Groklaw's Sean Daly has been busy getting more information for us about the recent announcement by Opera that it has filed a complaint with the European Commission against Microsoft. He's done two interviews. This one is with Thomas Vinje, the lawyer for the European Committee for Interoperable Systems [ECIS], who is helping to represent Opera before the Commission. You can listen if you wish to the audio [Ogg], and we have a transcript also. You can get the Audacity player to play Ogg audio files here and yes, it works on Windows and Macs too, and it's licensed under the GPL.
You'll likely recall Sean's interview with him immediately after the European Court of First Instance in September announced it had upheld the EU Commission's finding [PDF] that Microsoft had abused its dominant position.
The second interview Sean has just done for us was a group interview with Opera's CEO Jon S. von Tetzchner, Jason A. Hoida,
Deputy General Counsel, and CTO, Håkon Wium Lie. We hope to have that for you very soon.
I definitely understand much more about the complaint and why Opera decided to ask for relief, and I think you will too. By the way, notice that it is a complaint filed with the EU Commission, which will in turn decide whether or not to investigate and act; Opera has not sued Microsoft in any court. This isn't like litigation between two companies. That is only one of many misunderstandings that hopefully these interviews will clear up for everyone.
Mr. Vinje explains, among many other things, how this complaint could benefit everyone, not just Opera, if it is successful. Perhaps it has already had some success, eh? I read that Microsoft just announced yesterday that an internal test build of Internet Explorer 8 passed the Acid2 test, one test for standards compliance. It's completely unrelated to Opera's complaint, they say, and that may indeed be true. But as you'll see in the interview, compliance with the Acid2 test is one thing the Opera complaint specifically asked for. It's still the case that Microsoft didn't always have standards compliance even as a vital goal. And, without wishing to be unkind, some of us would like to verify with our own eyes.
Mr. Vinje mentions the Acid2 test, and so here's a little bit about the test for any who may not know, and you'll see how Microsoft expressed itself in the past about it, from IDG's Nancy Gohring's March 2006 article in Infoworld:
Opera initially proposed the creation of the Acid2 test as a way to highlight the lack of support for some standard HTML (Hypertext Markup Language) and CSS (Cascading Style Sheet) features in Internet Explorer (IE) and other browsers, said Håkon Lie, chief technology officer for Opera. The Web Standards Project (WaSP), an independent group that supports the use of standards in browsers, created and hosts the test.
To pass the test, a browser must be able to accurately display the test page. Opening the page with IE reveals a large red block with some spots on it. A browser that passes the test, however, will display a smiley face and the words "hello world."...
If more browsers support the features in the test, designers will be able to create better sites, he said. Developers often feel they can't use some available tools because they aren't supported in IE, the most widely used browser, Wie said. "There's a reluctance for developers to use features that aren't widely supported. The least advanced browser holds things back," he said.
Last year, Chris Wilson, a developer working for Microsoft on IE, wrote in a blog posting that Microsoft wasn't planning to ensure that IE7, the next version of IE that is currently available as a beta, could pass the Acid2 test. He describes the Acid2 test as a broad wish list of browser features that goes beyond standard CSS and HTML.
For historians and purists like me, let's look at the 2005 Microsoft blog entry referenced, to make certain we are quoting accurately:
In that vein, I've seen a lot of comments asking if we will pass the Acid2 browser test published by the Web Standards Project when IE7 ships. I'll go ahead and relieve the suspense by saying we will not pass this test when IE7 ships. The original Acid Test tested only the CSS 1 box model, and actually became part of the W3C CSS1 Test Suite since it was a fairly narrow test – but the Acid 2 Test covers a wide set of functionality and standards, not just from CSS2.1 and HTML 4.01, selected by the authors as a "wish list" of features they'd like to have. It's pointedly not a compliance test (from the Test Guide: "Acid2 does not guarantee conformance with any specification"). As a wish list, it is really important and useful to my team, but it isn't even intended, in my understanding, as our priority list for IE7.
