We’ve Done This Before, Honest!

May 16th, 2005  |  Published in old and busted  |  2 Comments

Dunetchka and Justin dropped Amelie off for the afternoon, so we got to see what it might be like having two little ones around the house.

Ben with a rakeBen was mildly freaked out at first. He kept a pretty close eye on Amelie, didn’t make much noise, and insisted on being close to either Al or me at all times. It was a little odd, just because he deals with other kids at daycare all day long. Must’ve been the encroachment of another kid on his home turf for the first time.

Unlike Ben, Amelie is one of those babies who seem to “just sleep.” So we went on a walk to pick up lunch and she was nodding off by the time we were home. We gave her a bottle, she fell the rest of the way asleep, and she stayed that way for a while. When she woke up, she played. No carrying on or anything. Much more pacific than Ben.

For as lo-maintenance as Amelie is, it was pretty strange having her in our care. She’s a sweet little girl, but we realized we didn’t really know what much of anything she did meant. And she’s still young enough that most of the gimmicks I’ve got that work on Ben don’t phase her: raspberries, lip-blubbling, ear-pulling and eye-rolling were all big busts.

Once Ben was up from his nap, he was a lot better disposed to the visitor. She sat in his old bouncy chair and he chattered at her and played with the toys that dangle from it. So evidently we didn’t scar him too badly.

Here’s hoping he’s also forgotten the confrontation over how and where the rake is a good toy. It’s o.k., for instance, in front of our house and away from Al’s freshly planted flowers and bushes. It’s o.k. on the sidewalk in front of the neighbor’s house. It’s kind of not o.k. in the neighbor’s rose bushes.

I managed to redirect him a few times, but the third or fourth time he headed right back for the roses I decided to take him to the back yard. He saw that I was trying to herd him, fell down to the ground, got a deathgrip on the rake and began to wail. I carried him and his rake to the back and tried to give him the rake back, but I’d ruined the whole experience for him and he was inconsolable. Bedtime wasn’t much later.

Is It Bendable?

On the recommendation of the ever-interesting Cool Tools, I picked up a 20q yesterday. It’s a plastic ball that plays 20 questions using a canned neural net derived from the one operating at .

> Last year, after 1 million rounds of 20 questions online, the neural net had accumulated 10 million synaptic associations. It has a 73% success rate of guessing what you thought. Burgener then compressed the 20Q code to run on a chip, and had the neural net select 2,000 of the most popular 10,000 objects it then knew about. He then had the neural net select out the most useful 250,000 synaptic connections related to those 2,000 objects, and hard wired that learning into the chip in the orb.

Unfortunately it doesn’t have much way to learn from games it plays with people: It’s strictly working from a fixed database/net. But it’s pretty good. Al & I have played with it a few dozen times in the past day. So far we’ve beaten it with water tower, firewood, DVD, jellybean, baby gate, banana, and dirt.

On the other hand, from those “losses,” it came really, really close a few times. For jelly bean it guessed candy, for dirt it guessed sand, and for DVD it guessed CD ROM. For firewood it guessed pine tree. For banana it guessed plantain.

The biggest wow we got out of it was when it correctly guessed “pacifier” from a bunch of questions that didn’t seem to go anywhere near the kinds of things you might use to describe a pacifier with its questions.

It has also scored on platypus, spaceship, racecar (not just any car), and pinecone.

Its most bizarre loss came when we had it trying to guess “bear,” and it first tried “cat” then came back five questions later (it gets a second chance at 25 questions) with “tabby cat.”

It couldn’t guess itself, but it did guess “handheld computer.”

None Dare Call it Chickenshit

I must have spent a solid 90 minutes last night trying to work up a suitable rundown of all the stuff I’ve read about the Wired News/Michelle Delio story this week. I got in over my head and decided to reapproach today by just providing some links and comments to get from the top of the story, wherein MIT Technology Review decided it had a rat in the house, down to the bottom, wherein friend Phil explained why the hell the whole thing matters.

For the sake of putting most of the story in order for anyone who didn’t notice it:

And here are a few representative links we can put outside the timeline:

The most important of all those links is probably MIT Technology Review editor Jason Pontin’s “Did Michelle Delio Fabricate Sources?,” which took the time to double back around once the site had washed its hands of her reporting to assess how Wired News and most of the blogosphere has reacted. His succinct summation:

> “The blogosphere has largely accepted the version of events that Delio and Wired News are jointly spinning: the conventional wisdom on slashdot and elsewhere has quickly become that unconfirmable sources are not the same as fabricated sources. Besides, every one makes mistakes.”

Then he calls it as close as he can:

> “As journalists, if we have any useful function at all, it is to represent the facts of the world as we find them–the rest is intellectual theater. Delio and Wired News are, in my opinion, sheltering behind a problem in logic: it is almost impossible to prove a negative. Put another way, Technology Review would have to speak to every human on earth to prove that Delio’s suspicious sources do not exist. But another logical rule, the law of parsimony, tells us the simpler explanation is that Delio made stuff up.

