Wordyard

Hand-forged posts since 2002

Archives

About

Greatest hits

All Web 2.0, all the time

by

Over at Technology Review, in “Homo Conexus,” James Fallows reports the results of a two-week experiment in which he cast off desktop software and cast his lot with a collection of Web-based (“Web 2.0”) applications.

He finds that the new world is best for stuff that’s natural to share: calendars (Google Calendar) and photos (Flickr). The Achilles heel is connectivity: For instance, he uses Writely to compose the article itself and, inevitably, his Net connection drops in the middle, forcing him to reconstruct his work from local backups that he’d cautiously been keeping.

Fallows also senses a “tragic” potential in the essentially trusting (and, he argues, perhaps overly idealistic) framework of the new Web:

Every bit of the Web enterprise operates on trust. Web-based commerce has gone as far as it has because of the surprisingly low level of fraud and error…. all this depends on the basic trust that messages will go through undistorted, unintercepted, and in general unimpeded.

If problems like privacy breaches or indentity theft cause that trust to break down, he suggests, the Web 2.0 era may lose its innocence, along with the trust that keeps so many of its wheels turning.

If you live and breathe this stuff — if you religiously read TechCrunch and store your bookmarks in del.icio.us — then you’re unlikely to learn much new here, or to find Fallows’ sympathetic skepticism very congenial. But for the rest of us, the article is a useful reality check — a reminder that a lot of what early adopters get excited about isn’t yet ready to cause a mainstream stir.
[tags]web2.0, james fallows[/tags]

Quote of the day: Microsoft’s Long March

by

From an interview with Steve Ballmer in this morning’s Journal:

You can’t replace Bill Gates, but I think the future of the company is brighter looking forward even than looking back. You never can go through the teenage years of being a company again in the sense of growing from nothing to something, but I think we can go from something very good to something great.

When did China get great? China didn’t get great under Mao Zedong. China got great under — in the recent years — probably got great under Deng Xiaoping.

And you thought Apple was the personal-computing revolutionary! If BillG was wearing the Mao jacket, then did that make Paul Allen Lin Biao? If Steve = Deng, does that make Ray Ozzie Hu Yaobang — and is he in danger of being purged?

What was Ballmer thinking?
[tags]Microsoft[/tags]

Outliners then and now

by

I am addicted to outlining as a means of organizing my work and life. (And no, the outliner in MSWord does not count, it’s a clumsy, kludgy horror that has probably turned off millions to the value of outlining.) I still use Ecco Pro — a long-orphaned Windows outliner — every day. (This old post has links to some of my writing on the topic.) I used Ecco to compile the research for my book, and I use it, GTD-style, to keep the spheres of my life moving in harmony.

Ecco is a fascinating hybrid of the pure outliner that Dave Winer pioneered in the 1980s and the free-form personal database exemplified by Mitch Kapor’s Lotus Agenda, which let you recategorize and invent new categories for your information on the fly. (Today’s tagging phenomenon is a latter-day version of the idea.) Chandler, the product whose development my book Dreaming in Code chronicles, started with the ambition of bringing some of these ideas into the present, though it has since evolved in different directions.

I was reminded of this complex software genealogy recently as I read a page that Winer recently linked to — a detailed chronicle, written in 1988, of how his once-popular outliners (ThinkTank and More) came to be developed. (I found it because Winer linked to it from another page about thinking about the Internet as an idea processor — which is also food for thought.)

I’ve never understood why outliners never found wider adoption. Is it just the curse of Word (once Microsoft “included” outlining in Word, however poorly, the market evaporated)? Is it that people associate outlining with boring work they had to do in high school composition class? Is it that the number of people who like to organize their thoughts in collapsible hierarchies is just not very high? But the alternative model of idea-organization tools, which provides you with more of a 2D or 3D space to place and link words and concepts (cf. The Brain and other “mind-mappers“) has never caught on in a big way, either. Maybe the vast majority of people are still too busy figuring out how to wrestle their computers into submission to concern themselves with trying to use them as (in Howard Rheingold’s phrase) “brain amplifiers.”

Many contemporary outliners (like Shadowplan) feel more like checklist organizers than tools for organizing large amounts of text. With the more sophisticated programs, one problem I have (I’m thinking here of tools like Zoot and InfoSelect) is that they are built like e-mail clients with separate panes — a pane on the left where you expand and collapse nodes, and then a pane on the right where you read the text associated with the node that’s highlighted at the left. This separates the “thing itself” from the “relationships between things.” That’s not the way my mind works: I want to see the things and their relationships — all at once!

