you any rights not specified in the Software Copyright Act of 1980. Accordingly, if all your actions with the software are permitted under the Software Copyright Act of 1980, you are not bound by the license; you are only bound if you do something the license permits but the law does not.
Now, the UCTIA is trying to change that, true. But the average home user does not license his software; he buys a copy, whatever the publisher tries to claim.
There's no "we" in team, only "me"
Heh, it's clauses like these that make me glad our building's vicious guard dogs can't read. :)
Um, that would be Nixon, in his 4th term. If you look carefully at the bits of scenes in NORAD, you can spot Henry Kissinger, Gerald Ford tripping on his way out of Air Force 2, and G. Gordon Liddy. Also, in the first issue, when the cops are investigating the death of the Comedian, check out the picture with Vice President Ford. That is such a great story, too. Take a look at issue 6, comparing the first page to the last page, working towards thie middle...
itachi
The key difference is in what kind of benchmark it is. Q3 is used to compare the capabilities of 3D accelerators, while SQL and Oracle benchmark tests are used to compare transaction throughput. Also, I agree on one point of yours; Q3's reputation as a benchmarking tool is not one of its selling points; the playability and excitement of playing it ARE selling points. That's why more people like Q3; Epic/DE/Legend has built a reputation of combining glitzy level architecture with strange weapons. People are turned off by that, since they know full well that most of the weapons could work (let alone exist) in real life.
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
Let me remind you that Oracle's License Agreement is the same in terms of that No benchmarking should be published without Oracle's authorization.
My point is Oracle is even more of an idiot, and I don't know why people are still use Oracle databases. MS SQL Server and IBM DB2 were rated the best in the benchmarking test.
It's all about good marketing.
... with a shitload of experience (including tuning Oracle statements to be blazingly fast), I can tell you EXACTLY why Oracle, MS et cetera prohibit publishing any benchmarks.
Any database has an incredibly complex program in the heart of it - SQL parser/optimizer that decides how each statement is executed. Select an inefficient plan, and you're toast. The plan depends not only on the statement, but also on the server parameters, current load, database setup and DBA policies (such as analyzing tables' and indexes' contents). I've had situations when statements running minutes on test server were taking hours when released to production (fixed by adding "hints", that force the optimizer to execute it the way I want it to). Database SQL Optimizer might be more complex and sophisticated than an underlying operating system.
Since my experience includes not only Oracle but some other SQL databases, I can say that different databases optimize some statements differently. There are many ways to do the same things in SQL. Good example will be Oracle with EXISTS construct more efficient than the IN one (you can usually code the same statement using either).
Thus, comparing different databases on the same SQL code MIGHT become a comparison of apples and oranges. If you take a code, optimized for Oracle (even if it is completely standards-based) and run it on Informix or DB2, you might get results that are inadequate thus "proving" that Oracle is better.
TPC is supposed to be a database-independent performance test, but in reality it's obsolete and allows database vendors to optimize their databases heavily to run just this test.
Another thing, all of the top results are not comparable either (even farther than apples and oranges, it becomes a comparison of an elephant (Oracle on Starfire) versus bunch of ants (SQL Server on cluster)). These configurations are for different tasks and can't be compared directly.
Tigers respect lions, elephants and hippos. Maggots respect no one. (C) S. Dovlatov
Last I saw, Oracle has the same clauses in their license, too.
It's amusing, too, that Larry would be talking about someone else's product only performs well on benchmarks. A lot of people probably don't remember that Oracle fixed their DB server up to excel at the TPC-C benchmark when it came out, and did it in a way that was worthless in real world tests. To put it bluntly, they did what they are accusing Microsoft of doing.
The only thing worse than Bill Gates is the legion of people who want to emulate him, but at the same time are envious of his success. Larry Ellison is the defacto king of that group.
Do you have ESP?
Oh god... someone shut the flashing red lights and the sirens off... For a lot of benchmarks, more independant organizations should be created and funded by all the companies who wish to have their products benchmarked, with permission given to publish their findings however they see fit. Funded by the companies that create the software? And what is Linux supposed to do? Besides that, benchmarking organizations are to busy and dumb to learn how to tweek a database system or an OS or anything else for performance. Better to have a public run off where trusted representives of the development communities involved arrive and tailor their own software to identical hardware platforms. And then document and publish every tweek and customization so that the results can be reproduced independently, just like in the scientific method.
It appears that Big Bad Bill does not want to leave the little guys alone. If Microsoft wants to remain dominant in the future, then the company should focus more on product quality, rather than squeeze every penny out of its competitors over lame copyright issues. The hack reported earlier today is a start, but what Microsoft should do is step back and reassess the situation the company is in.
Revealing this nefarious tactic is what Freedom of The Press is all about
Larry isn't 'The Press' though.
I agree that people doing objective reporting should be able to publish benchmarks, EULA be damned.
I also think Larry's got a better product.
I don't think, however, it is unreasonable for Microsoft to be annoyed with Larry for doing this when they have no way of knowing how misconfigured the MS SQL install is vs. how optimized the Oracle install is.
--
Some people have a way with words, and some people, um, thingy.
When you buy software, you are actually purchasing a license to use to the software, and vendors routinely insert clauses in their license agreements that help them meet their goals in the marketplace.
For example, Netscape prohibits publishing benchmarks in the license agreement for its Client software. The games developer Apogee (a.k.a. 3DRealms) even added a clause to its license agreement earlier this year that prevented negative reviews of its product! So Microsoft is really not acting unusually harsh in this regard.
True, but this is what marketing is built on and they know that. They've done it themselves, because:
There's lies, damn lies and statistics.