As you see, this issue isn't something that just fell off a truck on to the highway before a started Microsoft. It's been an issue for web developers and users for a long time. If you think it's just Opera that cares about this, go to Google and search for:
ACID2 "Internet Explorer" site:msdn.com
You'll see the many, many requests that IE comply with standards, so it could pass the test. And way back when IE 7 was released, Microsoft said it knew it wasn't compliant with standards, but that they'd try to get there. There was a 'but' however:
We understand that we are far from being done and we know we have still a lot of work ahead of us. IE 7 is a stepping stone in our effort to improve our standards compliance (especially around CSS). As an example, in the platform we did not focus on any proprietary properties – though we may try out new features in the future using the official –ms- prefix, following the CSS extension mechanism. We also work very closely with the W3C CSS working group (which I am a member of) to help clarify assumptions in our implementation and drive clarifications into the spec. I really like to thank everyone who helped us here.
So, almost three years later, they have, they say, internally gotten an IE 8 test build to pass the test. But as they themselves pointed out, that doesn't prove compliance across the board with all web specifications. How that will affect the Opera case is obvious on that prong, but of course, it depends on Microsoft actually releasing a product that verifiably does pass the standards test and is meaningfully compliant with other web standards and not later tweaking IE so when folks using other browsers arrive at websites, they'll actually be able to see and use them as intended, properly rendered.
If you think *that* isn't a problem, you are not me. I deal with it all the time and have for years. I don't know if the EU Commission knows it, but there are web sites that offer video, but only if you use Microsoft's Media Player, which I don't, because of the EULA and the privacy implications. So it isn't just that consumers couldn't buy Microsoft's operating system without Media Player bundled, it's also that due to Microsoft's dominance, web developers sometimes forget about accessibility for the non-Microsoft-using world. And there are more of us all the time.
So that is one of the issues Opera is bringing to the EU Commission, that consumers are being held back by Microsoft's long history of extending standards with proprietary alterations/additions/tweaks and refusing or neglecting to support web standards. Mr. Vinje also speaks about Silverlight and I really hope the EU Commission looks into what bundling Silverlight could do to the Internet.
The other issue Opera is raising, of course, is bundling, which is what makes it possible for a monopoly to ignore standards others would like them to use. While it's unknown if the EU Commission will care about standards as an antitrust issue, the bundling issue is very much more straightforward, as Mr. Vinje explains:
An Interview with ECIS's Thomas Vinje on the Opera EC Complaint,
by Sean Daly, December 18, 2007
Q: Sean Daly reporting for Groklaw, I'm here with Thomas Vinje who's representing ECIS, and we're here today to talk about Opera and more specifically, the complaint of Opera against Microsoft that was made to the European Commission. Perhaps you could just tell us a couple words about the complaint?
Thomas Vinje: Well, the complaint raises two issues. One is the bundling of Internet Explorer with Windows, which the complaint asserts is contrary to the principle established in the judgment of the European Court of First Instance of the 17th of September with respect to tying. So in other words the complaint states that the tying of Internet Explorer with Windows is unlawful under European law. And that's based rather squarely on the CFI's judgment of September 17th.
The second claim relates to standards noncompliance and the claim there, as I understand it, basically is that Microsoft has participated in several standard-setting processes related to browsers, for example with respect to cascading style sheets -- just to take one example, and there are not that many standards at issue here -- and it has participated in the formulation of the standard, it has indicated within the standards-setting organization that it would implement the standard, and it has publicly stated that it was complying with the standard, but the implementation has not been forthcoming, and indeed, IE is in many respects far from compliant with standards and adopts proprietary nonstandardized Microsoft approaches to the technologies that are standardized. Whereas other browsers, Opera included, Firefox as well are -- I couldn't necessarily assert that they are 100% compliant with standards, I don't know that -- but they certainly are far more compliant with standards.
