Special Treatment [LWN.net]

Special Treatment

Posted Oct 17, 2007 6:37 UTC (Wed) by man_ls (subscriber, #15091) [Link]

So you raised the issue of duplicativeness on the OSI lists. How come the Microsoft licenses are not duplicative? What is novel about them? (and please don't say "brevity", we are all adults here and can read reasonably fast).

Also, you speak as if Microsoft Corporation were the only feasible users, and their codebases the only feasible application, of Ms-RL and MS-PL. Were Netscape Communications the only conceivable users, and Mozilla Communicator the only conceivable use, of MPL, when it was written?

Again, it all hinges on whether the Microsoft licenses are duplicative (which they seem to be). The MPL takes an original approach to Free software (or Open Source if you wish), thus enabling its use for corporations which wish to have a weak form of copyleft. On the other hand, duplicative licenses which are widely used have their own merit (like Apache's, which is similar to BSD). But duplicative licenses with virtually no users have theoretically no place in OSI lists.

Anyway, I think I can reasonably speculate about Microsoft Corporation's intended use of those two licences: They'll experiment with the licences' use for maintenance of non-GPL commonses, and as a trial balloon for open source within that firm generally.

You are too kind. I can imagine a couple of worse things, like announcing: "Microsoft distributes MS Windows (C) (TM) (R) under Shared Source (TM) (R), which is a form of Open Source (OSI approved)". A convoluted way to associate their OS to Open Source, but one which might work in certain circumstances (propaganda for CIOs, procurement policies poorly drafted). And part of the usual Microsoft smokescreen for their illegal activities, like abusing their monopoly.

Our only consolation in such a scenario will be to cry "I told you so" in a moderately loud tone. OSI members will be denied even that.

Special Treatment

Posted Oct 17, 2007 14:53 UTC (Wed) by rickmoen (subscriber, #6943) [Link]

man_ls wrote:

So you raised the issue of duplicativeness on the OSI lists.

No, that is not what I said. I said I was one of those who commented on that point, and was among the minority who felt the licences are nothing particularly special. As to what other commentators felt, if you want to see what they said, Read The Friendly Archives.

As an aside, please note that duplicativeness in a licence is necessarily a matter of degree -- and also of opinion.

Again, it all hinges on whether the Microsoft licenses are duplicative (which they seem to be).

Hey, you have an opinion! You'll do well on the Internet (especially if you continue to omit your real name).

I can imagine a couple of worse things, like announcing: "Microsoft distributes MS Windows (C) (TM) (R) under Shared Source (TM) (R), which is a form of Open Source (OSI approved)". A convoluted way to associate their OS to Open Source

I believe you mean "to make them look really cheesy, and, more important, to risk bad publicity through further and particularly egregious public abuse of the term open source, in blatant bad faith towards OSI". I think they're too smart to do that, but obviously we'll have to wait and see.

To pass the time, I'd be willing to put a small wager on that matter (after all, I'm not sure they're that smart), but am uncertain offhand how to word it.

Rick Moen
rick@linuxmafia.com

Special Treatment

Posted Oct 17, 2007 18:04 UTC (Wed) by man_ls (subscriber, #15091) [Link]

No, that is not what I said. I said I was one of those who commented on that point, and was among the minority who felt the licences are nothing particularly special.

Sorry for the mistake. In fact your sentence is similarly worded but certainly different (raised a question vs raised the issue).

As to what other commentators felt, if you want to see what they said, Read The Friendly Archives.

:D I read the discussion for MS-PL, and this is what I gathered.

Chris DiBona: "Finally, why should yet another set of minority, vanity licenses be approved by an OSI that has been attempting to deter copycat licenses and reduce license proliferation?", posing the question. Brian Behlendorf: "I don't know yet if there has been an explicit rejection of a license up for certification, so I don't know if we've yet established how different a new license needs to be", so the issue is not clear. John Cowan: "I think (as I thought two years ago) that this is a case where the anti-proliferation rules should be set aside", effectively asking for special treatment. Bill Hilf: "There are already several hundred community projects that use these licenses, including over 150 Microsoft projects", making it a beautiful circular argument. I did not in fact find any messages from you speaking about redundancy.

