Wednesday, June 27, 2007

Just another business model, eh?

A little follow-up to the spat brewing between the OSI and CRM vendors that I blogged about yesterday. As I see it, the OSI is bothered by the failure of CRM licenses to provide some crucial - and here I'm going to use a word that the OSI didn't want to touch with a ten-foot pole - freedoms, that the OSI considers an important part of the concept of 'open source'. The irony of this situation is immense. The OSI sought to distance itself from 'ideological tubthumping', 'free software zealotry' or whatever else it was, to make the case for 'open source as a purely technical, pragmatic, business strategy. It was all about access to the code, because that, you see, just helps us write better software. But 'free software', the conditions associated with which are stated quite clearly in the Free Software Definition was to be disdained because that wouldn't be business-friendly enough (they don't like talk of values and freedom apparently). But being business-friendly has its downside as the OSI is discovering. Maybe if they had kept their values front and center, they wouldn't have needed to have dragged them out now, pointing to them, and asking for community policing to try and block the path of those corporate vendors, who staggered into this space, enticed by a new business model, and now just want to exploit it for all its worth. It never was just about having access to the code so that more folks could debug it. And this story tells us why.

Tuesday, June 26, 2007

What's in a name

There is a wierd little battle developing over the term "open source", between the Open Source Initiative, and a small group of vendors (CRM folks like Sugar et al.) that would like to use the term to describe their products, despite their licenses not being OSI-approved. Slashdot is talking about it, and unsurprisingly, focuses on Michael Tiemann's furious blog posting, which responding to Dana Blankenhorn's post querying how far 'open source CRM' could get from OSI-license terms, included this blast:
..there really is not room in the market for Yet Another Proprietary CRM system....THESE LICENSES ARE NOT OPEN SOURCE LICENSES. This flagrant abuse of labeling is not unlike sweetening a mild abrasive with ethylene glycol and calling the substance Toothpaste. If the market is clamouring for open source CRM solutions, why are some companies delivering open source in name only and not in substance? I think the answer is simple: they think they can get away with it.
Now, the OSI does not have a trademark on the term 'open source', it simply maintains the Open Source Definition, and a list of approved licenses that meet that definition. So, this battle, to say the least, promises to be interesting. Can the OSI pull off this particular kind of policing? (Perhaps it might have been easier if they had stayed with a corporate-unfriendly term like 'free software'? But I digress). The 'problem' is that licenses like the CPL is that as David Richards explains
the CPL limits EXTERNAL redistribution. That's pretty much it. So while our source code is available and is for free, while we allow it to be modified and the modifications licensed however the authors dictate (think BSD), and while it can be freely distributed internally by user organizations, we run afoul of the OSI because we do not allow, without our consent, external redistribution. partners.
And there are other restrictions, all of which fly in the face of the OSI's requirements of leaving certain decisions on modifications to downstream developers.

There is a reason behind RMS's obsession over terminology; it matters.

Tuesday, June 19, 2007

Its all just one, really

Tom Chance pens a thoughtful piece on Lawrence Lessig's views on the 'two culture economies': supposedly, an amateur, "sharing economy" and a second, 'traditional', full copyright "commercial economy" (how this "commercial economy" got to be the 'traditional' one is beyond me). If that distinction sounds dubious to you (and like it might have internalized, unwittingly, the very dichotomy that the restrictive culture that Lessig thinks he is contesting, wants to promote) then read the piece for yourself. I found Lessig's _Free Culture_ extremely persuasive when I read it, and its hard to reconcile his view of two separate creative economies with the vision that seemed to run through that book - that all of culture, not some subdivision of it, was reliant on sharing, cross-pollination, and innovation based on relentless derivation. I found that analysis exhilarating, and often used the examples of borrowing lines one heard on a train in one's writing or joke-making or conversation-starters in my classes in Computer Ethics (at Brooklyn College) as entrees into an extended discussion on copyright and patent issues in software. That vision, I'd argue, would reject the kind of dichotomy that Lessig endorses (and Chance's piece is an extended take on that riff, as it argues for a kind of interdependence and inter-relationship between the two that in the final analysis renders that kind of distinction meaningless to begin with).

Sunday, June 17, 2007

A stat and a confirmation

Buried in this news piece about the Linux Foundation Collaboration Summit, organized by the Linux Foundation, which featured discussions about the Linux support for embedded systems, power-saving, those patent deals, real-time Linux and so on and so forth, is the kind of nuggety technical stat that tickles my nerdy sensibilities, and I just know I'm going to be using it a party sometime soon:
Starting with release 2.4 and then 2.6 of the Linux kernel,Linus Torvalds and company have been issuing updates every two to three months. "We add 2,000 lines of code a day to the Linux kernel. We work on 2,800 lines of code a day," said kernel developer Greg Kroah-Hartman. "I've never seen the pace of change that Linux has shown."
While that "work on" line is a little vague, this passage is followed by some rumination on the kind of testing that happens on the kernel (not enough regression testing), and the way in which the structure of the Linux development model relies on its user base:
There's a tension between introducing new features and stabilizing them...what we really need is for the user community to help us track down bugs...The user base is far bigger than the number of kernel developers

Friday, June 15, 2007

Another deal?