So Microsoft, by participating in the standards-setting process, by stating that it will comply with them, has induced other browser producers to produce browsers that are compliant with the standards, whereas Microsoft has not done so because Microsoft obviously has a very heavily dominant share of browsers. And website developers and web application developers target IE and they therefore target their websites and their applications to the noncompliant "IE mode" and that means, then, that except to the extent that they are able to have a sort of "IE quirks mode" that is achieved, as I understand it, largely via reverse engineering, other browsers such as Opera do not work properly. In other words, websites and web applications won't work properly on them, because they are targeted to the IE proprietary noncompliant mode rather than to the standards-compliant mode that others browsers use. And that then advantages Microsoft, disadvantages the other browsers, and increases to some extent the applications barrier to entry that protects Microsoft's Windows monopoly.
Q: Well, I think I can understand the first point. The Court of First Instance ruling was very clear about what constituted tying by Microsoft. It was, I think we could say it was a landmark decision. Opera is Norway-based -- Norway is not in the EU, but they are a European company, I can understand why they would prefer to file a complaint here in Europe rather than in the USA. But as far as the standards go, it seems to me like a much stickier subject to get around. For example, the devil is in the details when we talk about standards. We're talking about CSS version 2 for example. Or we're saying OK, the Document Object Model, or we're talking about HTML version 4.1. And there's a certain level of technical expertise required just to understand what standards are being referred to and which standards Microsoft should be coding to according to the complaint. How does the complaint address that? Is the complaint very explicit about which standards are concerned?
Thomas Vinje: I believe the complaint is pretty explicit about which standards are concerned, and I understand that there is a test called the Acid2 test which tests, I understand legitimately, accurately, and completely -- the extent to which a browser complies with existing standards. And so, I believe that the approach they suggest is that the Acid2 test would be one which could be the benchmark for determining standards compliance. So it's not sort of pulled out of thin air, and it's not vague or uncertain, as it's presented, as I understand it. It's rather concrete, and the boundaries are pretty clear.
Q: Now, as I understood, Opera is not asking for monetary damages -- they're merely asking that Microsoft comply with the standards upon which they've contributed, participated. Is that the case?
Thomas Vinje: That's correct, yes.
Q: And how would this compliance be measured? Would these tests be the benchmark, or what would be the best way to say, well, IE8 that will come out in X number of months is compliant or not? How would you respond to that?
Thomas Vinje: I guess the suggestion is indeed that this Acid2 test should be the benchmark that the Commission should employ in determining compliance.
Q: Aside from the tying of Internet Explorer and the extensions, the Microsoft-specific extensions, the proprietary extensions, are there other examples of Microsoft conduct that Opera feels is damaging?
Thomas Vinje: I believe that it's limited to the tying and to the standards noncompliance. I'm not aware, at least, of any other aspects of Microsoft's conduct related to browsers about which Opera has complained to the European Commission. There may very well be other aspects of Microsoft's behavior that they're not particularly happy with, but I'm not aware of them, and I don't believe they've been complained about.
Q: Now, if a person uses a non-Microsoft browser which is not Opera, it could be Firefox or Safari, could there be any benefit to them from this case?
Thomas Vinje: Yes. Absolutely. I would think that any favorable resolution of this case for Opera would redound to the benefit of the producers and the users of all non-Microsoft browsers. If Microsoft is required to comply with standards, then users of Firefox and users of Safari would not encounter as many situations where websites, web applications didn't work properly on their browsers, and there would be a considerably more level playing field for competition. That would have some effect upon the applications barrier to entry protecting Windows.
And then - so I would suggest that consumers broadly would be benefited and, you know, I think one doesn't -- I think it's uncommon for anyone to file a complaint anywhere for totally altruistic reasons. I would have difficulty believing that if anyone were to assert it. But Opera has stated that it is doing this on behalf of consumers for the reasons I have suggested, and that rings true to me based upon my discussions with them and simply based upon the content of what they have suggested and asserted, and it seems to me that they are indeed pursuing claims which, if accepted by the Commission, would substantially benefit consumers, beyond users of the Opera browser.