--
A feeling of having made the same mistake before: Deja Foobar
Well, see, the problem is that the Oracle EULA contains the same provisions that Microsoft is accusing Oracle of violating. Now, Microsoft could retaliate and do the same thing that Oracle is doing, but that would be illegal according to Oracle's EULA. Instead of taking the approach that most Slashdotters would take to resolve something like this (which would be to go ahead and respond in kind, thinking that "it's okay to break the law, Oracle did it first!"), Microsoft's going through the courts. People here might think that two wrongs make a right, but that's pretty shaky law.
Cheers,
Stuff like this is the real danger of UCITA -- it would put a chill on any real criticism of published software. You want to say something bad about Microsoft SQL Server? You have to get Microsoft's permission first (which, of course, they won't give you). IF you say it anyway, they slap you with a law suit.
UCITA == adios, First Amendment.
--
Tired of FB/Google censorship? Visit UNCENSORED!
The EULA for most hardware CAD design software (titles like PowerPCB & Orcad) reads (in summary) like this:
-You can't use this software in any manner designed for figuring out how fast it is.
-You can't benchmark this software. You can't publish benchmarks of this software. You can't sell it to someone who will publish benchmarks. You don't own the software, but if it's stolen, and someone makes benchmarks with your copy, you're liable.
-By agreeing to this license, you agree that
-We may inspect your premises with a software agent team at any time, 24/7/365. If you are not there to assist us with entering your premises, we may let ourselves in.
-We may recall your license electronically at any time, for any reason.
... i could go on but my fingers hurt.
Software that sells for 100k a copy doesn't mess around with wimpy licenses like the one at issue.
The real problem is the UCITA.
--
What happens when you outlaw guns
I've changed many online agreements to suit myself and clicked submit. If the server doesn't protest then the same logic that says one click is enough should apply the other way too.
In fact, many actually put the agreement in an editable textarea so you can change things even more easily. So far none has refused things.
Make sure you keep a printed copy of the resulting edit.
Cheerio,
Link.
Not allow benchmarks to be published without permission seems to be the standard for the database companies. I know Microsoft, Oracle, Sybase, and Informix all do this.
Not true for DB2 from IBM. Okay - so I work on DB2 for IBM Canada but I would be seriously worried if we prohibited people from running, testing and comparing our software against the competition. Oracle's license makes a nonsense of the already restrictive license agreements that most commercial software comes with. But ask yourself this - why would a software company not want you to run benchmarks on their software?
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
Are TPC benchmarks useful? Really when you think about it it's more a benchmark of how well the clustering and hardware performed during the test. The top spot which Microsoft holds on a Compaq machine is a 192 processor Xeon cluster with a huge number of dual processor clients etc.
These benchmarks are useful - they give the performance teams at IBM something to aim at and keep us on our toes in development :-)
On a more serious note, they give you a good idea of how these databases scale when you are talking about SMP/MPP systems. Oracle suffers in large clusters of machines because they have a shared disk array system, so there is always an overhead keeping all the parts of the cluster in sync. Both DB2 UDB and SQL Server have a shared nothing architecture, which means you have to carry copies of certain data structures around the network, but you only take it where it's needed. This wins when you get to 8 or more nodes in a cluster and gets better as you add more. Having said that, it's interesting to compare that SQL Server TPCC benchmark done on a 192 processor cluster scoring aroudn 520000 transactions against the 128 processor DB2 UDB benchmark (same CPU type) scoring 440000. Take your pick as to which is the faster database (hint DB2 UDB scales nearly linearly as you add cluster nodes).
Not that I like a bit of shameless self promotion ;-)
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
Change "could" to "couldn't". Yes, I forgot to use Preview, slap me with the steel ruler.
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
(I'd have replied in email but you don't have one listed.)
and now slashdot has cut off the end of your sig, too..
I guess these days I've been demoted from hacker to hackee. B-)
taco was laughing about you, specifically, (regarding the truncation of your user name) when he gave the talk at IT recently.
Just so he was laughing WITH me.
Maybe I should give up and pick a new handle.
(He once offered to make an exception to the "no changing handles" rule due to my being savaged by the software revision, but I couldn't think of a good replacement and couldn't resist the sigline pun. Perhaps I should think harder and then see if the offer is still open.)
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
"We must abolish every right of the consumer to fairly evaluate every product on the market" Yes and of course Oracle will evaluate MS SQL very fairly
Comment removed based on user account deletion
Shrub was coined, I believe, by the columnist Molly Ivans -- since he is a 'Junior Bush'. You might even say he is genuine Texas home-grown dope. At least his father worked hard for what he achieved and is an intelligent man. Shrub hasn't had to work for anything in his life. The only 'successful' business deal he was ever involved in was a sweetheart deal set up by Daddy's buddies and financed by Arlington, TX taxpayers (I'm talking about the TX Rangers baseball deal of course.)
---
Comment removed based on user account deletion
Comment removed based on user account deletion
I'm still trying to visualize the EULA written on the paper band wrapped around the silverware/napkin.
By breaking this seal, you agree.........
This could provide a watermark for challenging EULAs. After all, publishing your own benchmark is akin to "fair use" of copyrighted materials. It is truly remarkable that companies like M$ and Oracle place draconian restrictions in their EULAs that prohibit things that are ABSOLUTELY guaranteed to someone under copyright law, such as prohibitions on reverse engineering and fair use commentary in the form of published benchmarks.
Myriam Miedzian, who has published a book on boys and violence, recently wrote a column in the Baltimore Sun in which she reported that as a kid George W. Bush used to kill frogs by putting firecrackers into them, tossing them in the air, and watching them blow up.
Now I'm looking forward to the day that Firstone only licenses their tires to me.
Remember, you only bought the rubber the tires were made out of, not their IP (the tire shape/tread), which can be revoked at any time.
