I didn’t get to do much for Yahoo Hack Day, but I did get to help a coworker a teeny bit with an implementation of Y! Search for social web sites, including Facebook. There could be some interesting repercussions from that, so I won’t say more now. But what did surprise me is how many Yahoos are active on Facebook.

Myself–I’m still a Facebook curmudgeon. But mostly I simply haven’t had the time to check it out, or figure out the value proposition of accepting an invitation. -m

Speaking of podcasts, last week I unsubbed from Rocketboom, the show having officially become unbearably advertising-swamped. It feels good (but not as good as getting that hour-per-week of my life back from Diggnation).

Possibly coming soon: unsub from Security Now, instead of fast-forwarding through half of it at present. -m

EXSLT support coming to Firefox 3.0. Python Flyweights. Timeline of MSFT engagement on document standards. RDFa Primer. And not that this is a conspiracy blog or anything, but strange things are afoot at Minot AFB, hours from where I grew up. -m

My Copious Free Time(tm) has been filled lately by two different evaluation projects. One is the 2nd Annual Writing Show Best First Chapter of a Novel Contest, for which the first round of judging is just winding up. The main benefit for contest entrants is that every submission gets a professional critique of at least 750 words. But additionally, each submisison gets a score on a 50-point scale, based on:

• 10 points for Story. Is it a compelling read with a great hook? Are we engaged?
• 10 points for Style. Is the writing smooth and tight, without awkward constructions, extraneous verbiage, and redundancies?
• 10 points for Dialog. Is the dialog natural and does it move the story along?
• 10 points for Character. Are the characters interesting? Do we care about them?
• 10 points for Mechanics. Are grammar, spelling, and punctuation correct?

I’m also attending some classes aiming toward becoming a Certified Beer Judge (details on Meadblog). This isn’t as fun as it sounds. (Well, OK, maybe it is…). The idea is to build up better sensory perception so that my personal brewing and cooking projects can benefit. But the upcoming test is 70% written essay questions like “Identify three distinctly different top-fermenting beer styles with a starting gravity of 1.070 or higher, and describe the similarities and differences between the styles”. 30% of the test is based on actual tasting and filling out a tasting sheet. Of interest, the scoring here is also based on a 50-point scale:

• 12 points for Aroma.
• 3 points for Appearance.
• 20 points for Taste.
• 5 points for Mouthfeel.
• 10 points for Overall Impression.

The interesting part is that there’s similarities between the two tasks. For both, I need to work off of physical paper, not in my head on on a computer screen. For both, I first “skim”, building an overall impression, then dig down into individual categories to assign a score for each one. Then I step back and look at my numbers, and check whether everything makes sense and accurately records my impressions. When I’m satisfied, I add everything up and am done.

Most day-to-day problems aren’t so well structured or normalized, but nonetheless, I find myself tackling all kinds of problems with a similar approach. There you have it. Writing and drinking beer make you a better person. :) -m

It’s too easy to get absorbed in all the terrible things happening on the news. But not everything is like that. Take 7 minutes and watch this. -m

• Everyday Life in Ancient Mesopotamia, Jean Bottéro
• Mesopotamia: The Invention of the City, Gwendolyn Leick
• The Structure of Scientific Revolutions, Thomas S. Kuhn

Yeah, they’re related. -m

I just finished an online version of SICP, the famous computer science text Structure and Interpretation of Computer Programs (link to full and official text online). What do I mean by “finished”? Well, there are online video lectures (link to iTunes-ready RSS video feed), expertly delivered by SICP authors Sussman and Abelson themselves in 1986. Beyond just watching the lectures, I took careful notes so I have something to refer back to later. I didn’t get the full intense college-class experience–no pop quizzes for instance–but it’s still helpful. Recent Python and XSLT I’ve written has been influenced for the better.

It’s interesting to peruse the reviews for this book on Amazon. They seem to mainly fall into gushing 5-star reviews, or else ‘yecccch, I’ll never use this stuff’. Both are correct. As Harold Abelson says to start of the first lecture, computer science is neither about computers nor science. The point of this endeavor isn’t about “programming”. You won’t find many Scheme jobs on LinkedIn, for instance. It’s all too easy to get pulled into the world of trench warfare programming, so it’s good to be able to step back and survey broader theoretical issues.

