I attended the O’Reilly Emerging Technology Conference, or ETech, once again this year, and, distracted by other projects, did a very poor job of blogging about it. (You can read about the excellent EFF-sponsored debate between Mark Cuban and Fred von Lohmann, on the YouTube/Viacom lawsuit, here and here; Raph Koster spoke about magic as the underlying structure of game-play; and Danah Boyd gave a wonderful talk titled “Incantations for Muggles,” about the relationship between technologist-wizards and the rest of the human race — Koster took notes on it.)

The conference, as you may have heard, was abuzz with discussion of the Kathy Sierra saga — she’d been booked as a kickoff keynote speaker, but cancelled at the last minute, understandably spooked by threatening comments posted on her site and a couple of other blogs.

Sierra’s plight set off an immediate and vast blogstorm. There was much introspection and self-questioning about the onslaught of invective, nastiness, vicious taunts and obscene threats that sometimes emerges online, and seems especially targeted at women; there was also something of a rush to judgment to point fingers at particular bloggers whose sites and posts might (or might not) have encouraged the posts that caused Sierra such grief.

A prodigious number of people seemed to feel they had to weigh in immediately on this ugly situation, though virtually no one (yes, including myself) seemed willing or able to take the time needed to explore, in detail, what had actually happened and who had done what. I still haven’t seen any fully reported-out piece on the events — the coverage in the S.F. Chronicle seemed creditable, but it didn’t unravel the toughest questions: who was stalking Sierra, and was there in fact any relationship at all between said stalker(s) and the well-known bloggers she called out in her wounded post?

Sitting in a conference without the time or resources to do any reporting of my own, I thought, shoot, there’s no way I can know enough about what happened to add anything to the conversation. Of course comments like those Sierra encountered are, and should be considered, beyond the pale; Sierra deserved sympathy and support. But the storm of anger and the rush to judgment her post sparked represented, I thought, a failure of forethought. Running a blog provides the constant temptation to shoot off at the mouth. Sometimes, though, when you just don’t know all the facts, considered silence is golden.

The irony here is that this was supposed to be ETech’s year of fun and games.
(more…)

Last night’s Berkeley CyberSalon panel was great fun — thanks to Eric Allman, Chad Dickerson, Lisa Dusseault and Jaron Lanier for sharing their thoughts, and to everyone else who showed up with their time, their questions and their trenchant comments. (I had not known that both Jaron and Lisa have very recently become parents, and if I had I might not have presumed to ask them on the panel, but I’m very glad I did, and grateful that they put aside their parental duties long enough to participate.)

One of many highlights for me was the moment when Eric hauled some old inch-thick booklets out of his briefcase — one contained the entire set of Internet (then Arpanet) protocols from the late ’70s, bound in a mere single volume; the other was a complete list of the network’s users. My, things have changed.

There’s a downloadable podcast/MP3 file of the whole event if you’re curious.

This coming Sunday, March 4, come on down to the Hillside Club, where, from 5 to 7 p.m., at the Berkeley Cybersalon, we’ll be talking about why software is (so often) so hard.

For this event, practically in my backyard and as part of what must be the longest-running technology salon around (I started going over a decade ago), I didn’t want to just give another spiel about Dreaming in Code.

Instead, it’s a panel discussion, with Eric Allman of Sendmail (program and company), Chad Dickerson of Yahoo’s Developer Network, and Lisa Dusseault of CommerceNet (and formerly of the Open Source Applications Foundation).

It was originally scheduled for yesterday, 2/25, but we realized that counter-programming against Oscar Night was foolhardy — even for an event as geekish as this.

This Web 2.0 conference is almost exclusively focused on the business end of the phenomenon. And the defining characteristic of this iteration of the tech-industry business cycle is that virtually no one other than Google has gone public. It’s as if the excesses of the dot-com bubble left the very term “IPO” tainted. It’s also the case that this time around, the market, very sanely, isn’t that keen on supporting IPOs for companies that haven’t demonstrated profitability. So the “exit strategy” of choice today for startup companies isn’t to go public; it’s to be acquired by Google or Yahoo (or maybe AOL or Microsoft).

Yesterday evening, Barry Diller advised a questioner who asked how you could “build value” today: “Don’t sell it. Just ride it. Equity is built by holding on. Sometime you gotta sell a little of it. But hold onto it if you have something of value.” (Here’s more on Diller’s talk.)

This advice has its limits, however. A successful Web service start up reaches a point, if it manages to attract millions of users, where it has to start getting really good at things like datacenters and customer support. Maybe that’s not what the founders are interested in. Or maybe it’s dauntingly expensive. At that point, selling out to a Google or Yahoo makes perfect sense. These companies are explicitly and unashamedly in the business of doing outsourced R&D for the big guys. That’s “building value,” too.

