… has to be satisfactorily.
Displaced to the second rank: procrastination. I liked that one because it plays such a major role in my life, and my native language (german) has no word for that concept. It’s nice to learn a new foreign word and realize: “Oh, so that’s what I’m doing 90% of my time.”
[John Sowa:](http://www.w3.org/mid/4426BF93.1070503@bestweb.net)
> RDF and OWL are legacy systems that must be supported, but
semantics, pragmatics, and ontology are where the action is.
RDF and OWL are too limited, clumsy, and inefficient to support
any serious work in knowledge representation and reasoning.
An enormous amount of effort in the SemWeb literature addresses
workarounds for getting up to the level where AI was in the 1980s.
> My recommendation is to import anything in those languages into
Common Logic and do the real work with CL-based languages.
This comes from one of the top authorities in the knowledge representation and reasoning field.
RDF and OWL are a compromise between two camps, KR/reasoning one one hand and web/data integration on the other. My tent is in the latter camp, and I blame all of RDF’s problems on the former. I hadn’t really realized that some folks on the other side are unhappy too.
(My first thought, on reading John’s statement, was: “Don’t feed the trolls.â€)
Here’s Danny’s continuation of the “SemWeb types should embrace OPML/RDF is too hard” thread, in which he takes a step back to paint the big picture.
The goal is the Web of Data, Web as a platform. An RDF-based semantic web could be one way to get there. Emerging “naive” technologies like RSS and OPML and microformats and structured blogging are other paths that lead into the right direction. So are heavyweights like Microsoft’s WinFS. Danny does a great job connecting the dots between the current web and this jumble of technologies.
What’s the clear path? The point where I disagree with him is whether RDF is the “clear path,” the one that will ultimately lead to the goal on an optimal route. To stay with his forest analogy, we’re following a map that was drawn by early trailblazers with incomplete knowledge of the territory and is now cast in stone as W3C Recommendations. It’s a magnificient and elegant map, seducing in its apparent simplicity. But as we make our way across the territory, we waste endless time fighting obstacles that were not on the map.
Meanwhile, the “naive” folks are exploring different corners of the forest from the ground level. They waste time in areas where we have long established a clear trail, they often take the easy road downhill en masse just to find out it’s blocked, but they adapt, they learn, they optimize locally, and new folks who embark on the pilgrimage follow their lead instead of trusting our map.
We are believers. Our trust in the map is rooted in its good-looking theory, and we follow it even if the day-to-day evidence suggests it has major flaws. Correcting these flaws is extremely hard because they have been cast in stone as standards and we already have invested so much, we are committed, we can’t just backtrack and try another route.
Phil Jones said the same thing in less colorful words and stresses that the “naive” types explore and fight and mess up and learn in public:
With communities of thousands of users depending on the format and tools for their everyday needs, there’ll be “many eyeballs” identifying problems and suggesting workarounds and fixes. The naive community will probably address the problems in order of actual pain they cause, undoubtedly leaving dozens unsolved altogether, but solving the big ones to most people’s satisfiction in a fairly short time.
Do you remember? So, my fellow believers: Do you remember when you first beheld the map in all its glory and became convinced that this was the path to follow? Do you remember how you figured out the hard way, step by step, that it’s not so simple in practice? It’s time to stand up and point at things and say: “This was not on the map!”
In many ways, “geeks” are the canaries in the New Economy’s coalmine. Programmers and knowledge workers often operate as free agents in the digital economy—self-employed or contract workers with little job security and a constant need to reinvent themselves for new employers. Working at home or remotely, they are overwhelmed by a barrage of e-mails and media inputs, lack the structure and community provided by conventional offices, and must erect or erase hard boundaries between their personal and professional lives. Such a vacuum of external supports and structures means that such workers must find new systems for setting goals, defining next steps, and managing the “project” of life.
This is from an excellent article about the rise of the self-help phenomenon, and in particular about GTD, the productivity and time management strategy that has attracted a cult following among computer folks (including me).
(via 43folders recent links)
Here’s a short screencast demo of Dabble DB, one of the more exciting offerings in the endless stream of new Web 2.0 apps. It’s a web-based database, aimed at casual users. Microsoft Access on the web, kind of.
It’s not public yet, but the screencast shows off the features very well: rich import and export options, schema evolution, very polished UI.
Roger Jennings has an in-depth look at the Web-based DB marketplace.
I wonder if they will offer an API that makes it possible to build a database-driven web site with a Dabble backend. That would be awesome, even if it’s read-only.
Now imagine something like Dabble DB where you can have virtual tables whose items are loaded from an RSS feed or from something like a SPARQL query. I’m pretty sure that we’ll get something like that in 2006 or 2007, maybe not from Dabble, but from some other confluence of Web 2.0 applications and crappy data formats. That’s the kind of idea that brought me into the Semantic Web world, and it’s good to see a workable solution approaching on the horizon.
Just noted that one of my posts shows up on the first result page when I google for “perfectly organized.”