In Ecco, as in More, you’ve got the full text of each node right in front of you, in place in the outline hierarchy. This allows you to use the tool — as I understand Dave Winer does — as a primary writing environment; it also allows you to dump huge amounts of information into the outline efficiently, move big pieces around easily, and swoop quickly from a top-level overview to the finer details.

Today Mac users can adopt OmniOutliner, which has a feature called “inline notes” that begins to move it toward the model I prefer. If I were using a Mac every day I’d also check out Eastgate’s Tinderbox, Circus Ponies Notebook and VoodooPad. Windows users can still get Ecco for free. In the new world of web-based apps, there’s not a lot of activity yet — though there is a rudimentary AJAX-based outliner called Sproutliner. 37 Signals, the “small is beautiful” web app company, has a lightweight listmaker called Tada List, along with another product that’s sort of a free-form personal info manager called Backpack. And then of course we come full circle back to Dave Winer, who has created the Web-based outline format OPML (the OPML editor is here) for constructing and sharing Web-based outlines.

I don’t know if outlining software will ever take off, but to me it feels like a natural way to use a computer. I will keep using Ecco until they invent a version of Windows that won’t run it, and I suspect I will outline until the day I die.

POSTSCRIPT: Doc Searls’ technography from Bloggercon IV is a good example of outlining in action. He wrote about it here.
[tags]outliners, pims[/tags]

Newassignment.net: new-model journalism

by

Jay Rosen has posted a detailed sketch of a new, non-profit venture in the “citizens’ media” (or “networked media”) realm that he is calling NewAssignment.Net. The idea is to create an institution online where people can contribute dollars to fund reporting projects they’re interested in. These projects will in turn be pursued by paid reporters and editors working creatively with information and contributions flowing back to them from the Net. Foundation seed money gets the thing off the ground; money from the crowd keeps it going. Old-fashioned editorial processes mesh with newfangled feedback loops and reputation systems to produce something new and unique.

Jay is one of the bright lights in this area, and I’m looking forward to what he comes up with — especially since some of the issues and problems he’s exploring are similar to the ones I’m working on at Salon these days.

Rosen’s description makes it clear that he’s seeking to create an institution where many traditional journalistic values persist and shape the work being done in a novel mode. In particular, there’s the idea that the reporters are going to go out and ask questions and consider all the information flooding back to them from the Net and determine the truth as best as they can — even if that truth is not what the people ponying up the cash wanted to hear.

This, to me, is likely to be a major friction point for NewAssignment — which will doubtless be avowedly nonpartisan but which will not be able to insulate itself from the fierce political divisions that shape so much online discourse today.

At Salon, we don’t make any claims to nonpartisanship but do maintain our own tradition of journalistic pride, and a commitment to fairness and giving the “other side” a say, and a belief in telling the story as you find it, not as your political preferences might dictate it. This has regularly placed us at odds with at least some of the readers who are funding our stories with their subscription dollars. (The relationship is not quite the direct quid pro quo that Rosen envisions, where individual site visitors put their chips on specific stories, but emotionally it seems similar.)

So, for instance, in the wake of stolen-election charges in Ohio in 2004 we had Farhad Manjoo — one of the most talented, hardest-working and open-minded reporters I’ve ever worked with — devote a lot of time to exploring the story. He’d done significant reporting on the topic in the past. His conclusion — as our headline put it, “The system is clearly broken. But there is no evidence that Bush won because of voter fraud” — was well-documented and carefully delineated. But it wasn’t what many of our readers wanted to hear.

Ever since, Salon has had a steady trickle of disgruntled subscribers cancel on us, citing these stories as a factor. It’s never been enough to make any difference to our business, and it certainly won’t stop us from doing further reporting on the subject, and presenting our findings accurately. But it’s disheartening. And I think that NewAssignment may face some similar tensions if it ends up reporting on topics that people have strong feelings about, which it must if it is to matter.

The sample story Rosen walks us through to explain his new idea is one about wild variations in drug prices from one place to another. The assumption is that some people who are upset by what they perceive as unfair, rigged drug pricing might be willing to help fund such an investigation. But what happens if the reporters come back and say, gee, it turns out that the drug companies are innocent here, the fluctuations are actually the result of [some other factor]? (I’m not saying I love drug companies. This is just an example.) Will these citizen-journalism sponsors want their cash back?