My copy of the EULA _does_ say that! Ahhhhhh!
-- Never make a general statement.
Ahh! But is showing a live demonstration the same thing as 'publishing benchmarks?'.. I don't think so. It's simply a public viewing of reality.
Of course Microsoft can sue. SQL Server is just plain better than Oracle. Has anyone actually tried to use Oracle? It's a nightmare compared to SQL Server.
At any rate, thank goodness for competition. Have you seen those numbers? Without compteition, we'd probably still be in the dark ages of 100,000 tpmC. Hah.
You quitting proves that the karma kap worked. The most annoying of the whores shut up. --CmdrTaco
Then who is the press???
Like others have said, I hope Oracle fights this one. If a corporate interest could claim legal control over product reviews that made them look bad then there could be a lot of lawsuits out there.
Of those possible, the potential for lawsuits from drug companies and food manufacturers scare me the most.
It's all about the basic fact that corporations are not citizens.
-C------------
"Cut word lines. Cut music lines. Smash the control images. Smash the control machine." - William S. Burroughs
I believe the license for the electronic structure program Gaussian contains a similar clause.
BTW, MS-SQL was originally a port of Sybase; but MS forked the code around version 4.x and the quality has been going steadily downhill ever since. Sybase managed to screw up pretty badly with version 10, but their 11.x and 12.0 versions are highly impressive.
Why is it that the proponents of "one nation under God" are so eager to get rid of "liberty and justice for all"?
As I see it Larry just reacting in his own way to the shameless marketing semi-truth from M$. I would probably react like this if somebody would do something wrongly suggesting that his product is superior to mine. TPC test published by Redmond was done on 183 (!)SQL 7 database servers and compared with one Oracle server. If M$ would test their Access or Excell in this configuration then thay would have the same result for Access or Excell. All Larry was saing that if you make yor real life choice, use real life tests. I do not see him as an angel, but I applaud him for what this one. That is why he is who he is. Way to go Larry.
However, I think in Microsoft's case, the EULA could be defeated in court because one could argue that the contract is not voluntary. Contract law invalidates any contract that was entered involuntarily by either party. Since many people need software that forces them to use Microsoft operating systems and software, and there is no alternataive in many cases, one could argue that the contract was entered involuntarily.
I'm not a lawyer, but I would love to see someone try this and win!
The only good weather is bad weather.
"Because you don't like MS's EULA for SQL Server, you should not buy it. If you agree with the EULA, go ahead and buy it"
In the USofA, this issue has been fought out three times: in the 1930's, 1950's, and 1980's, each time involving Consumer Reports magazine and their right to publish reviews of products sold on the open market. IRRC Consumer Reports essentially tied in the 1930 case and won the 1950 and 1980 cases. The basic issue being: do consumers give up all 1st Amendment rights when they participate in a business transaction? And the answer, at least as of 1988 or so, being NO. But we all know that these things can change, and I guess that since we are on Internet time we couldn't wait until 2010 to go through this process again.
[Note: I am not a big fan of CR in general, so don't flame me for their sins, but in this case I think they are absolutely right].
sPh
I don't understand why this isn't a free speech issue. If this was a government entity, not only would WE be all over them, but three or four watchdog groups and probably a fair portion of the population.
This isn't the only place this is happening. I'm hearing about legal cases where people aren't allowed to say anything to others about their own case. I don't know about you, but this is FAR more worrisome to me than web censorship on public machines. Somehow, contract and other branches of the law are overriding people's right to free speech.
I'm not sure where to draw the line. There's times when voluntarily agreeing not to speeak up about something is probably good. NDAs, for example. But perhaps the time is coming when we're going to need something besides the first ammendment to protect free speech. Congress might not be able to make any law prohibiting free speech, but corporations can sure draw up contracts and suits that are doing the job.
Libertarianism is rich wolves and poor sheep playing gambler's ruin for dinner.
Remember kids, all this BS was made possible when a judge decided that loading a program from disk to memory was also considered "copying."
A current problem I'm having with SQL 7 is that it pretty much dies when pulling ~1-10MB BLOBs (in SQL server)
No user can do anything when another is querying a BLOB.
The term is UCITA.
Uniform Computer Information Transactions Act.
Learn it, know it, conquer it.
[
I'm kidding BTW..
But think of how much FUD we can eliminate if we simply add to the GPL that you can't say anything bad about Linux before getting a Slashdot review...
This is stupid... You should be able to benchmark and publish the results...
I don't actually exist.
Hmm if Slashdot folowed Microsofts idea...
New Slashdot AUP.. You can't post anything inflamitory about Slashdot w/o the permition of Slashdot Staff...
Fair? No... But identical to saying you can't publish benchmarks with out Microsofts permition...
Yes Larry is weasel.. as proof.. he has the same requirement.
You should be able to publish benchmarks of products and you shouldn't need permition to do so.
You are free to buy the product you just aren't free to say it sucks if it sucks...
I don't actually exist.
If UTICA gets passed
What does a smallish city in New York have to do with this?
--
--
"Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next.
>Yes and of course Oracle will evaluate MS SQL very fairly
Hay if we are to believe companys with a vested intrest in Microsoft (such as editors of Windows mags) evaluate Linux fairly why not..
Being more sereous... No we shouldn't believe they would..
But should we also belive that Microsoft would give permition to an unbies review?
Sure why not... yeah and Oracle can be trusted to do an unbies review...
I don't actually exist.
>It is interesting to note that the EULA for >Windows says nothing of the sort.
>If MS really is the great satan, AND they
>have a durable monopoly, why don't they?
Key word here is durable..
Microsoft is a monopoly.. Most Linux advocates will say it is NOT durable...