> “This is serious stuff. We chose to remove all of Michelle Delio’s stories from technologyreview.com–including the three stories which proved accurate. I do not want to publish a journalist whom I am reasonably sure is a fabricator.”

“[A] journalist whom I am reasonably sure is a fabricator” is about as far as anyone wants to push the matter, and Wired News has made it clear that there is no “clear, bright line” to be found in the matter. It’s going to pick out the bits it can’t verify and carry on. No word on whether Delio will continue to have a role on the site, no apparent intention (despite some strong hints from Penenberg) to judge on Delio’s overall truthfulness.

When the Jayson Blair scandal broke, journalists began asking how Blair’s fabrications had escaped the presumably built-in correction mechanism of the very people who were being misquoted or lied about. The AP did its own survey, and came up with this:

> Those who said they failed to report errors had a variety of explanations: They doubted newspapers cared about mistakes or would listen to them. Navigating a newspaper’s corrections system would take too much time. The error was so obvious that surely someone at the newspaper would correct it. They believed inaccuracies were intentional in journalism that glosses over the fine points and hypes storytelling.

It cited a Karen G. Johnson of Otis Orchards, Wash:

> “she does not consider the errors she sees to be ‘mistakes’ but rather ‘deliberate embellishments or fabrications to make the story more interesting.'”

Assorted axe-grinders are fine with this state of affairs. A national news media that’s not credible in the eyes of its readership because of perceived indifference or, worse, a widespread belief that lying is simply what reporters do to spice things up, can’t fulfill any meaningful role in the national dialog.

Wired News probably seems like a tiny concern because, let’s face it, it isn’t first to the stories very much, and even when it’s second or third it isn’t bringing a lot of extra insight. It’s a brand name and some PageRank, and The Gawker’s take on the matter seemed only mildly forced:

> Delio, a freelancer for Wired News, disputes the contention, saying that – wait a second. Wired News? Wired News? You mean those guys who run stories like ‘Blogger Jason Kottke has gone pro’? This is what weŐre getting all worked up about? Hell, we assumed they made that stuff up as part of the job.

But it’s part of a bigger pattern, and I’m going to give Phil the final word, taken from a mail he sent to scoop this week, because he’s the rightest of the lot when it comes to explaining What it All Means:

> The broad effort to undermine the fundamental concepts of credibility and

truth is real. We’ve been coaxed into blanket disregard for civic

institutions like the government and the information media, to the point

where only saps and chumps stick up for them. “Aa, they’re all liars,”

people say. But they’re all not, actually–only the ones for whom the truth

is a rhetorical opponent, an enemy. Propagandists and liars don’t need to

prove credibility–it’s enough to make us believe there’s simply no such

thing. When we have no reliable source of fact, they’re free to play on our

hunches, fears, and prejudices. And all they need is for attractive

anchormodels to politely hold the door.

Sign from Gawd

I downloaded the Ubuntu installer CD today, burned it, booted up my clone (which currently has an extra 80GB to play with), and discovered that, well, Ubuntu can’t tell there’s a USB keyboard there. So it can’t install. The BIOS knows it’s there, because I can configure the BIOS with it at boot time. Windows knows it’s there, too. I can even bail out at boot with a quick press of “F8.”

When I travel back in time to the me of four years ago and think about the review I’d write, I’m guessing that a.) I wouldn’t have a USB keyboard because I generally knew better than to depend on one as a Linux user anyhow and b.) I wouldn’t have thought twice about shrugging the installer’s failure off. I might have mentioned it in the review, but I wouldn’t make a big deal out of it because part of the he-man Linux reviewer culture involves having a.) a ton of spare parts laying around and b.) remembering to namedrop modprobe.

As it is, though, I’m not the me of four years ago and I’m thinking that if every Linux reviewer were required to simply stop the review at the point a common piece of hardware that worked well with the competition failed to be recognized and halted an installation, my Ubuntu review would be about two paragraphs long. Depending on if I padded it with some introductory crap about Ubuntu being based on Debian, but supposedly friendlier, etc.

Anyhow… I plugged in a PS2 keyboard (because I have a few in a bin, which is sad), and rebooted. Once I tapped “enter,” it booted a full kernel and that managed to autodiscover all the USB stuff.

Once I finished up install, I pulled the CD out of the drive, rebooted, and got a “GRUB Error 18.” A little Google told me that I can’t boot into either Linux or Windows now because Ubuntu’s version of parted hosed my partition table.

But you know what? It picked up DHCP like a champ!

Anyhow … some digging around netted me my XP boot CD, I dropped into rescue mode, ran FIXMBR, and put the whole thing behind me.

Ubuntu couldn’t have run “Battle for Middle Earth” anyhow.

Comments are closed.

© Michael Hall, licensed under a Creative Commons Attribution-ShareAlike 3.0 United States license.