UPDATE: Jeff Jarvis’s post about NewAssignment provides some tidbits of interest about the new media venture he’s been dropping hints about for a while, named Daylife. But I wonder about his comment: “We must explore new business models to support coverage of news and this is one of them.” It strikes me that the not-for-profit, institutionally-supported model Rosen has picked — perfectly reasonably — is good for many things, but maybe not so good for exploring new business models. Yes, there are sustainable nonprofit models, and maybe NewAssignment will turn out to be one of them; but it seems to me that Rosen’s plan is more about delivering a proof-of-concept for important new ideas about networked journalism than it is about building a business. If I’m wrong, I’m sure he’ll let me know!

[tags]Jay Rosen, citizens media, newassignment.net, Salon.com[/tags]

Yahoo goes scrum

by

Buried at the tail end of yet another which-portal’s-on-top? feature in the Times is this interesting tidbit about software development practices at Yahoo:

Meanwhile, Yahoo says it is now trying to emulate Google’s faster method of creating products. Like most big companies, it used to develop software by first creating a comprehensive design that defined how features would be written and tested. Instead, it is now trying what is known as a scrum method, where it will plan, build and test parts of a product every 30 days.

“We may not know how everything fits together,” Mr. Patel said. But by creating partly completed products that can be shown to customers, “We can get insights from users and react to that over a three- or four-month period to put it all together,” he said.

Scrum is a species of agile software development in which the development team, among other things, holds quick daily meetings and delivers new bits of functioning software on very short schedules. It’s all about “moving the ball forward,” scrum expert Ken Schwaber says.

[tags]software, software development, yahoo, google, scrum[/tags]

Dabble launches

by

Dabble, the new service for sharing Web videos that Mary Hodder has been developing, just launched. Think Flickr for video, but without the hosting of content and a more sophisticated focus on sharing “finds” than Youtube offers. I’m looking forward to experimenting with it.

[tags]dabble, video[/tags]

The blog rebooted

by

Almost exactly four years ago, on July 22, 2002, I started my first blog. Blogging felt natural to me since I’d been writing for the Web since 1994 and self-publishing since 1974 (originally via mimeograph).

My blog was part of a larger blogging program I’d put together at Salon, in partnership with Userland. It was the tech-downturn doldrums — an era when every time we at Salon opened the papers or fired up our browsers we knew that someone, somewhere, would be predicting our imminent demise. And there wasn’t a lot of extra cash at the company at the time, so the blogs program was chiefly a labor of love, launched in the wee hours. I did the CSS, wrangling Salon’s home-page design into Radio Userland templates, all by myself (which anyone who knows anything about CSS can probably tell with a single glance at the unruly code).

I loved Radio Userland at the time for the way it combined a blog publishing system and an RSS reader. But times change; Userland put its energy into other products; Salon Blogs produced many great blogs but not a substantial change in Salon’s business; and my blog settled down from the program’s focal point to a personal-publishing bullhorn.

Several months ago, in anticipation of Salon’s plan to build a new platform for users to contribute their own writing, we closed off new signups to the old Salon Blogs platform. Today I’m moving my own blog to a new home, here, at Wordyard.

I’ve managed to export my whole four years’ worth of archives (over 1000 posts, averaging about one per weekday for the whole timespan) to WordPress. (For those who care, I used the Radio Userland exporter, which pops out a plaintext file in Movable Type export format; edited that file to make things like titles and categories work; then imported into WordPress.) The comments, alas, will remain back at the original Salon Blogs location, where they will continue to be available.

With this move, I plan to blog somewhat more vigorously, and to provide more posts about my forthcoming book, Dreaming in Code, as its January 2007 publish date nears. I also look forward to leveraging some of the great features and plugins created by the WordPress open-source community.

If you subscribe to my RSS feed in Bloglines (the reader I’ve been using daily for years), the transition should be transparent — Bloglines will do the flip for you, you don’t need to touch anything. If you subscribe through other feed readers or services, you’ll have to resubscribe to the new feed address, which is here.

More anon!

Mashup Camp 2

by

Yesterday I spent the day at Mashup Camp 2. I missed the first one last winter, but what I read intrigued me enough to make a point of showing up when it came around again.

The two relevant things here, one having to do with mashups, the other with that word “camp,” which is really a proxy for the whole “unconference” movement of which this event is a high-profile example in the tech world. (Mashup Camp organizer David Berlind wrote about the first event’s experience with the format back in February.) Let’s start with that.

When I showed up at 9 a.m. down in Mountain View, at the Computer History Museum, the conference had no schedule — just an open grid on an eight-foot-long pad at the front of the meeting hall. An hour later, several dozen developers (and some “API providers,” a k a vendors or company reps) had introduced themselves, proposed sessions, posted the sessions on the grid, and presto, there it was, a conference schedule.