Q: Now, if I wanted to play devil's advocate for a moment, I could say a browser came along two or three years ago and stated that they would someday like to have 10% market share -- that was Firefox. And when that came out, everybody said "It'll never happen, it's a 96% IE world, that won't happen." And yet, today, we're in a situation where Firefox has made substantial gains and apparently directly against Internet Explorer. And yet, the information I have is that the Opera brower has had fairly flat growth over the past few years. Now, Firefox, as far as I know, is a standards-compliant browser. Why would Firefox be gaining market share against IE and not Opera?
Thomas Vinje: My understanding is that Firefox has indeed made some substantial progress. I think it's perhaps about 14% market share today. Having looked at the numbers, it appears to have plateaued there, perhaps even lost a bit of share recently. And those who are more expert in this area than I have suggested that Firefox has gained traction largely among users of Open Source software, and those who are "geeks," and technically adept, and not the majority of Microsoft software users. And it certainly seems from everything I've been able to read and ascertain that it's highly unlikely that Firefox will go much further in terms of achieving market share, and indeed, again, it seems to be perhaps losing a bit.
And Microsoft retains over 80% as I understand it of the market, and it has obtained that market share through what I would certainly suggest is unlawful conduct through the tying of IE with Windows. I have a very, very difficult time believing that Microsoft would have anything approaching its current market share absent the tying with Windows. So even if IE has made some -- excuse me, Firefox has made some marginal gains by virtue of certain circumstances which apply to it and a certain segment of the market, that does not in any way negate the fact that Microsoft has achieved these dramatic market shares as a result of illegal conduct.
Perhaps I could just remind you of -- I'm sure you recall them, but I could state them nonetheless -- the criteria identified by the Court of First Instance for unlawful bundling. And they are that there are two separate products, and it's very clear that Internet Explorer and Windows have been, and remain, separate products. I don't think that that can be disputed.
Secondly, the tying product, in this case Windows, must be in a dominant position. With 95% market share, that can't be disputed, and indeed the Court of First Instance found it to be so, and Microsoft indeed admits it to be so.
Thirdly, there must be coercion in the sense that it's not possible to obtain the tying product without the tied product and indeed here it is not possible to obtain Windows without also obtaining Internet Explorer. And finally, there must be foreclosure of competition. And pretty clearly, there has been a rather dramatic foreclosure of competition, in the first instance, of course, vis-a-vis Netscape, and Netscape was, of course, by far the leading browser, the pioneer in the market really, and it was excluded effectively from the market, and I think that it is pretty indisputable that it was excluded as a result of the tying behavior. I think that anybody, frankly, who argues contrary to that is making a claim that it's difficult to make with a straight face.
Q: Now, what would be, to your mind, a good remedy ordered by the European Commission, assuming that they agree with Opera? For instance, we saw three years ago that there was the unbundling of the Windows Media Player and big surprise, none of the retailers decided to stock it and even though it's theoretically on sale, I have never succeeded in finding it on sale, XP-N. So what do you think would be the best remedy possible?
Thomas Vinje: That's of course a very important and a difficult question, and clearly, neither Opera nor the Commission nor anyone interested in a restoration of competition in these markets would wish to see that sort of ineffective remedy adopted again.
I think one needs to distinguish between two types of situations. One being the situation with respect to products that have been tied unlawfully for a period of time and which have become very heavily subject to the network effects that result from such tying with a monopoly platform, in this case Windows, and in light of those network effects, to really open the market for browsers, for example, up to true competition on the merits. So that there would be head-to-head competition, on the merits, between Opera, and Firefox, and Safari, and IE, and the tying -- the aftereffects of the tying to Windows would not affect the competition in that market -- you know, that's very, very difficult to achieve at this stage in light of the longstanding tying of Internet Explorer, all the investments that have been made by PC manufacturers surrounding it with respect to add-on software, and all sorts of things. You now, the network effects have taken hold very strongly in many ways.