But a fradual monopoly is still a monopoly..
"The Great Satan" suggests no other company is treated like Microsoft...
RedHat, AoL, Sun, Oracle, Commodore (before it died), Atari, Apple, Corel, IBM and on and on and on...
You only hear more about Microsoft as so many people aren't allowed to use anything else...
For me it's just an annoyence than my tax dollers go to buy shotty software when it allready has BSD and Linux available for free. (this annoyence extends to HP-UX BTW and isn't issolated to Microsoft)
I don't actually exist.
This is pathetic. Software is defended in the courts by corporations and organizations (and often by /.ers) ad nauseum as speech. Expression. Communication.
Speech is allegedly protected in the US; this applies to any other speech, and certainly any criticism of that speech (with a few basic restrictions). So why is this forgotten with EULAs? Because time and time again, corporations are gunning to use the tide of law to grant more "personhood" status to their products and themselves than anything near 98.6 degrees farenheit. What gives? It's bad for business, it's bad for the consumer, and it's bad for government to facilitate this kind of intimidation. I suspect that this kind of attempt to develop US copyright laws will fail- it's too absurd here, although perhaps not too absurd for the Belgians.....
It would be nice if the did outright violate the liccens.. then if Microsoft did sue.. Oracle would win and BOTH liccens would be voided..
It's an obveous bash to say "Do you really want to use a product that has this in the liccens?"
but it's far to commen practace...
The suck part isn't Microsoft.. Microsoft evil evil HISSS.. but that.. for the first time... everyone really dose do it...
Microsoft wouldn't be the only ones hurt by this..
In fact I doupt Microsoft with be injured much at all if this were considered illegal..
I don't actually exist.
That's right because, as we all know, when you sit down at the bargaining table with Oracle or Microsoft, you negotiate as equals. No imbalance of power here; no sir, no way.
This is news to me. I see a lot of stuff in the copyright statutes about distributing copies and derivative works, but no mention of conditions on use. In fact, when other industries, say the publishing industry, for instance, have tried to put conditions on the sale of things like books (e.g. minimum retail price, prohibiting resale), they have gotten slapped down.
Even the software industry doesn't have the chutzpah to claim that these license terms can be justified under copyright law. Instead, they try to rationalize them as "voluntary" contracts. However, the law does, as others have pointed out, recognize the concept of "unconscionable" (and therefore unenforcable) contract terms. Do these terms qualify legally as unconscionable? It's hard to say; nobody knows for sure until you fight it out in court. I think it's rather telling, however, that for all their sabre rattling, software companies always seem to keep these matters out of court. I think it's equally telling that they feel they need to get special law passed to legitimate their license agreements.
Well, except for the minor detail that the GPL places restrictions only on copying the software, while the commercial guys put restrictions on using it and (even more offensively) talking about using it. With as many times as this difference has been pointed out on slashdot, I'm amazed that anyone who has been reading for any length of time can remain ignorant of it.
-rpl
Its not as if either party is on the right side of this issue. Sometime ago I was evaluating different database systems for a project. Postgres had a nice page benchmarking it against various other sytems. They had comparisons with Mysql, Primebase, Openbase and several others. What they did not have was comparisons against Oracle or Microsoft's server. They stated right on the page that the licensing agreement of both companies forbids publishing of benchmarks. What we have here is two repressive giant corporations trying strangle each other to death. In course of their battles they're stepping on a whole lot of little people. Meanwhile, I'm still wondering if this provision would infact stand up in court. In no other industry are their licensing agreements that forbid the independent testing of one product against another and publishing the results. If there were, Car and Driver and other such magazines would've been out of business a long time ago. Not to mention Consumer Reports. The question I would have for a judge is "What makes software different?" Actually, that's something I would have to ask on many different issues. What makes software different than other tools and devices we purchase in the real world?
It's a pity that Ellison is merely demonstrating, rather than actually publishing, the relative performances.
That means they could defend a suit by claiming they weren't in violation, rather than having to claim the clause is unenforcable.
Of course that's to be expected - since they had such a clause before Microsoft.
Which brings up a question: Microsoft apparently put the clause in in reaction to Oracle's using it. Perhaps they'll sue in order to break such clauses themselves.
Granted their current product's performance has sloth in places that allows Oracle to come up with a benchmark that makes them look bad relative to Oracle. But benchmark hacking is a well-known art. So even if Microsoft IS generally slower they can no doubt come up with benchmarks that make them look better - or buy experts to modify their products until they ARE faster. And they can afford a lot more advertising. So it may still be in their interest to kill such clauses.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
"Because you don't like MS's EULA for SQL Server, you should not buy it. If you agree with the EULA, go ahead and buy it. Larry is just weasel anyway, so why should there be any defence for him and his corporate thugs?"
Yes! We must abolish every right of the consumer to fairly evaluate every product on the market! Those who accurately depict a product, good or bad, MUST BE PUNISHED!
Oprah Winfrey vs. Texas Cattle Ranchers. Enough said.
/Brian
I do. You buy most products outright. With software, you generally buy a (revokable) license to use the product.
Conceptually, it's something like purchase as distinct from leasing (although it's leasing with a one-time payment.)
That's also why EULA's can say that you can't sell the software, or rather that the buyer can't run the software. The license is generally non-transferable.
This next song is very sad. Please clap along. -- Robin Zander
Sorry to jump OT for a moment, but this was my favorite part of the Spinal Tap movies.
As a heavy metal guitarist, I've actually seen people do things exactly like this, some even went so far as to pull off the knobs on their amps, move them a quarter turn forward, then put them back on so that they could say, "Look, I can go a full 25% above 10. I'm REALLY loud!"