Mashup Camp instant schedule grid

There had been no arguments over process, no disputes, no grandstanding or boring throat-clearing. Part of that was the result of deft moderation by Kaliya Hamlin (she writes about the event here); part, no doubt, was the nature of the attendees — this was primarily an engineering conclave, after all. If we’d been talking about Iraq, something tells me the process might have been bumpier.

In the pop culture world, “mashup” means creating a new work by combining elements of two (or more) existing works. (Danger Mouse’s “Grey Album” — the Beatles’ White Album meets Jay-Z’s Black Album — is probably the highest-profile example in music to date.) In software, a mashup is a new program or service created by wiring up two or more existing programs or services.

Web-services mashups can be remarkably easy to hack together and provide immediately gratifying results — the canonical example was the Craigslist/GoogleMaps mashup that Paul Rademacher made last year, placing the Craigslist for-rent ads on Google’s map service. At Mashup Camp, developers got the opportunity to show off their projects during a “Speed Geeking” event (modeled on speed dating) at which visitors in groups of a half-dozen wandered from table to table to hear five-minute demos. Here’s a full list of the participating demo-ers.

I didn’t come away with the sense that any one of the projects I saw was going to change the universe. But put it all together and you got a window onto a simpler, faster, and perhaps smarter approach to software product development — one that trades in the virtue of from-the-ground-up consistency and thoroughness for the even more compelling virtue of “getting something working fast.” It’s software development as a Darwinian ocean in which large numbers of small projects are launched into the water. Only a handful will make it to land. But most of them required so little investment that the casualty rate is nothing to lose sleep over.

Fallows, PIMs and Chandler

by

James Fallows has been writing thoughtfully about computer software for longer than most of us have been using it. Years ago he wrote a definitive paean (long online here but apparently no longer) to Lotus Agenda, Mitch Kapor’s legendary personal information manager. (I say “a” rather than “the” because this program evoked such loyalty from smart writers it actually ended up with two definitive paeans; the other was by Jimmy Guterman.)

In the new issue of the Atlantic, Fallows writes about two latter-day PIMs — Microsoft’s OneNote and Chandler, the long-gestating project of Kapor’s Open Source Applications Foundation, the tale of which forms the central narrative of my book, Dreaming in Code. He interviewed me for the article; though most of our conversation wound up on the cutting-room floor, I did make it into one paragraph. I wish the article were online (there’s a stub here, but the full piece is only accessible to subscribers). But I couldn’t ask for a better venue for my first distant-early-warning book publicity. Here’s the relevant graph:

Despite substantial follow-up grants from foundations and universities, the team developing Chandler has so far released only a partly functional calendar application. Scott Rosenberg, of Salon magazine, became an “embedded journalist” on the Chandler project from 2003 to 2005 in order to investigate why good software is so hard to make. (His book about Chandler and complex software design, Dreaming in Code, will be published in November [now, January]). “It is taking a long time, but anyone who writes off Chandler is being short-sighted,” he told me. “They are on a quest.”

Fallows asked me whether I thought the book had turned out to be a comedy or a tragedy.

“Neither,” I replied, thinking furiously on my feet, my brain flashing back to my decade as a theater and movie critic. “It’s an epic!”
[tags]James Fallows, Dreaming in Code, Chandler[/tags]

Craigslist’s money left on the table?

by

Brian Carney’s Wall Street Journal piece about Craigslist wants to know why Craigslist isn’t maximizing its revenue:

One industry analyst has estimated that Craigslist could generate 20 times that $25 million just by posting a couple of ads on each of its pages. If the estimate is to be believed, that’s half a billion dollars a year being left on the table… Google has turned unobtrusive text ads into a multibillion-dollar revenue stream. And posting a Google-type ad or two next to its search results wouldn’t cost Craigslist users one thin dime. So why not cash in?… If Craigslist does what its users ask of it, and Craigslist doesn’t need or seem to want all the ad revenue it declines to collect, maybe we, as end-users, should ask them to post some banner ads and give us the money instead.

Carney is either failing to see or deliberately ignoring a simple element in the equation here: The absence of ads is one of the key factors behind Craigslist’s phenomenal success. No barriers, no annoying popups, no distractions, none of the gaming and manipulation that Google text ads increasingly invite. Instead, simplicity and effectiveness — and trust.

Of course Craig Newmark and Jim Buckmaster could turn on the ads and rake in some cash, short-term, but they would undermine what they’ve built and compromise the principles that have served them so well to date. They’ve clung tightly to those principles, against the conventional wisdom, and doing so has served them too well to stop now.