On the discussion for MS-CL there was nothing I could find. In short, the issue was not discussed at great length because it was not considered relevant.

Hey, you have an opinion! You'll do well on the Internet (especially if you continue to omit your real name).

I don't see what my real name has to do with the issue: as far as I can see my driver's license might read "Man Ls, sq" and it wouldn't change the fact that I read the licenses and they did not seem to offer anything new.

I think they're too smart to do that, but obviously we'll have to wait and see.

Have you never heard Ballmer speak in public? He will stoop as low as necessary to make business. I mean, they only need to know their target customers; and in that they are immensely smarter than either you or me, witness their commercial success.

To pass the time, I'd be willing to put a small wager on that matter (after all, I'm not sure they're that smart), but am uncertain offhand how to word it.

Let's see. If Microsoft ever publishes the words "Windows", "Shared Source" and "Open Source" in the same sentence then I win, and you will pay me a beer next time you visit Spain. Let's make it two years; after that you win.

Special Treatment

Posted Oct 17, 2007 21:39 UTC (Wed) by rickmoen (subscriber, #6943) [Link]

(On the mistaken characterisation, not a problem. Glad to clarify.)

I read the discussion for MS-PL, and this is what I gathered.

That was definitely a fraction of it, but certainly not all. I vaguely recall there having been a bit more, as the thread arose several times.

You'll probably notice that substantive discussion of proposed licences on license-discuss tends to get slightly swamped in (1) sparring between two or three licensing trolls delighted at the attention and everyone else, (2) net.randoms dropping in and posing questions about real and theoretical licence problems, and (3) various people arguing about gods-knows-what varieties of disputes regarding copyright, contract, trademark, and patent law.

Also, it's not unknown for a licence to get proposed and garner almost no comment from anybody, presumably because either the proposer or the licence text just didn't interest people. For example, Socialtext's radically toned down badgeware licence (CPAL) got exactly one objection, from Matthew Flaschen, that the mandated display of original-developer info (if the technology supports such a display) for "sufficient duration to give reasonable notice" still arguably violates OSD#3 (Derived Works) because logo accumulation through code reuse could become unreasonable and impractical. I answered: "I do agree with the concern about logo accumulation in derivative works, and have voiced it myself, but feel that some sense of proportion should be applied, in judging the proper extent of that concern. Yes, an important aim of open source is to make code borrowing lawful and practical (to the extent that licence compatibility allows), but, realistically, Nicholas Goodman's logo-overload scenario is (even under the early badgeware licenses with fixed-size logos on every page) a severe (yet amusing) over-exaggeration of what could ever actually happen in the real world."

OSI's Board approved CPAL without a great deal more discussion, and (IIRC) its report cited my comment and not a lot else. Partly, this was because nobody had much else to say. Partly, it's because the related issues had already been thrashed out in earlier discussion of the preceding MPL 1.1 + Exhibit B licences. My point is: Sometimes, you find few on-list comments because the proposal in uncontroversial, other times because it's just not interesting, other times because most relevant discussion occurred earlier in only peripherally related threads.

Thus, the best way to understand the general thrust of license-discuss discussion is to read it on an ongoing basis. Looking up The Friendly Archives is very useful, but might miss some.

Anyway, Matthew Flaschen and I (among others) explicitly opined that both Ms-RL and Ms-PL are clearly OSD-compliant, and did so both before and after one of those licences was renamed from its earlier naming. (Thus my point: You didn't find that.)

