Free For All

Some writers on the Free software movement speak as if the kernel hackers, security experts and fanatical sysadmins who drive Linux and other Free operating systems are martyrs -- folks who may code out of love, but who ultimately are on what could be seen as a suicide mission rather than a milk run. The typical free software guy (and in fact, the typical software guy, period ) gets treated as a one-dimensional character, with the projects they work on reduced to meaningless blurbs. Peter Wayner knows better -- he takes to heart the notion that history is written by the winners, and proceeds to write history. Read on to see why I'm recommending his new book Free For All to my father. A correction: Theo de Raadt (whose name I had originally mispelled, sorry Theo) pointed out that I'd slipped in "Open" where I should have said "Net." Apologies to all involved in each.

Free For All author Peter Wayner pages 340 publisher Harper Business rating 8.2 reviewer timothy ISBN 0066620503 summary From-the-trenches history in the making, a survey course on how, why and when Free software took over.

Future Perfect Free For All's subtitle ("How Linux and the Free software movement undercut the high-tech titans") well expresses the attitude that Wayner lets filter through every page of this book. Wayner writes as if from the perspective of a computer historian 10 or 20 years from now, mentioning casually the tools and methods which allowed (past tense) the Free software movement to flourish as if dismissing in many cases the overwhelming dominance of closed software today. Most desktops, it's true, are running some version of Windows, and despite the popularity of Linux and the BSDs, there are still chickens left to hatch before the count. But in the 1920s and 30s, there were still plenty of horsecarts, too: Wayner proclaims that the internal combustion engine of the day is the virtual engine under the hood of our computers.

It's a forgiveable act of hubris, though, considering that Wayner also points out the plentiful high ground that Free software has newly gained, recently regained, or never lost claim to, and it's a convincing list. Slashdot readers, for instance, may know that Apache serves the majority of today's Web sites, but does the average Barnes and Noble browser, even in the computer section, know just what Apache is? This book wastes few opportunities to point out areas where Free software is the obvious best choice, not just a grin-and-bear-it low-cost alternative to something better.

Historical perspective Wayner sets most of this book in the 1990s -- the reference to Linux in the title makes that a clear and sensible decision -- but makes frequent and welcome trips back in time to temporal locations from the age of Big Iron in the 1960s to Richard Stallman's 1984 GNU Manifesto. To those of us born in the 1970s or later, these episodes serve as welcome reminders of all the history we can learn of only through such means.

To that end, the book offers details and anecdotes about the creation of the Unix and Unixlike operating systems that are on the rise now, from the post-breakup copyright battles over the original source code of AT&T Unix to the serendipitous ignorance of Finnish student Linus Torvalds, who didn't know that there already was all-but-the-polish of a free Unix system already available.

It's not the case, though, that the entire Free software community is presented as one big happily family. More like an extended family with skeletons in several closets and some bickering both around the dinner table and otherwise, but for all that a generally harmonious bunch. The issue of licensing, and of hotly debated terms which might seem to an outsider hopelessly semantic, are raised at several points. Wayner contrasts Richard M. Stallman's vision of Free software (whether you see it as humble or grandiose) with the viewpoints of Eric Raymond, Bruce Perens and others. Stallman and the GNU project seem to get the lion's share of attention, with the obvious justification that without the GNU tools, a free Unix workalike would seem like a quixotic dream.

The time-shuttling approach that he takes with each chapter brings a benefit that makes this book an easy one to put down (for a few minutes, at least) -- it means that each chapter stands as an interesting monologue on some aspect of the Free software movement, and can be read as an enjoyable short essay. Taken together though, the chapters don't just entertain and milk nostalgia from silicon: they make a good case for the premise of the title. Ironically (if you see it this way) this means undercutting some of the arguments that Microsoft is a monopoly. Perhaps Microsoft was a monopoly, but the cut is made and the tree is toppling.

Interestingly, among the copious information about the origins and present state of the various BSD projects (Net, Free, Open), Wayner speaks a good deal about the whispered-about (and shouted-about) animosity between OpenBSD project leader Theo de Raadt and the developers of the other BSD varieties. While de Raadt spoke openly with Wayner, and the NetBSD developers seemingly did not, what ermerges is a slightly more interesting picture than I've seen before about this, and it confirms some positive things I've heard about the whole OpenBSD project. (A project which I think has caused improvement in many other software projects with its unyielding security focus.)

Minor Gripes Wayner's writing is informal -- no stiff upper lip here. That's not a bad thing, but the prose slips regularly into casualisms and jargon, parts of which work better than others, but none so distracting to detract greatly from the story being told. (As if I'm one to complain about that!)

The other problem I have with the storytelling in Free For All is the litany of rhetorical descriptions of hackers which are introduced in order to refute them for no clear reason. No, not all hackers have long scruffy beards; Yes, RMS and Alan Cox do. No, not all hackers are pale and anti-social; Yes, some of them are. Maybe its just that I've heard these things said before so many times that it just doens't seem relevant any more. Perhaps many of these "human interest" elements really will fascinate readers who'd not considered them before.

Recommended Reading Who should read this book? I mentioned that I'm recommending it to my father, for the simple reason that this is one of the few books I've seen which are down-to-earth readable but still meaty enough to walk away from with a satisfied feeling, not like you've just been Dummied.

In fact, it reminds me of Stephen Levy's Hackers, in part because it shares a sense of exhiliration and admiration for the people involved, as well as a freewheeling, back-to-the-story-in-progress story telling style. The hackers who make the BSD projects run, and the Linux kernel expand and shrink as code is cultivated and reined in, may be inspired software geniuses. But they share in the unglamorous, painstaking dogwork as well as the glory, and beam a kind of virulent enthuiasm for the cool stuff they're constantly on the cusp of. Wayner gives an over-the-shoulder peek at what that means which doesn't require a C.S. degree, and serves as its own character glossary.

Add this book to the pile that includes Hackers, The Secret Guide to Computers and Open Sources for readable, fascinating, fun computer history that's also relevant for your pointy headed boss.

You can purchase Free for All at Fatbrain.

Comments

NYT article by Peter Wayner

[HugoRune]

This guy also wrote this article for the New York Times. It's pretty scathing of the WIPO and the MPAA. Good to see that this message is being published in the mainstream press.

(don't forget that if you want to avoid NYT registration there's always username:cypherpunks, password:cypherpunks)

Original ideas there are many

[tylerh]

What have this gang of people done that isn't a "workalike"

The Internet

Internet Explorer. IE started life as Mosaic, one of the original browsers. Like all of the origninal browsers, Mosaic was open source. Microsoft bought the browser idea from its Open Source inventors.

Apache. This is the direct descendant of the original web server (it too was open source), and it dominates the web. Microsoft has tried to copy Apache's functions, but has had a tough time keeping up with Apache's pace of innnovation.

sendmail . Essentially all of the email that goes across the internet does so thanks to sendmail. The orginal (open source) developers now also run a company, but the orignal accomplishments all happened open-source.

BIND The Internet works on IP addresses (eg. 135.23.43.121). Any time you type a URL (letters) into your browswer, you are using BIND. This was invented open source (the B is for Berkely).

TCP/IP These are the two protocols (among others) that make the internet possible. In a sense, they define what is "internet." Developed and implemented open source

Eric Raymond addresses "creativity" issues in his essays.