So, you know, would an unbundling -- a total unbundling of IE, so that in other words Windows were offered only without IE -- that's a possibility, at a price differential, in other words, to OEMs there would be price differential -- would that work? Perhaps. But in circumstances where there has been longstanding unlawful tying, I would think that it would be appropriate to consider a "must-carry" remedy as one that might have a chance of undoing to some significant extent the adverse effects for competition of the longstanding unlawful behavior.
With respect now, just turning, looking at the issue more broadly, as I think is important to do, beyond browsers and beyond the Opera complaint, to the issue of tying more generally, there are serious issues on the horizon. Silverlight is one of them. Silverlight is of course a runtime and rich internet applications are run on Silverlight. Adobe has announced a competing product to Silverlight which is cross-platform. In a sense, you might say that there's at least a possibility of Microsoft facing a situation in the near future similar to that it faced in '95-'96 with the advent of the Netscape browser, where computing might move to the Internet via rich internet applications on a cross-platform basis and that would render, would commoditize the underlying operating system and that could really truly open up the operating system market finally to genuine competition and threaten Microsoft's monopoly in the operating system market.
So, you know, Silverlight is not bundled today, but one could foresee the possibility of an exact rerun of that which occurred with respect to IE: to bundle it with Windows, to bundle it with Office, potentially, or other similar mechanisms, and that could eliminate competition from competing products and face off that potential threat to Microsoft's platform monopoly. So with respect to that sort of tying, potentially a true unbundling remedy might work because the network effects haven't yet taken hold because the unlawful tying hasn't yet occurred and so one doesn't have to deal with aftereffects; longstanding aftereffects in the case of IE, of unlawful tying. So I think to make a long story perhaps a little shorter, one needs to distinguish perhaps between the two situations of products that have been tied for a long time and products that haven't yet been tied.
Q: All right. What is the next step going forward from here? A complaint has been given to the Commission, they will look at it, is there a timeframe? Have they said "We'll get back to you"? What's the next step?
Thomas Vinje: Well, the Commission has no deadlines for such things. It's up to them. You know, they certainly are under pressure from their Commissioner (laughter) to deal with things quickly, and so, you know, they presumably won't dilly-dally forever, but they have no particular deadlines. The only thing the Commission has said publicly is that it's received the complaint and that it will study it carefully, and it added words which it usually doesn't add -- because that's usually all it says, it says, "We've got it, and we'll read it" -- it added, it said, "We will study it carefully, especially in light of the recent jurisprudence" -- in other words, the CFI case. And indeed, it seems to me rather difficult to dispute, actually, that this Opera case is not on all fours with the Windows Media Player case. The boxes that had to be ticked and were found to be ticked by the Commission and then the Court of First Instance in the Windows Media Player case are, I think, indisputably "tickable" here as well. So it seems to me, to use American parlance, to be rather a slam-dunk case in that regard. That doesn't mean the Commission will address it immediately, but it certainly has the legal tools available now to dispose of it rather rapidly so it wouldn't be surprising to see it go fairly quickly. But I have no inside knowledge on that and just can't make any predictions on that.
Q: One last question. Since the CFI decision in September, there were negotiations between Steve Ballmer and Neelie Kroes, she announced that there would be compliance. What's your level of information concerning the Microsoft compliance? Is there good news, is the protocol documentation coming? Is it positive so far?
Thomas Vinje: I'm afraid I really have no information on that score. There are remaining problems with the license agreements that need to be worked out and the potential licensees have been vigorously trying to obtain agreement from Microsoft to certain necessary changes to the license agreements which frankly shouldn't be very controversial, but have not yet been agreed. And frankly, it hasn't been possible yet really to discuss them. So that remains an issue. And only when those issues concerning the license agreements have been resolved will it be possible for the potential companies who, the companies who might potentially be licensees, I think who intend to be licensees if things work out -- for them to look at the documentation and see whether it is indeed substantially better and how close it is to being fully compliant with respect to accuracy and completeness. So I don't have much news on that score, I'm afraid.
Q: All right, Mr. Vinje, thank you so much for taking time with us.
Thomas Vinje: Thank you very much.