John Cowan opining that "the anti-proliferation rules should be set aside" for Ms-PL doesn't signify, for, among other things, the fact that OSI has no such rules: It has only an intent (and subcommittee) to develop some. (Note also that Cowan didn't opine that he considered Ms-PL duplicative within the meaning of the subcommittee's recommendations. He just felt that a procedure that he mistakenly believed to exist should be waived.)

I don't see what my real name has to do with the issue...

Well, other people very likely do. I expect to have to stand behind what I say, and therefore try to take great care (human frailty, limited patience, and caffeine deficiency permitting) to make sure I can. When others, like you (but far less so than many, to be fair), start asking me from behind cover of pseudonymity to answer their carelessly mistaken misrepresentations of my posts, I think it only fair to point out the asymmetrical nature of our discussion: You can take sloppy, somewhat disreputable potshots at my posts with impunity, create the expectation that I'm somehow obliged to disprove them, and if necessary walk away clean from your nick if it ever proves embarrassing. I can't; I'm posting as me.

But I didn't mean to harp on that (it's just because you asked, and besides, I gladly believe you that it was accidental), so let's please move on.

Have you never heard Ballmer speak in public? He will stoop as low as necessary to make business.

This would be a relevant objection if I'd said that Microsoft Corporation (or Steven Ballmer) were ethical. I did not. I said I said I believed them to be "smarter than that".

I don't think I should have to do all of your homework, so, consider taking a few minutes figuring out what you'd do if you were OSI President and observed that Microsoft Corporation were going around invoking OSI's name in public attempting to suggest that its proprietary software is in some fashion open source. Novice PR problem.

"Memo from Joe Sixpack, Microsoft maintenance division, Madrid: Windows is mentioned in the company's Shared Source Web pages, and is not open source. Please send a copy of this memo to Rick Moen, and advise him that 'man_ls' and I will be each having half a beer at his expense." ;-)

Rick Moen
rick@linuxmafia.com

Special Treatment

Posted Oct 17, 2007 22:19 UTC (Wed) by man_ls (subscriber, #15091) [Link]

It would really make me very sad to part with this pseudonym (apart from the fact that I have paid for the privilege, being a subscriber and all). In fact I'm the same old man_ls / Alex Fernández you had a little riffraff with a couple of years ago. That is precisely why I said that my real name doesn't matter: you didn't remember it from last time and that's perfectly fine, because who cares.

OK, you think that it would just be a PR matter that can be dealt with, and let's hope it is so. We have all seen Microsoft do worse things (as in their infamous antitrust cause), and we have seen how it helps their friendly press and their astroturfing. We have also seen it fail (as in the Office Open XML fiasco). Maybe it will not work next time, or maybe they will not even try. It is true that e.g. with Windows CE (which has a "see but don't touch" license) it has not worked at all.

Anyway the licenses are approved, so we can only wait and see. And pay that beer if you ever come to Madrid and I can show you an unambiguous case of :D

Special Treatment

Posted Oct 17, 2007 23:43 UTC (Wed) by rickmoen (subscriber, #6943) [Link]

Alex, I honestly do appreciate that reminder.  And you're of course right, that people's
reasoning should absolutely stand or fall on its own merits, and in that sense (among others)
"real names don't matter".  FWIW, I've actually worked hard to protect the right to comment
pseudonymously or anonymously longer than I'd care to remember.  And I'm pleased to report
that we are, in fact, overall in total agreement (not that an honest difference of view isn't
OK, too).

In particular, your point is well taken about your stable "nick", and there are many I admire
greatly, including, say, "Fyodor" of nmap fame, who I doubt writes grocery cheques in the
Dostoyevsky character guise he adopts for software purposes.  ;->  If I've seemed unduly
suspicious, I do ask your pardon, and hope to remember you better next time.  (As you'll note
in the thread you linked, I've seen pseudonymous personae used for some extremely scurrilous
things, especially for ankle-biting personal attacks against better-known open source people,
so perhaps you can understand my wariness.)

Rick Moen
rick@linuxmafia.com