But staying independent is more fun. Look at GoDaddy CEO Bob Parsons — a colorful ex-Marine whose “I’m just a dumb guy who flunked fifth grade” serves as cover for a shrewd business mind.

Parsons told the conference about his near-IPO experience earlier this year; he said he spent $3 million preparing his SEC filings and courting bankers, only to discover that the bankers and the financial press were focused exclusively on “short term accounting paper profits.” Parsons is a believer in operating cash flow instead. He’s proud of his company, with its 920 support reps actually answering the phone when customers call in about the domain names they’ve bought. But all he heard from the public market’s representatives was, “When are you going to get your customers to use self-help so you can cut your support staff?” So he pulled the plug on the IPO.

I cringed at Parsons’ unabashed enthusiasm for ads that plant his logo on the chests of well-endowed women (”"Because that’s where every guy would be looking”). But most Web 2.0-style execs could learn something from his understanding of the basics of retail psychology: “People love the convenience and speed that comes with the Net. But when it comes to resolving problems or learning features, people much prefer to deal with other people.”

From Eric Schmidt’s argument that network-based computing would prove irresistibly more reliable than alternatives, to Jeff Bezos’s pitch for Amazon’s on-demand storage and computing services as a means to “let people spend more time and dollars on the differentiated part of what they’re doing, less on the undifferentiated,” to Microsoft exec Debra Charpaty’s presentation about the nuts and bolts of building and running datacenters, one focus of this Web 2.0 conference has been on the server side of the old client/server dichotomy.

Web services are great, the argument goes, but don’t forget about what it takes to deploy and maintain them. “The Cloud” is a nice metaphor for everything that’s “out there” on network-based services, Charpaty argued; then she showed slides of endless racks of machines and squat, windowless buildings sprouting on desolate flats, and declared, “This is the real cloud.”

In one sense, these vast, electricity-hogging, heat-dissipating, cycles-generating structures are the new mainframes. Yet they are also the nerve-centers for an approach to computing that’s more distributed than ever before.

How do the businesses at the heart of the Web industry manage to juggle their determination to dominate the increasingly centralized business of providing the new basics, like storage and raw network-based processing, with their professed dedication to the values that shaped the personal computing industry that gave birth to theirs — values like freedom of speech, individual empowerment, and the unlocking of personal creativity?

That’s the big question underlying all the other controversies more visible on the surface here, like Net Neutrality or intellectual property or open APIs or data mobility.

While we clink glasses over House victories and bite fingernails over squeaker Senate races, here are some notes from today’s sessions at Web 2.0.

Thirteen new companies offered five-minute pitches for new products and services at the “Launchpad” event here.

The one that jumped out at me, unsurprisingly, given my history of interest in personal-information managers and the focus of my book on one such project, was Stikkit. It’s a personal-information manager (and sharing tool) built around a sticky-note metaphor. It looks like it has a heritage stretching all the way back to the old-fashioned “terminate and stay resident” note-taking programs like Sidekick and free-form PIMs like Lotus Agenda. Stikkit is led by Rael Dornfest, who I know from his work organizing many editions of the O’Reilly Emerging Technology Conference. I made a note to myself to explore it further tonight, but it appears to be down at the moment. More later!

I was also intrigued by Klostu, an attempt to create a “super-social network” linking together the separate islands of the “Boardscape” — the thousands of disconnected message boards across the Net. This strikes me as smart: there were tons of communities sharing stuff online long before anyone had coined the term Web 2.0, and it makes a lot of sense to serve them.

The presenter for Instructables, a site featuring user-contributed “how-to” projects, repeatedly emphasized that his service’s most important feature is the passion of his users. He’s right: more than spiffy software or innovative business models, that’s what makes any Web venture — “2.0″ or not — matter.

Here are the rest of the projects:

Omnidrive and Sharpcast: Two different approaches to syncing stores of content across multiple machines and devices.

Turn: Automated ad targeting.

Sphere: “Less geeky” blog search.

Adify: Instant advertising networks.

3B: Three-dimensional, walk-through Web browser.

ODesk: Hiring market and distributed management system for software developers.

Venyo: Reputation management service for bloggers.

Timebridge: Outlook add-on for meeting scheduling.

There is a decided atmosphere of hysteria in the halls here at the Sheraton Palace Hotel, where the Web 2.0 conference — suddenly rechristened a “summit” on its eve — has gathered for the third time.

Outside these walls, the ancien regime of a corrupt Congress may be about to crumble. But here, the crazed talk is all about startups and mashups and buyouts. If last year’s event offered what I called a strong whiff of “heady eau de dot-com,” this year we are drowning in the bubbles.

When I checked in at the conference registration, the woman working the desk apologetically asked to check my photo ID. Earlier, someone had jumped the desk and tried to steal some badges.

Later I’ll post about the companies from the Launch session. And we’ll see how the network holds up for liveblogging some of the talks.