I just find it funny that no one ever picks on guitarists when they do shit like that, yet everyone jumps all over someone when they pull something that makes them look like a fake 3l33t h4X0r d00dz. Like I've said before, it must be the stage lighting. I have yet to see a geek with good stage lighting.
Bite my yammer.
"I mean, if I wanted to publish database results, does this mean I can't write a report about it?"
No, you can't write a report about it. As others have noted, Oracle has this condition in its licence too.
The bottom line is that if you don't agree with the licence terms, then you don't have to use the software. As the creator of the intellectual property it is your privilege to apply whatever conditions on its use that you see fit. This applies equally to free and commercial software
In this respect, Microsoft is being no more evil than rms.
Are TPC benchmarks useful? Really when you think about it it's more a benchmark of how well the clustering and hardware performed during the test. The top spot which Microsoft holds on a Compaq machine is a 192 processor Xeon cluster with a huge number of dual processor clients etc.
:}
How usefull is this data in the real world where I don't have $10million to waste? Great when the newest processor comes out I'll run the TPC test and I'll have the top spot till someone on a totally different platform with totally different hardware does the same thing. Great very useful for me.
75% of all statistics are made up on the spot!
Trevor.
Has Ellison explicitely signed the EULA with Microsoft? Does he explicitly talk about benchmarks made without permission from Microsoft? Is not he expressing his opinion and the opinion of his company? Hey Redmond? Where is that 1st Amendement that lays in the Constitution of the country you reside? Maybe, inside a company it may not work, but this is public speech!
Besides why are you always doing things upside down? Where is the written license for Win98 on IBM comps? Russian Law demands it. So what should I do with those CD's that came with these computers? Why you violate the law by stating on your User's Guide:
"No part of this document can be reproduced or transferred in no form and no means, no matter electronic or mechanical, including photocopying, record on magnetic carrier, if there is not written permission from Microsoft"
This is a part of User's Guide for Win98. I'm citing. Do you read? CITING! And on Russian Law I have the DAMN right to do it!
Do you know what this stuff reminds of? Scientologists. Microsoft seems to be turning into another high-tech church.
The DMCA makes unauthorized reviews illegal. Unlike an EULA, it is legally enforcable in your country. If you don't like it Larry, then either take it like a man or donate some of that hard earned cash towards the EFF.
Story Time. I was driving home from the university last week, and came across a similar "contract" while shooting down I-71 at around 85mph. This one was on the back of a large truck hauling rocks:
"Remain back 200 feet. We are not responsible for any resulting damage from falling rocks."
Give me a break - These are scare tactics. If a truck dumps a load of rocks on to you, you do have a claim against them. Same with AOL: prove malice, fraud, or negligence, and you could probably sue for more than termination of your account.
43rd Law of Computing: Anything that can go wr
Not like people haven't built systems to score high in benchmarks before. Revealing this nefarious tactic is what Freedom of The Press is all about, particulary in spirit. Larry's got Bill's buds on this one.
--
A feeling of having made the same mistake before: Deja Foobar
Obviously, I'm half-joking here. Only half, as if it does state in the license agreement that you have to have permission to publish benchmarks, then they really have no choice but to sue. They will lose, as the article indicates that Ellison is _not_ publishing his results, merely demonstrating them. Others are publishing reports about his (Ellison's) demonstrations. I don't fault Ellison for making his remarks. He is the head of a rival company. I just hope MS does sue. They need to lose another case.
Eric Gearman
--
Atomic batteries to power! Turbines to speed!
OK, if you have property and lots of nice things, it isn't going to be easy to put all this stuff into jeopardy...
But - surely! - somebody with enough computer expertise to do the benchmarks
...and nothing to lose (e.g. because all of his/her gear is in a spouse's name... or owned by some untouchable trust... you get the idea)
...will eventually come along and write such a report as most of you fear to write/publish
...and will place it on some great off-shore web server, e.g. along with all those illicit copies of popular but expensive books and/or software.
Ideally, of course, someone will win a legal arguement that it's in the public interest to know the answers to questions that arise (and get answered) in benchmark reports
...or that the rights of an individual to create/publish such a report are greater than the rights of any corporation to supress it.
Oracle has the same rules. They prohibit the publication of benchmarks of their database too. I'm guessing it's for that reason that they're trying to circumvent microsoft's restriction without out and out breaking it. I've seldom seen a more odious term in a EULA.
--
First off, the big dirty secret is that databases are completely different. Even if you have a portable application, it doesn't matter because you still have to port all of your database stuff. In an ideal world, you should be able to evaluate one database, then switch to another, just like swapping in and out parts pretty easily. You could then compare databases on the basis of performance.
This is my sig.
Oracle and DB2 have similar clauses in their license agreements that forbid the publication or disclosure of benchmarks without express written consent of the company.
Yes, it's a load of crap, but it's not a "New Innovation" from Redmond. They weren't first, and they actually didn't start to put this into licenses until SQL Server version 7, last year.
AFAIK the clause about benchmarks isn't valid in europe. this clause would put too much restrictions on consumer organisations.
//rdj
No one can understand the truth until he drinks of coffee's frothy goodness.
--Sheikh Abd-Al-Kadir, 1587
Does this strike anyone else as just plain dumb? I mean it's like a small child going to mommy cause the bully told the whole class he picks his nose. IMHO Microsoft should worry less about who's saying what about what, and worry more about actually making a decent product that would stand up in these demonstrations.
I'm a bassplayer. no one picks on guitarists when they do stuff like that because we're too busy laughing at 'em; as if they were fooling ANYBODY. Everyone knows you have to actually relable the '11' on there (or '12' in the case of my SVT)
"Everyone is entitled to their own opinion, but not their own facts."
