Clearly you can recognise the fact that IBM promotes a lot of Free software, whereas Microsoft does not.

The same could be said of Novell, yet some people still consider it legitimate to engage in exposition of those issues where their choices are questionable.

The issue-driven agenda is similar to ours (ignoring brands, handling law/conduct), but you cannot leave out any company’s history altogether.

One also shouldn’t blinker there focus to only those areas of computing which interest them. IBM has been friendly towards Free Software on the desktop, but has their history with regard to the mainframe arena exhibited a similar level of freedom advocacy?

I would caution against admonishing those who advocate software freedom on the mainframe merely based on IBM having made significant contributions towards Free Software on the desktop. I certainly disagree with Mr Mueller and Mr Maynard on several issues, but a company’s contributions to the Free Software community only goes so far towards excusing their more disfavorable and exclusionary conduct.

The GPLv3 may give you rights to get a GPL work and run it through a defective trash compactor (eg, run it on Windows even if Windows is not behaving as expected) if that is what tickles your fancy; however, compiling GPL using a proprietary compiler (eg, a Microsoft compiler) which produces object code you can’t otherwise derive and then distributing without the corresponding source (to include instructions on creating that object code) is not allowed.

I’ve seen projects that use Microsoft tools to build the Windows binaries and then put these binaries up for download. This is going against the letter and spirit of the GPL by distributing something without sufficient source code to allow you to predict its behavior.

OK, I wrote a followup about reasons for distrusting Florian, at least from the FOSS POV.

I’d like to think I am issue driven as well, but there are multiple issues in this saga that I think deserve attention as a FOSS supporter.

I’m still waiting for you to give balance to your attack on IBM by attacking Microsoft for moving in on new territory and for using patent threats against many open source users.

>> If they only wanted to get rid of software patents because IBM has so many of them, that wouldn’t play any role.

Florian, Microsoft takes out more software patents than almost any other company.

Let me know the day they come to you saying they want to abolish software patents because their position is quite the opposite as revealed by deed, by their briefs in support of them in the Bilski trial, and by internal memos (eg, from the Comes exhibit).

I agree except that I’d go further. Remove all the GPL software from Windows, and watch Linux and other open platforms become a lot more attractive to more people. Notice that Microsoft is trying to get all GPL to work on their platform and then is additionally supporting a class of open source projects that get special rights and capabilities only when run on Microsoft’s own platform. Microsoft knows that they have to win the application game (get them to work on their platform, exclusively if possible) in order to preserve their proprietary platform.

A more important point is that open source over a proprietary platform that is in active development is not open source. The user doesn’t have control or predictability except as the closed platform vendor sees fitting to their business plans.

I think the GPL might not be acceptable in these cases where the underlying platform is not standardized (eg, because it is in flux and with features not publicly documented) nor open source. [I'll have to reread the GPL later. I mentioned this above "GPLv4".]

You stated you find mixing open and closed source to be acceptable. It’s different to use a proprietary application here or there with mostly open source, for example, than it is to use a proprietary platform since, in the latter case, everything becomes proprietary by virtue of the substantial dependencies on the proprietary platform that every single piece of software running on it has.

You can isolate and lock down and limit what any proprietary application can do when it’s on an open platform like Linux, but you can’t have control of even a single application (except to the degree tolerated by the platform vendor) when you are on a closed platform like Windows.

It’s difficult to know what RMS had in mind without knowing the exact words and more context.

I am not against antitrust enforcement were warranted, but it should not be replacing one abuser with a greater one that is more of a threat to open platforms.

I am wondering why you aren’t speaking out, as a FOSS supporter, against the growth of monopolist proprietary platforms into new territory?

It’s this inconsistency, as a FOSS supporter, that is troubling other FOSS supporters who look at your actions.

>> If action is taken against them, it’s a clear signal to Microsoft and its OEM partners.

That is correct. You can’t send a signal to Microsoft that they will be allowed to go into new markets as an excuse to weaken other players that happen to be strong in those markets.

Microsoft is (or should be) restricted because of their position with their Windows platform. This is the signal we should send them, in addition to any other signal we send to IBM and anyone else.

Did you mean to imply that Microsoft is for open standards that are fully specified while companies like IBM “actually” are not?

>> The European Commission’s decision in the Microsoft case addressed both trade secret and patent issues, and I said I wouldn’t want that new European interoperability law to stop short of what was achieved in the Microsoft case.

I think I looked over that decision. I am not a lawyer and don’t deal with Microsoft products.

I did get the impression that the decision did not require Microsoft to document all aspects of the protocols or full semantics of their API.

>> I think that EU initiative will also be real litmus test because some of those who advocated the principles we (as opponents of software patents and proponents of FOSS) cherish only when they viewed it as a means of attacking a competitor will now struggle with the idea of having to open up, too.

IBM has problems with a proprietary platform being pushed by Turbo Hercules (the company). You should too when you consider that this would be a growing of power of an abusive monopolist.

I think, as a supporter of FOSS (as you just claimed you are), that you should attack Microsoft on par with your attacks against IBM. Microsoft is trying to grow their proprietary lock-in and reach into new markets.

>> because in the worst case a weak outcome due to lobbying by the likes of Apple and Adobe could also result in more Microsoft behavior of the kind none of us would like to see.