Obviously, Google doesn’t have a clue.
Update: Today I’m in the top ten for “cluttered homes”. Google may be slow on the uptake, but ultimately gets it right …
A nice little productivity hack for small teams: Todoque is a simple shorthand for writing down tasks, questions and ideas. Example:
1. Alice<Bob [Q]:How much are the tickets to the Opera? [A] They are $200 and you need to book by tuesday.
2. Mary<Alice [T]:Ring the bank and arrange 4th mortgage.
3. Bob<Mary[T]: Can you see if you can unify the tribes of Afghanistan into a singular cohesive democratic body.
Q is for question, A for answer, T for task, I for idea, E for event. The “<” shows who assigned the item to whom. Resolved items are crossed out.
I can imagine this to be pretty useful in email and wikis. If used consistently, the items can be automatically parsed into summary lists. This also reminds me of the //TODO and //FIXME comments often seen in source code.
(via 43folders recent links)
Perl is the Linear A of the computing world.
(from my silly little outburst in the comments over here (in German))
Why is Perl so successful, despite its obvious flaws? Danah Boyd:
“Coolness” is about structural barriers, about the lack of universal accessibility or parsability. Structural hurdles mean people put in more effort to participate. It’s kinda like the adventure of tracking down the right parking lot to get the bus to go to the rave. The effort matters. Sure, it weeds some people out, but it makes those who participate feel all the more validated. Finding the easter egg, the cool little feature that no one knows about is exciting. Learning all of the nooks and crannies in a complex system is exhilarating. Figuring out how to hack things, having the “inside knowledge” is fabu.
Often, people don’t need simplicity – they want to feel proud of themselves for figuring something out; they want to feel the joy of exploration.
Bugs make technologies seem alive, particularly if they’re acknowledged and fixed. They give texture to the environment and people are impressively patient with it if they feel like the architects are on it. It makes the architects look vulnerable which brings them back down to earth, making them real and fallible, but giving them the opportunity to do good. They let the benevolent dictator really serve the people.
Well, the quote isn’t actually about Perl. It’s about social community sites like Friendster and MySpace, and why teenage kids prefer MySpace’s bugs, ugliness and idiosyncracies over the Friendster’s streamlined simplicity and uniformity.
But I find it revealing. It’s the same with technologies. Being inaccessible can be cool, it’s a way of distancing yourself from people who are “not in the know.”
Just like parents who don’t comprehend what their kids are doing on MySpace and become afraid of sexual predators and cry for the lawmaker, the Java and C# and PHP developers can’t comprehend why the Perl disciples put up with all that apparent insanity around their language of choice. Now we know: It’s their way of rebelling against their colleagues.
It’s not just Perl: Perl is certainly an extreme example of this technology subculturalism, and making fun of the whale gut wranglers is cheap and easy. But are the rest of us any better?
I’m certainly not. After all, would I still be involved with RDF and the Semantic Web if it wasn’t for all those fine people at #swig, Planet RDF, the SemDesk workshops, and jena-dev? Would I have bought a PowerBook if all the alpha geeks hadn’t switched a year or two earlier? I look down on folks who don’t use a real text editor. And I certainly enjoy my Web 2.0 group therapy. I often advise newbies on open source mailing lists to “just get the latest code from CVS and compile” – knowing full well that this will be a major struggle for them. But it’s a rite of passage: You just can’t be part of the software developer tribe if you haven’t compiled a project from CVS.
Why do technologies get adopted or fail? If a technology gets widely adopted or falls by the wayside is not a function of its quality alone, we all know that. Sometimes it can seem as if quality doesn’t matter at all. The IBM PC triumphed over the Mac, Windows triumphed over OS/2, VHS over Betamax, C++ over Objective-C, Smalltalk over Java, PHP over Python, the list is endless. Many explanations have been offered. Worse is better – correctness, consistency and completeness can be sacrificed for simplicity, said Richard Gabriel. It’s all about marketing – using dollars or individual genius – said Steve Yegge. It’s about hitting the 80/20 point, said Tim Bray. I humbly submit: It’s also about being cool, being different, belonging to a tribe, expressing yourself.
Returning to Danah Boyd’s Friendster/MySpace essay: If you want your technology – be it a language, a commercial web site, an open source project, a standard – to succeed, then part of your job is to create an environment where your users and customers have the space to bond into a tribe, express themselves, be different and develop and share their cultural values. Because we all are rebellious teenage kids at heart.
(If you don’t follow the social software space and have no idea what Friendster and Orkut and LinkedIn and MySpace are about, then go read the essay. The socio-technological revolution of my generation was the Web. The next generation’s revolution will be the thing whose early incarnations you can see today at those sites. And you can watch it brewing in Danah’s writing.)
(On starting a side business:)
Many people worry that they’ll languish in obscurity. Don’t worry about having a great idea that no one knows about. Worry about having a bad idea that everyone knows about.