... like this about sites wanting to publish bad reviews of Dikatana? Isn't this type of action bad for /everyone/ involved, as it skews the products image favorably, when there probably is a better performing product available?
"This amp is special, see all the knobs go up to 11, that means it is one louder than other amps"
I hereby proclaim that Microsoft may not say anything bad about Linux unless they have the express permission of Linus Torvalds.
Michael
http://www.buymeasportscar.com
http://www.buymeasportscar.com
No, really! I'm serious!
It could be very interesting to watch Microsoft try to bully a company that it can't just buy.
"... the advance of civilization is nothing but an exercise in the limiting of privacy" - Janov Pelorat
Pistols at 20 paces!!
Is anyone else sick and tired of two guys with more dough than most third-world nations engage in pissing contests like this. Nobody cares. Nobody.
Well except for clueless manager types who judge the power of software by the number of zeros in the pricetag.
Blech. Signatures.
"Let the market decide" sounds to me as if the market needs the ability to choose, and the information to properly make that choice. That way, superior products and producers are rewarded, and inferior ones punished, to the Darwinian limit if necessary.
It's pretty clear that Microsoft (and others, as well) attempts to limit both information and choice.
How the heck can ANYONE from the commercial software side call the utterly ruthless Darwinian meritocracy of Linux/Open Source, "Communism"?
The living have better things to do than to continue hating the dead.
Oracle is great for scalability and reliability. However it's a bitch to administer and will turn around and bite your ass if you make the smallest mistake. Sybase has equivilent scalability and reliability and is much more administrator-friendly. DBA isn't a job for the timid or stupid - regardless of the platform you need to have a high level of technical competance. Managing a large Sybase installation can be done by a single DBA; in my experience, the equivilent Oracle installation would require 2 or 3 people.
Why is it that the proponents of "one nation under God" are so eager to get rid of "liberty and justice for all"?
Pit them against each other on Celebrity Deathmatch. The card would read: Larry "Shogun" Ellison vs. Bill "BSOD" Gates
If it's been done before, I'd like to see the vidcaps.
does mySQL even handle transactions yet?
mySQL is free, and you get what you pay for.
"If I were allowed by company xyz to tell you that their product sucks I would, but since I am not allowed, I won't tell you that is sucks"
Opus: the Swiss army knife of audio codec
Two different issues here:
1. Did Ellison actually violate the EULA. While the article doesn't provide much information, it doesn't sound like he did. To truly publish a benchmark, you've got to give specific numbers. If Ellison did that, even in his speech, he has published his numbers. If he said that Oracle has run benchmarks (and even if he ran them side-by-side and let people see), it's a big stretch to say that he published the results. He is really characterizing his company's experience in running the benchmarks -- providing conclusions that are based on the benchmark results. I don't think that's the same thing.
2. The other issue is whether the EULA provision is enforcable. I don't have the first clue on that issue, but it certainly seems suspect. After all, it comes under the category of a company disclosing its own experience with a product. One imagines something akin to telling a company that they can't tell tell anyone that a product hasn't worked well for them. I can't believe that a Court would allow that sort of restriction. The qustion is where to draw the line.
You mean 2 or 3 sybase DBA. You only need 1 Oracle DBA to administer one or more Oracle DB.
"Failure is not an option, it come bundled with the software"
We already know that a company can write a license that is *signed* by both parties that limits free speech in exchange for products/services (in the forms of NDAs). When UTICA passes, it then allows those click-thrus to be just an enforcable too, so by clicking through an install you could forfeit your rights on that piece of software forever.
I'm hoping Orcale stays on the ball and defends this all the way in court; a software license cannot trump 1st Amendment rights.
"Pinky, you've left the lens cap of your mind on again." - P&TB
"I can see my house from here!" - ST:
If you ask me too. There's a general principle in contract law (and a license agreement is a form of contract), that terms of a contract that are "unconscionable" will not be enforced by a court of law. This principle is broad and poorly defined, but is totally up to the courts in their discretion to decide, which means you have to have the resources to fight it out in court. I imagine Ellison just might be able to...
Most "unconscionable" contract terms are economically unfair, often because of severe inequity between the 2 contracting parties. However another common reason for finding a contract term to be "unconscionable" is that it is "against public policy". Many states have laws that prohibit contracts that force their residents to litigate in other states or using other state's laws, calling such terms "against public policy". A more detailed discussion (written for New York State lawyers) is here.
I think a contract term that prohibits someone from disclosing "trade secrets" is certainly enforceable. For a product that has very limited distribution, a "gag" clause that attempts to prohibit the transfer of technical information to a potential competitor is probably enforcerceable in most forms. However, in this case, any and every serious competitor has a copy of the product or can easily obtain one, and the "benchmark gag" is designed to allow Microsoft to control information available to potential customers on a critical purchase point - how well the product works. If MS changed the clause to require that the benchmark conditions be fully and fairly disclosed, that would probably be enforceable.
So were you by any chance one of these pranksters?
The reasoning behind it is that the testing is to be performed under VERY precisely defined & controlled conditions, to prevent unscrupulous people from skewing the data.
In light of the Mindcraft debacle, most slashdotters should be able to at least understand the motivation involved, if not the practice itself.
Imagine using two computers for a database benchmark, but using a *slightly* different model hard drive in one (from the very same drive vendor, even) but with a 10% slower seek time. You better bet that benchmark will be skewed, and it would be HIGHLY unlikely anyone would catch it.
Everyone knows Ellison shoots his mouth off -- a lot. On the other hand, personally, I have more respect for the guy with cojones than whiny Bill threatening to sue.
God, threatening to sue. Pathetic. That's a sure indicator, Bill's panicking.
Anyway, I work for Oracle but don't speak for them, etc.
Part of the Second American Revolution!