First, my apologies for falling behind on this project — travel and freelance work ate into my writing time. From now on I should be able to keep more to schedule.

This week concludes our Dijkstra marathon. Next week we’ll tackle Knuth’s retort to Dijkstra, “Structured Programming with go to Statements”. The week after, it’s on to Mitch Kapor’s Software Design Manifesto.

Thanks to everyone who has posted comments; you’ve kept some informative and spirited discussions going in each of the previous Code Reads postings, and I, for one, am learning a lot!

Beginning tomorrow afternoon (Tuesday, 11/7), I’ll be at the Web 2.0 Conference, from which, wi-fi willing, I will try to do some live or semi-live coverage. If I can keep my eyes away from the election returns…

Steven Levy came to Sylvia Paull’s Berkeley CyberSalon at the Hillside Club tonight to talk about the iPod and his new book, The Perfect Thing: How the iPod Shuffles Commerce, Culture, and Coolness. I haven’t read the book yet (Farhad Manjoo has, and his Salon review is a wonderful meditation on what, both good and bad, the iPod is doing to the experience of listening to music). There’s a nice excerpt online in Wired; Levy’s also got a blog on the topic.

Levy started off by largely disavowing his superlative title. Of course, he admitted, the iPod is far from perfect, from its too-easily-scuffable skin to its too-confining conception of digital rights management. He said the device represents more of a “perfect storm,” a perfect summation of all the issues that arise when a medium goes digital.

I have to say I didn’t find this too persuasive (maybe he makes a better case in the book!); it might be better just to say, “Book titles are chosen to get your attention,” and move on. Because everything else Levy has to say about the iPod is fascinating, amusing and important.

Levy sees the iPod’s shuffle mode as the key to its meaning — so much so that he got playful with the book, writing each chapter as a discrete unit so the whole book could be put on shuffle mode. There are four different sequencings of The Perfect Thing out there; no telling which one you’ll get. (Once upon a time, in my previous life as an arts critic, I did something similar in channeling the spirit of John Cage for a review of a celebration of his music.)

He asked the Hillside Club crowd how many listened to their iPod with shuffle on; I’d say about half the audience raised their hands. I wasn’t one — though I find shuffle an amusing novelty, mostly I love digital music for the control it offers me, the chance to be my own DJ, so why would I want to go random? After listening to Levy, I think I’ll try it more; he made a good case for seeing what interesting juxtapositions turn up between the music you’ve chosen and the moment you’re experiencing.

I asked Levy whether the pro-shuffle and anti-shuffle tribes divide by age, hypothesizing that maybe a forty-something like me is still rebelling against growing up listening to bad radio, whereas a younger person who grew up with digital music might be craving more serendipity. But Levy said he hasn’t noticed an age skew between pro- and anti-shuffle-ites (he’s a bit older than me and is a shuffle-ite himself). He guessed that it’s more like the division between people who have the patience to organize their lives around PIM (personal information management) software and those who can’t be bothered. That makes sense — the PIM devotees (I’ve long been one) would also have the patience to program their own listening.

Levy also talked about the strange experience people have when they find that their ostensibly random shuffle mode seems to play favorites; for him, Steely Dan just kept on showing up. A column he wrote on this topic evoked a torrent of amusing email, some of which he read. Deeper investigation among mathematicians led him to conclude that Apple wasn’t lying when it said that shuffle really is random — and that the experience people had of shuffle “favorites” is actually a statistical phenomenon known as “clustering” that turns up in nearly any random distribution.

Lee Felsenstein asked Levy about what the iPod’s triumph has done to narrow public space, now that so many of us are walking around with our own private soundtracks. Levy’s answer made sense for a New Yorker: “When I’m on the subway, I don’t really intend to do much social networking.” But what about outside of dense urban conglomerations (the kinds of places Steven Johnson celebrates in The Ghost Map)? Do we need more alienation in the cookie-cutter exurban communities where human connections get more and more tenuous? The “don’t bug me” message is useful on mean streets; but out in the vast wasteland, iPod-induced solitude may be worth worrying about.

I’ll be heading up to the OOPSLA conference next week (I’ll be there all day Tuesday and Wednesday). It’s in one of my favorite cities in the world, Portland; and it’s a chance for me to brush up on the software realm as my book nears publication.

OOPSLA stands for “object-oriented programming, systems, languages and applications”; it’s a venerable conference dating back about 20 years, and it serves as something of an epicenter for the more visionary or radical tradition in the software development world.

I attended OOPSLA two years ago, and it was one of the highlights of my book research — with presentations from all sorts of interesting people, including Alan Kay, Jaron Lanier, Ward Cunningham and Richard Gabriel. This year’s speakers include Brenda Laurel and Charles Simonyi.

If you’re at the conference and want to chat, leave a comment here or drop me an email.