Microsoft wants to strike a yet another 'we-wont-sue' protection deal with Red Hat, but so far, the only interest is on their end. Red Hat's response:
We continue to believe that open source and the innovation it represents should not be subject to an unsubstantiated tax that lacks transparency
I'm finding this series of 'protection deals' increasingly bizarre; Microsoft must realize the terrible PR this is creating (on the other hand, they haven't worried so much about terrible PR in all these years). If, as Microsoft claims, the issue is interoperability, then surely there are plenty of other ways to address it - like the kinds of technical solutions that folks on both sides would be willing to work on. Why are these deals needed? How does striking them address the so-called 'intellectual property' issue that Microsoft claims to be worried about? Are they going to sign deals with each and every commercial Linux enterprise? But perhaps the most damning questions is:
if interoperability and protecting customers are Microsoft's ultimate goals for the deals...why is the company so interested in Linux companies that ultimately don't affect that many customers?
The damning answer to that, comes from Bruce Perens who says:
Here is Microsoft out collecting the losers in the Linux business and paying them money so they can...paint open source as music pirates out there using Microsoft technology without a license...I think they're out to scare people.
I hadn't realized Microsoft had effortlessly segued into the horror movie business.

Thursday, June 14, 2007

Can I sit in?

Jonathan Schwartz invites Linus to dinner (in response to Linus' comments on GPL V3 as a possibility for the Linux kernel if OpenSolaris goes that route). But more importantly, his blog post, in doing so, sparks off a blizzard of comments that make for superb reading and remind me once again, of why the FOSS community is so much fun to observe. Everyone's heart is on their sleeves, the flaming is high-class, the points made are sharp, the technical content is high and some of the most important issues that hold the community together (reciprocity in this case) always make an appearance. Check it out. (I found Linus' comments on the kernel mailing list excessively cynical but your mileage may vary).

Wednesday, June 13, 2007

A change of heart?

Well, well. I didn't think it was possible, but Linus has (gasp!) actually admitted the possibility of the Linux kernel moving to GPL V3 - thanks to OpenSolaris. While he still remains a bigger fan of GPL V2, this self-confessed pragmatist offers reasons grounded in precisely that philosophy for his change of heart:
If Sun really is going to release OpenSolaris under GPL 3, that may be a good reason" to move Linux to the new licence, Torvalds said in a posting to the Linux kernel mailing list on Sunday. "I don't think the GPL 3 is as good a licence as (GPL) 2, but on the other hand, I'm pragmatic, and if we can avoid having two kernels with two different licenses and the friction that causes, I at least see the reason for GPLv3.
Of course, technical issues could still interfere in the proposed intermingling of Solaris and Linux code (whose advantages include "adding Solaris' ZFS (Zettabyte File System) storage software or DTrace probing utility to Linux or adding Linux's broader hardware support to Solaris"). One last splash of cold water from Linus:
They'll not be releasing ZFS and the other things that people are drooling about in a way that lets Linux use them on an equal footing...To Sun, a GPLv3-only release would actually let them look good, and still keep Linux from taking their interesting parts, and would allow them to take at least parts of Linux without giving anything back...And don't get me wrong: I think a truly open-source GPL 3 Solaris would be a really, really good thing, even if it does end up being a one-way street as far as code is concerned!
While not everyone in Open Solaris wants to move to GPL, the fact that Sun has done it before with Java means it isn't impossible either.

Tuesday, June 12, 2007

Affero is out for discussion

The first discussion draft of the GNU Affero General Public License (GNU AGPL) is now available. Based on version 3 of the GNU GPL, the AGPL includes terms to ensure that users who interact with software over a network can receive the source for that program. The ability of webservices provider to not provide source has long been a contentious issue (noted on this blog previously, and remarked on by Tim O'Reilly as "pockets of proprietary opportunity"). Now comes the free software community's response to this issue. I expect the discussions on this to be quite interesting largely because the question of how the user "interacts" with the software in question is a tricky one (how direct such interaction should be remains vague) and also because it seems web services would provide greater opportunities for circumvention of the AGPL's terms. Still, with the draft out for discussion, some of these issues should get sharpened. With web services on the up and up, something like the Affero was going to be required eventually, and it poses a very particular challenge to get its terms clear and legally workable.

Tuesday, June 05, 2007

Eben holding forth has been running a series of video clips of an interview with Eben Moglen. I'm linking to just one of them as you can pick up the rest from the same spot. Check them out. Moglen speaks like, as someone said to me once, "Moses come down from the mountain". He clearly seems to think he has the gospel, and I'll say one thing, if preachers were as passionate and made as much sense as he did, I'd be religious by now. (OK, I'll stop with the religious comparisons, I know it makes folks nervous, what with FSF zealotry and all, what?). Anyway, Eben holds forth on all the fun stuff: hacking the Novell-MS deal, free software, meeting RMS, monopolies and common-based production.

Sunday, June 03, 2007

RMS on why V3 is good for you

RMS writes pretty well and here he is, doing his best to explain why upgrading to GPL V3 is a good idea. There are some useful clarifications in here (I'm not sure whether it will do much for those folks that have very fundamental disagreements with V3 but it will definitely help those who are confused by some of the discussion that the new version has generated). First, RMS points out that despite incompatibility between V2 and V3
GPL version 2 will remain a valid license, and no disaster will happen if some programs remain under GPLv2 while others advance to GPLv3.
He then goes on to explain why this incompatibility should not be such a nuisance, for
license incompatibility only matters when you want to link, merge or combine code from two different programs into a single program. There is no problem in having GPLv3-covered and GPLv2-covered programs side by side in an operating system.
He then goes on to explain its stance against Tivoization and DRM. And closes with a brief discussion of that patent deal:
Microsoft made a few mistakes in the Novell-Microsoft deal, and GPLv3 is designed to turn them against Microsoft, extending that limited patent protection to the whole community. In order to take advantage of this, programs need to use GPLv3. Microsoft's lawyers are not stupid, and next time they may manage to avoid those mistakes. GPLv3 therefore says they don't get a “next time”. Releasing a program under GPL version 3 protects it from Microsoft's future attempts to make redistributors collect Microsoft royalties from the program's users.