Where are we going, and why am I in this handbasket?
Isn't publishing benchmarks of something covered under fair use? I mean if you go to a restaurant, can someone tell you that you can't let anyone out there know how good the food is? No. Benchmarks for a DB server are excatly the same thing. I don't see how a lisence agreement like this can stand up in court. Then again, IANAL...
Ñ'
Welcome to the new war. I'm thinking that the 'war' between Oracle and SQL Server is going to be like the war between Netscape and IE. It will be interesting to see who comes out on top.
From the article, they are offering 10 million to "anyone who could get any application to run on Microsoft's TPC-C clustered-database configuration that the Transaction Processing Council." Perhaps MS should take them up on this bet.. it could be fun to see.
-Frijoles-
... alltrough both eula prohibits it i can say: MS sql sucks - oracle rocks. Not for a speed reason - for the stability reason - MSsql _WILL_ crash whenever given a reason or not, especially when the ones who must take care of them are sleeping or are out of town. or both.
--
1% APY, No fees, Online Bank https://captl1.co/2uIErYq Don't let your $$$ sit in a no-interest acct.
From http://www.spec.org/osg/jbb2000/docs/faq.html#Q14:
Q14: Can I compare my SPECjbb2000 results with my TPC-C results?
A14: No, absolutely not. SPECjbb2000 is memory-resident, uses totally different data-set sizes and workload mixes, performs no I/O to disks, and has no think times. It has a different set of run and reporting rules, a different measure of throughput, and a different metric. Such comparison would be a SERIOUS violation of SPEC's run and reporting rules and of TPC's "fair use policy." Violations are subject to penalties.
He'll state MS's benchmarks then downplay them as invalid and turn around and tout his benchmarks. Thats rediculous and just silly. This whole database war is ugly and I don't particularly care about it but Oracle is playing just as dirty if not more so than MS. I really hate their stupid 96% of fortunes e50 run oracle or whatever that commercial says. Who cares? Thats like saying 96% of fortunes e50 run notepad. That doesn't make notepad any better than anything else, just that its being used somewhere in that company. Uggh ... i think i've seen enough MS bashing for one day, then this comes along. Thanks /.
Databases are really complex beasts... You can't just take an untuned database (installed out of the box) and expect to get the best performance you can. And it's doubtful that Oracle would spend much time tuning a SQL Server installation when the goal is for the Oracle database to win.
It'd be akin to Microsoft bringing out two sealed white boxes and saying "One run windows.One runs Linux. Notice how windows is 1000% faster at typcial tasks when compared to linx?" without mentioning or letting it be known that the windows install is running on a 1 GHz Pentium III with 128 megs of ram, a gigabit ethernet card and a RAID 0 array of 4 internal disks, whereas the linux box is a Pentium 60, with 16 megs of ram, 10 megabit ethernet, and a 600 megabyte IDE drive.
Both of those computers would look identical from the outside.
For a lot of benchmarks, more independant organizations should be created and funded by all the companies who wish to have their products benchmarked, with permission given to publish their findings however they see fit. Any other benchmark, really, is suspect. As is any system installed for the express purpose of showing how slow a given piece of software is.
I see no reason to limit this restrictive concept to software. If we can apply it there, it is only fair to apply it to any manufactured product.
Non profit organizations like Consumer Reports (among others) better fight this one.
--- -- - -
Give me LIBERTY, or give me a check.
I would guess that Larry Elison will not fight this. As everyone else has mentioned, most of the larger database companies have this provision in their EULA, Oracle being no exception. If Larry fights and wins, well he hasn't done his company any big favors, as now he will have also invalidated his own companies ban as well. I would guess he would be quiet now, as he has made his point, the press has noticed, and he doesn't have to say anything more to keep this story alive. The last thing he probably wants is to have that clause of EULA's tested.
Then again, IANAL and maybe Larry Elison doesn't care.
Gavin Fischer
Live by the sword, die by the sword.
If ever there was a reason to not use either product, this is it. I've not included Oracle in several projects because of this behavior. I figure the $150k Larry didn't get doesn't matter to him but does help the competition. I guess now I need to avoid MS SQL too.
Companies that hide behind 'you don't know enough to benchmark our product' need to slither back into the slime whence they came.
But don't you think that this is interfeering with my ability to choose the right product for my application?
If MS (or anyone) doesn't let anything negative be said about their product, you're not getting the full picture.
Next time you read a review of a car, imagine if they were only allowed to say things that the car manufacturer approved. Every car would come across as a Rolls Royce or BMW. You wouldn't be able to decide till you bought one, and even then you wouldn't be able to tell your friends how it preforms.
-- IANAEG - I am not an elder god.
You open source monkeys should all be running postgres anyway.
On the other hand, if a rock is bumped off the truck due to poor roads or shocks or whatever, and dings your windshield as you tailgated the truck, you don't have a strong case - there's traffic safety issues, and the fact that the truck has posted cautionary signs reduces their liability. Now if the load was improperly secured (maybe the truck could have been covered?) as determined by applicable law, then you have a case, but otherwise, you're SOLed.
At least in this case, the 'contract' is more solid than those offered by software companies. Even in cases where the company knew that the software would crash and cause damage and did not take sufficient steps to fix that problem, (analogous to the truck dumping the rocks on you), they have been able to avoid all liability. "We are not responsible to any damage that our program may cause".
"Pinky, you've left the lens cap of your mind on again." - P&TB
"I can see my house from here!" - ST:
Lotus finds that Microsoft is touting a ridiculous benchmark study that shows Exchange a gazillion times faster than Domino in serving POP mail. Of course, methodology is bunk -- never mind the fact that no one would ever use these products in these configurations for these purposes. But headline says, simply, "Exchange a gazillion times faster than Domino." Lotus plays the cease-and-desist game, gets the claims modified and clarified but desired pain is inflicted.