My views toward Lisp/Scheme have shifted as well. Before, and for about the first half of the lectures, I would have talked vaguely about how Lisp has an elegant purity but unusable syntax. Lots of Irritating Superfluous Parentheses, and so on. By the end, I have a lot more respect for the language. I admit I was rather floored by the metacircular evaluator lesson, where a suitably fezzed Gerald Sussman writes an entire Scheme interpreter–writes in Scheme–on a blackboard.
Prior to that, I went through an MIT course on differential equations by the engaging Arthur Mattuck, picking up where my electronics left off. Prior to that, a book called Problem Frames, about fully analyzing problems before diving into solution space.

So I’ve been keeping busy. It seems like these sorts of things run in cycles for me, with a full cycle taking around two years. So I’m really curious: how do others manage “continuing education”? What have you learned lately? How do you learn best? What should I look in to next? Comment below. -m

I fell asleep one night while reading Ray Kurzweil, and had this crazy dream where the internet called me up (over VOIP, naturally) to complain that none of my web pages made sense. Par for the course, I thought at first. But then I told the internet a few things, to let me worry about my own domain of concern; he/she/it grappled with a response when a loud noise awoke me–my chirping alarm clock. I reached over to pound the Snooze button, but I stopped when my eyes focused on the display, which read in segmented LED letters: I rtFm. -m

The approximately seven readers of this blog have probably already heard this, but just in case: I have a new role at Yahoo!–working on next generation search.

Lots of details are still falling into place. For now I describe it: “Imagining, specifying, prototyping, developing, and evangelizing next-generation web search experiences leveraging the full and unique capabilities available within Yahoo!”

In many ways, this is a logical stepping stone after oneSearch, and I’ll be dealing with lowercase semantic web issues more now. Expect the focus of this blog to shift accordingly (though I’m still interested in mobile and will make note of important happenings.)

Search On! -m

Everyone gets so much information all day long that they lose their common sense. –Gertrude Stein

…the solution to the overabundance of information is more information. –David Weinberger in Everything is Miscellaneous.

Weinberger’s book is a great read, taking you to lots of different places–from a prototype Staples store to the underground Bettmann Archive, and meeting a variety of different folks from Linnaeus to Dewey. It’s the kind of book that attaches itself to a particular idea and riffs on it at length, covering lots of details and implications that show great insight, but yet seem obvious after reading.

The book posits three “orders of order”. The first I’d call the problem of atoms. Arranging physical things, whether a library or your silverware drawer is primarily limited by the physical realm. If you have only one copy of a particular book, it has to go on a shelf somewhere. If you have two copies, you need to decide whether to keep them together or separately, possibly increasing the chances of losing track of one.

The second order I’d call solving the problem of atoms with more atoms, the prototypical example being a library card catalog. It’s still physically limited: a card can only hold so much metadata, but at least it’s easy to have multiple cards for a given book, making it easier to find something based on your choice of title, author, subject, or something else. Notably, many of the early online efforts have been straight translations of the second order into the digital realm.

But the third order is something else altogether, solving the problem of bits with more bits, as the leading quote indicates. I get the impression that some sites, like Amazon and del.icio.us, are getting closer to Weinberger’s third order, but none have fully achieved it yet. The third order fully blows the doors off of the constraints of atoms, which we’ve spent the last few thousand years developing and getting used to.

If you’re geeky enough to be reading this here, you’ll be familiar with many of the lines of thought found in this book: Wikipedia vs. Britannica; implied vs. concrete hierarchies or ontologies; centralized vs. decentralized control; Semantic Web vs. “smushiness”.

My complaint is that for a book that talks about the three orders of order, the text itself is firmly tied to first order atoms. Sure, there’s a ongoing blog with tagging, comments, etc. but to actually read the text, you need to find your way through a book store, card catalog, or online bookstore to get it. Sure, there’s no third order of economics (yet), but still I would have liked to see something more.

So, what does this all mean for a company like Yahoo!? That’s the question I’m working on now. Stay tuned. -m

Havi Hoffman at Yahoo! gave me an advance copy of David Weinberger’s Everything is Miscellaneous. I’ve been a fan of Weinberger for years–I even quoted him in my book.

I’m just getting in to the reading. I’ll report back when I have more to say. -m