Some heat, some light

So, here is this post by Hugh McLeod over at GapingVoid that by itself is not so interesting, but manages to generate an interesting discussion. So do check it out. In part, that discussion centers on the supposed value that proprietary vendors (its about Microsoft originally, but his general point applies more broadly) provide, which makes them the preferred choice even when software is provided free of cost. [Basically, its the question, why FOSS? all over again] I'm not sure I have anything to add to that discussion, but I noticed this article by Matt Assay, which contains the following very useful lines (which are of direct relevance to software web services that run on top of free software):

I dislike the implication: the only way to make money from open source is by being a parasite. If the only way to make money from open source is by co-opting others' work (and giving nothing back) and spewing ads at them, then let me off the bus now...I don't believe that this is the only way to make money with open source...I like GPL-based models where you monetize the software directly. But I'm willing to accept the Web 2.0 angle, provided that Web 2.0 companies stop treating open source like a free good to be plundered, and rather as a valuable resource to be replenished. It is in their interests to do so. Google gets this better than most, though it has done little to replenish some of the core projects from which it derives value (e.g., Linux). My hope is that the Web 2.0 world will recognize that open source will only be available to "plunder" to the extent that these companies, like IBM and others before them, give back. Generously. Not out of charity, but out of self-interest.

Just the facts on FOSS market share

You will notice a link to David Wheeler's FOSS references page on this blog. Its quite an amazing collection of articles and well worth checking out. On that page, Wheeler has, for a long time, had an article which, in my opinion is quite possibly the most comprehensive quantitative study on FOSS market shares. As Wheeler puts it,

My paper “Why OSS/FS? Look at the Numbers!” is a massive collection of quantitative studies on free software, with the goal to “show that you should consider using OSS/FS when acquiring software”. It has a large set of different studies grouped into the categories market share, reliability, performance, scalability, security, and total cost of ownership.

The good news is that Wheeler has updated that piece, (first put out in 2005). Check it out (and read Wheeler's post over at the Free Software Magazine for what has changed in the meantime).

GPL and commercial advantage

I have a feeling I've talked about the commercial advantages of the GPL on this blog before, but I'm too lazy to go take a look in the archives. In any case, here is an interesting piece that reiterates that claim (one of the surefire certainties about the writing in this area is that the same points will be made again and again, the same canards spread, the same refutations made - this particular repetition suits me, so I indulge in it). The piece's narrative follows a general trend. It notes how some commercial vendors prefer licenses other than the GPL because the GPL is "too restrictive"; it then goes on to point out that the reciprocity requirement of the GPL actually makes tons of commercial sense, because it requires those that draw upon you, to give back. Here is an interesting excerpt:

Enterprise Content Management vendor Alfresco recently decided to move from its Mozilla Public License based (which is similar in many respects to the Apache license) to the GPL. Instead of seeing the GPL as being something that would handcuff their business, Alfresco has taken the view that it will actually help their commercial business. Matt Asay, vice president of business development at Alfresco, told me that the risk of a forked version of Alfresco's GPL code would be a positive thing for the company. "We would be ecstatic if someone forked the GPL version of Alfresco because then they get to go off on their fork and develop their own system but we would also benefit from the work that they do," Asay commented. "If we can't compete based on the work that we're doing on our own code as well as benefiting form the work that a fork would do on theirs, then we don't deserve to be in business."

I've read very similar sentiments expressed in a post by Matt Assay (provocatively titled "Microsoft + GPL = Match Made in Heaven"), which, while including language about "inhibiting competitors" that I find problematic, includes this gem:

Microsoft needs to ditch its weird view on the GPL. It used to call it anti-American. It’s actually the exact opposite. It is the most American of open source licenses. Microsoft could embrace it and continue to pull in its billions…and what could be more American than crass materialism?

A word on gNewSense

I use Ubuntu at home and work, which isn't a completely free distro, and have often debated the question of whether this sets back free software (mind you, I maintain a dual-boot machine that has XP on it, which I use for watching some streaming video that requires, grrr, the dreaded IE/MP combo, so I'm definitely not pure as driven snow). Still, I've often dreamed of going to a completely free distro, so recently, I went ahead and asked at the Free Software Magazine how distros like Blag, Ututo and gNewSense were doing. I received a very useful reply from another reader, which I reproduce in full (in case anyone is thinking about going, like, totally free):

I'm using gNewSense 1.1 at the moment, and it works really well. I definitely recommend downloading the live CD and trying it out. There's nothing I can't do with it that I couldn't with Ubuntu (unless I really wanted to download the proprietary ATI driver in Ubuntu Dapper, which let me have 3D acceleration but caused various other problems). I did buy a new wireless card (the Linkysys WUSB54G v4) so it would work well with gNewSense using 100% free software, without any proprietary firmware. To get my old one working with Ubuntu I had to download and compile the driver anyway (which contained proprietary firmware), and it worked terribly. This one works fine out of the box in gNewSense and Ubuntu. I haven't tried the other fully free distros such as Ututo, but I've heard they don't work as well. Judging by gNewSense though, I expect that's because of factors other than a lack of proprietary drivers.