Lotus marketing looks into the issue and discovers that its EULA does not prohibit this behavior. Everyone else's EULA does. Lotus changes EULA.
Chalk one up for the bad guys. We'd all like to live in a world without locks and alarms, but it ain't practical.
I'm looking over the wall, and they're looking at me!
I do. You buy most products outright. With software, you generally buy a (revokable) license to use the product.
No, you buy a physical copy of the installation media for the software, and a piece of paper with a "license agreement" that the company tries to convince you that you've agreed to, under dubious cirucmstances at best.
Contracts (such as, say, licenses) usually provide for a balance of the rights and responsibilities between the licensor and the licensee. That's why both parties have to sign a contract; if either party doesn't like it, they can try to renegotiate or, failing that, refuse to agree and therefore have no claim to the product or service being contracted for. But if you do sign, you know what you're getting yourself into.
Software companies try to tell you that you're accepting terms simply by opening the package, and that they've already signed the agreement by making you the box available to you. (And why shouldn't they sign? They have the deck stacked firmly in their favor!)
And, since you're accepting the terms of a license to a product that you haven't even looked at yet, they're going to tell you that you have no reason to expect any quality or stability from its product and that it doesn't have to work as advertised. Oh, and you can't try to inform the public of your problems with the software, or figure out how it works in case you want to fix it yourself. Oh, and that you can't sell your physical copy of the installation media to someone else if you get fed up with it. Oh, even if you are okay with these restrictions, the company can change or terminate the license unilaterally anyways.
Why would anyone agree to terms like these? This is not an "agreement"; anyone who tries to convince you otherwise doesn't have a very high opinion of you. And why should they? They got you to "sign" an agreement saying they don't have to give you anything more than you've already got, and are laughing all the bank to the bank with your money!
Given that the 99% of the license agreements out there are not printed on the back of the box in big, clear text for everyone to read before you open the box^H^H^H^H^H^H^H^H^H^H^H^Hagree to the terms -- gotta have room for all those pretty screenshots of what the software might do once you actually install it -- you're effectively being coerced into the agreement. One might be able to make a case for fraud, as well.
I hope that shrink-wrap licenses do get tested in court and are laughed out of the room. Because anyone who would try to enforce such terms on a piece of hardware certainly would be. (Are you listening, Digital Convergence?)
Jay (=
I'll use your mentality again in an example that might hit a little closer to home for you. If Natalie Portman put an EULA on her personally autographed glossies that said:
That mentality of your's doesn't fly very far now does it.
Shouldn't you be posting comments on the "Microsoft Cracked" story and calling everybody a zealot and so on? But wait, the story is all over the news channels...
--
Information wants to be beer, or something like that.
I find this issue to be slightly blown out of proportion. In the end we are dealing here with a rock thrown from one corporate giant's balcony, to another one's window.
Bill and Larry here are playing their little game...It's good to demand for criticism and freedom of speech, but frankly I'll pass on hearing comments about SQL Server if they're coming from Oracle.
Can anyone tell me (being a simple minded European), if might-be US president Bush Jr. really wants a free market and not too much government interference with the economy, does that also mean he will fight these laws?
Obviously these laws do have a large impact, no?
I have no idea why, and frankly it sucks. It's also Oracle, IBM, Informix, and the entire DB gang.
Infoworld did an artical on this few years ago, and then they invited everyone to bring in their DB and choice of hardware. IBM brought in an AS/400 box, MS with NT, and so forth. Dang, only if I could find the artical...
III.IIVIVIXIIVIVIIIVVIIIIXVIIIXIIIIIIIIVIIIIVVIII
I'd really like to see that one go to court, it's a great example of how out of control license agreements are getting. Ever read any? In the MS Office 2000 agreement, it basically states that if your business should fail due to a bug or flaw in the software, Microsoft isn't liable. Or America Online "Your sole remedy of any dispute with AOL is the immediate cancellation of your account," I got confirmation from AOL on that one, it basically means that you can't sue them, unless you can catch them in a breach of contract. I say all of this because as many would agree, license agreements are getting out of hand, many times when you break the seal on a CD case, it says that by breaking the seal you agree to the enclosed agreement ... the one that you can't read until after you break the seal.
How soon until we will be giving up our first born to these agreements?
Also, with that MS one from office, wouldn't that be something like General Motor's selling a call with a license agreement that says "If at any time this vehicle should suddenly explode for no apparent reason, GM is not responsible." Yea, right, Ford in being sued over their explorer and Firestone about their tires, such a thing would never hold up in court, and I for one think it's time for someone to challenge some of the ridiculous license agreements out there.
Lets see GM tell Consumer Reports that they don't want CS crashing cars and telling people how they fair.
Help Brendan pay off his student loans
Don't you love UCITA; it will give these rediculous clauses in license agreements even more legal footing. This is only the tip of the iceberg.
ByteMyCode.com: A Web 2.0 code sharing community.
Look, it's larry that's being a jerk. He can't take it that Microsoft owns all the top benchmarks set up by and independent (truly) testing organization that's been around a while. And if you look at the bang for buck figures, it's not even close. all TEN of the top ten are SQL server (some version) see the results yourself As for an app demonstration, I can write an app that will run faster with (piece of crap) microsoft Access than Microsoft SQL server.
---
DO NOT DISTURB THE SE
"You cannot write about your Operating system crashing without prior expressed permission from Microsoft"
And then they'd say if you don't like it -- don't use it. Linux aside, Windows has the majority of the desktop market share, and such logic will leave little choice to most of the people using PCs at home.
Ñ'