Are you saying that Microsoft, despite the EU taking care of them as you appear to think is the case, is a threat if Adobe and Apple are not similarly restrained?

Or are you saying that Apple and Adobe pose the same threats to consumers today as does Microsoft?

I want to make this perfectly clear: my agenda is issue-driven (hence my consistent positions over all those years), not company-specific.

RMS and I opposed Oracle’s acquisition of MySQL and to neither of us the fact that Microsoft also made submissions critical of the deal (see the European Commission’s decision, which contains lots of quotes from them) was a reason to change our position on the subject. RMS’ positions and mine were driven by FOSS concerns. That doesn’t mean we were wrong only because two proprietary software companies (Microsoft and SAP) also presented arguments against antitrust clearance from their (different) angles.

It’s the same for me now with the mainframe antitrust (abuse of dominance) case. I’ve made a lot of effort to look into the mainframe market. In addition to studying a lot of write-ups from both sides of the debate, I have held conversations not only with the founder of the Hercules project but with many others, including Bernhard Kaindl, a fellow anti-software-patent activist who developed the mainframe version of SuSE Linux, which is still the market-leading z/Linux distribution. I have concluded that strengthening Hercules (and the Hercules-based TurboHercules offering, which is equally available for GNU/Linux) does not in any way harm the interests of FOSS. On the contrary, I believe it strengthens FOSS directly as well as (because of the issues involved that are also relevant in other markets) indirectly.

Over time I will certainly explain that on my blog in greater detail to address concerns some may have due to IBM-driven propaganda.

Being issue-driven rather than company-focused is the only way to have consistent positions and the credibility it entails. That’s what I’m known for in Brussels, and I would never want it any other way.

If Microsoft called me and said they’re sick of all those patent troll lawsuits against them and they’ve understood (again, because a long time ago they knew) that software patents are a bad thing, would I close my door? No. I would say, “welcome to the club, what can we do together for this good cause?” And I would then ensure that I don’t become a gun for hire but that I would have all the autonomy in the world to pursue what’s right. I wouldn’t accept a muzzle of any kind. If they wanted to specify a kind of framework for the positions I take, I would tell them to read my Creative Commons-licensed ebook, No Lobbyists As Such.

If they only wanted to get rid of software patents because IBM has so many of them, that wouldn’t play any role.

But if they asked me whether I could support software patents or downplay the essential role of FOSS to ensure effective competition in the software market, I wouldn’t make myself available.

I believe I have now made a good-faith effort to explain my approach to the important issues that have been raised.

I see a need for antitrust intervention against IBM because I see exactly those principles at stake.

To keep it focused, let me just use the quote above from you on “leveraging” as an example. IBM leverages and thereby reinforces, expands and exacerbates its mainframe monopoly by disallowing the use of z/OS on non-IBM hardware. If action is taken against them, it’s a clear signal to Microsoft and its OEM partners. If IBM gets away with it, think of the implications this may have for Microsoft’s dealings with hardware companies.

The term “open standards” is actually used much more aggressively by companies who are actually against the concept when their own core business is concerned, such as IBM.

I certainly don’t want trade secrets to stand in the way of interoperability. The European Commission’s decision in the Microsoft case addressed both trade secret and patent issues, and I said I wouldn’t want that new European interoperability law to stop short of what was achieved in the Microsoft case. Elevating that principle from a case-by-case decision (which is court-validated but only by the Court of First Instance because Microsoft didn’t appeal to the ECJ) to a real law would be major progress with respect all companies in the industry, including but not limited to Microsoft. I think Apple and Adobe are particularly good examples of companies that need to open up and under the current legal framework they might have strong points to make that they’re not “dominant”.

I think that EU initiative will also be real litmus test because some of those who advocated the principles we (as opponents of software patents and proponents of FOSS) cherish only when they viewed it as a means of attacking a competitor will now struggle with the idea of having to open up, too. Fortunately, the first case was about Microsoft (and supported by Adobe, for instance) and not about them. It will be very hard for them to oppose Neelie Kroes’ initiative and I guess they will do it through backroom lobbying. We should all keep an eye on that and it makes that legislative process (which has not yet formally begun, but it’s on the right track in this “prenatal” state) a critical mission because in the worst case a weak outcome due to lobbying by the likes of Apple and Adobe could also result in more Microsoft behavior of the kind none of us would like to see.

By definition, patents are broad monopolies (anti-competition), and, for numerous reasons, this is particularly stifling to many individuals and unworkable for the creation and distribution of increasingly sophisticated software products.

Software is a product of the mind, without the bounds nature places on material goods. Patents on fiction or on math, for example, would also be very disruptive to society and advancement.

This is saying that prior to Microsoft trying to leverage software patents very much, think back 10 or more years (eg, to the US DOJ cases from the middle 90s and Microsoft earlier abuses), that they did not present anti-competition problems.

The major anti-competitive tool for their extensive platform has been trade secret. For example, the inevitable constant flux of semantic changes that occurs as the software is changed and grown, and the integration only available to their applications.

Another major problem is leveraging the large size of their distribution platform (Windows) in order to strike deals with re-distributors which lead to biases against competing platforms (eg, Linux).

Again, Neelie Kroes recognizing that the EU government should move to open platforms is excellent, as it solves numerous problems and potential complications from quasi-solutions.