Here is the page where I asked my question, which features an interesting discussion in its own right, about the propriety of using not-completely-free distros. More on that later.

From a plugin to innovation

So Microsoft's Open Source Labs has sent in a Media Player plug-in fix for Firefox (it "[it] shows another level of interoperability and eagerness in working with the Community"). You can check out the features offered (and of course, there is the gentleman in the comments who asks, "This is cool, but is there some chance of Microsoft creating an official Linux version of the Windows Media codecs so that people using Linux machines can view content on webpages that is encoded with WMx?"). This reminds me of yet another conversation in Seattle with folks from Port 25. I had asked the question on whether as part of their strategy of "working with" or "contributing to" the "community" they planned to send in code to FOSS projects. At that time, the response was whether or not I counted all the other instances of helping FOSS applications run on Windows, or the interoperability work counted as "contributions". They do in one sense, but the most fundamental contribution, and one that would help Microsoft best understand the culture that they seem to quite deeply misunderstand, would be to immerse themselves in it, and allow themselves the luxury of stepping away from thinking of it in terms of only business and technical models, and instead to think about it as a genuine passion for people. Thinking about it as a passion doesn't mean regarding them as nuts, or "extremists", or any other loaded term, but instead to try and understand the nature of the motivations that could have led to the FOSS dimension in its current shape. Innovation is driven by so many different motivations that it is simplistic in the extreme to imagine that it always takes capital to get it rolling (and shouldn't the copious literature on FOSS put out by business departments have convinced them otherwise?). MS often seems worried that not only will their business collapse if they were to go open-source, they just wouldn't get any code written. I don't know about their business (but I suggest they'd be surprised by how friendly the GPL would be to them) but I know that on the latter count, they needn't worry. Code will get written - and a lot of it will be very good. Writing code is a creative act, and the mysteries of how that act of creation kicks in, is one we're still working on. What we do know at the moment is that the answer most definitely is not "if no capital gets mobilized, nothing gets created"

Predictions galore

Oh, no, heres one more "the open-source bubble is going to burst" story. And guess what, to make this even more credible, its an open-source CEO (a successful one) speaking. But what does he have to say? This:

"Right now, open source is hot," says Rod Johnson, author of the Spring Java development framework and CEO of Interface21, the company he founded to market it. Most open source projects are supported by an army of volunteers who buy into the hype, but "capitalism will inevitably reassert itself" and developers will find they need to put more effort into steady jobs and private lives, leaving "open source zombies"--unsupported, unmaintained projects--he predicts.

Really? Whats the argument here? Or is this just prophecy? "The hour draws nigh when capitalism will assert itself". Newsflash: capitalism has already asserted itself in the world of free software and open source (like, maybe 1998?). And, "open source projects are supported by an army of volunteers who buy into the hype"? (Methinks this reporter has brought into the hype - so in that sense the subject and reporter are in sync). And I'm mystified by what it could mean that developers will need to put "more effort into steady jobs and private lives"? All those people working for open source businesses must crave the kind of job security the rest the corporate world has (like those 17,000 folks Citigroup laid off last week). Finally, just to strike fear into all of us, there is the prospect of "open-source zombies". Scary. But at least we know they are out there. What about all those closed-source zombies, confined to the campuses of their proprietary headquarters? Who knows what they're capable of?

Another take on the MTS-07

Pia Waugh blogs on the Microsoft Technology Summit. Pia has done a good job of providing the kind of synoptic take I was planning to write, so go ahead and check it out. I'd commented on her blog but somehow that comment was lost, and in any case, it was inadequate so I'll try to write more here sometime soon.

You can't get away from 'it'

So, as might have been evident from my past few posts, I attended Microsoft's Technology Summit from March 26th to the 28th. I posted a blogroll a few posts below so that you can see what other attendees blogged on live, and on some of their feedback. I also posted a link to an interview at their Open Source Labs (full disclaimer: my trip was paid for). But in this post, what I want to do is take a slightly higher-level look at what I think MS was trying to achieve with this summit. In its most basic form, it was to get the word out to technologists about what they were up to. Yes, its all smoke and mirrors, but in any case, its worth commenting on.

Some problems: First, this attempt at communication took place in a form which too many attendees at the summit felt was too half-duplex. Too much talking by Microsoft folks, and not enough amongst the attendees, (or listening to them), or facilitating a structure that allowed for more discussion. All of this got said to MS on the second day itself. Secondly (related to the first), part of MS's agenda at this summit appeared to be to present a kindler, gentler face to that part of the FOSS community they think they can do business with: "FOSS - FSF zealots" was the subset MS seemed to want to work with. The only problem with this approach was that often the way in which MS folks spoke seemed to indicate that any discussion of FOSS outside of software engineering or business models (and within that they only seemed to want to talk of revenue streams as being exclusively license-sale driven) makes them nervous as it might be what those "open-source philosophers" talk about. While I can understand the standard reason for being so leery ("like, we might get infected by left-field ideas") surely, MS cannot imagine that they can successfully co-opt FOSS without attempting to tackle adequately that part of FOSS ideology that is not covered by software engineering or business models. Or can they? Related to this was a question that I put to folks at MS's open source labs: do they plan to contribute to FOSS projects? It might seem wierd to them, but ultimately, if MS wants to learn about what makes FOSS tick, they'll have to jump in head-first.

"You can't handle the kernel!"

So, one point of contention (amongst many others) that came up in the course of a conversation with Sam Ramji, who heads Microsoft's Open Source Labs, regarded the usefulness of access to the code. As Ramji put it, "You want the kernel? You can't handle the kernel!". I've heard this one before, and its fairly simple to express. Don't hype access to the code when most users (insert your favored percentage here), simply can't code, or even if they didn't wouldn't want to inspect the code, and even if they did, wouldn't be competent enough to understand it, and even less competent to fix it. Fair enough. But what substantive point about free software follows from this, and does it lessen the normative weight of the request made in the Free Software Definition? I don't think so. Note that access to the code is a precondition for two out of the four freedoms (freedom 1: to study the code and adapt it to your needs, and freedom 2: the freedom to modify the code and release improvements). Now, note that the competence argument that is being deployed here simply says "you can't use the freedom to study the code and adapt it to your needs, because you aren't competent enough to - and concomitantly, access to the code also means nothing if you aren't able to modify the code". But again, the access to the code is merely a precondition to these freedoms; nowhere is the claim being made that access to the code is sufficient for code study and improvements. Those latter activities will only follow upon inspection and modification of the code by competent programmers, and here is where the crucial point lies (pardon me if its obvious to you): with code available to me, I am free to take my ignorance elsewhere (including presumably to the original programmers) and to ask more competent folks to help me. What access to the code most importantly secures for me is the freedom to pick and choose (within some reasonable bounds) who fixes my code and when.

As I indicated in my interview at the Open Source Labs, there is a free market out there in the free software world. Its a free market for programmers, able to bid on coding jobs with their eligibility determined by their portfolios (or some other demonstration of excellence).

Israel and the GPL

A detailed note on an Israeli case involving the GPL. There are very few legal proceedings involving the GPL, and every one of them is worth their weight in gold for the attention they focus on licensing issues, the chance to revisit the terms of the license and their respective implications. Even if this case is not of the dramatic "will-the-GPL-survive-this-latest-challenge", its useful in all these ways.

A trio on the GPL

RMS, Torvalds and Novell comment on the new version of the GPL V3. No surprises in any of what got said in there (except, perhaps, Torvald's less-than-flame-like-response).

Interview at MS's Open Source Labs

While visiting Redmond for the Microsoft Technology Summit, I visited Microsoft's Open Source Labs (the Port 25 you folks might have heard about), and was interviewed by Anandeep Pannu. Here is the link for the interview (you'll notice a reference to that part of my life that has to do with aviation history - check it out if you like). I spent most of the chat talking about the book, trying to come up with a reasonable synopsis of each of the chapters. I also got into a feisty conversation (later that night) with Anandeep's boss at the labs, and will try and post my summary of that particular interaction a little later.

Blogroll for MTS-07 Participants

So as indicated in this blog a little while ago, I was invited to the Microsoft Technology Summit, which ran from March 26th to the 28th. Comments later, but first, a blogroll of some of the participants.