It's not in the news as I write this, but reliable sources are reporting that Benoit Mandelbrot has passed away.
Benoit was the father of fractal geometry. That's the geometry of objects that do not have integer dimensions, such as most natural objects, like the seashore, most plants and trees, etc. It's also the geometry of the border of the famous Mandelbrot set whose depiction has worn many a home floating-point unit down to the nub. See Wikipedia for a list of Benoit's accomplishments, and if you haven't read his The Fractal Geometry of Nature, it's well worth a look.
I interacted with him a bit when I was a manager at IBM Research in Yorktown in the Computer Systems department (I think that was its name then) and he was an IBM Fellow in the Math department. Much of what I know was from someone who had worked for him there, who I hired away (that person's choice) to work on parallel software.
Aside from being obviously brilliant, he also jealously hung onto all possible credit from everything done in his department. (This was a major reason the person was interested in leaving.) He probably considered this normal and his due, especially being from a European academic background. It wasn't the rule at Yorktown, though, where most managers were, well, managers, who generally took credit only up the management chain, not outside it.
At one point, he expressed annoyance at being given an IBM Outstanding Achievement award, stating that once someone has a higher award (IBM Fellow, numerous math society awards), one is not given a lesser award.
My most outstanding memory of him: One of the few times I spoke with him, about getting paperwork done on that person's transfer, he shifted slightly in his seat and noisily farted. Either his social skills were less than polished (they weren't), or he was expressing an opinion (certainly not impossible). In either case, I guess I'm a member of an elite group -- those farted upon by Benoit.
He certainly made a major contribution to the world. Given the obvious photogenic character of his discoveries (see the book), I'm surprised there haven't been 67 Discovery/Science Channel re- re- discussions of it, like there are for other photogenic areas of science. Maybe those will show up now. Too bad. He would have loved to be the subject, and seen them.
At a really good Rockies’ game last night (see-saw lead, tied 9-9 in 9th, Rockies won in bottom of the 10th. Go Rockies!) (but dang, they’re probably out of it now) I learned a side-betting game that proved to be fun. I wasn’t told a name, so I made one up for the title of this blog.
This is a side-betting game played among a group of people watching the game.
Initially, everybody antes up some agreed-on amount into the pot. $1/person is common, and I’ll assume that below. Larger amounts could clearly be used.
The pot is passed from person to person among the group, in an order determined at the start, in a cycle. We just passed to the person in the adjacent seat to the left, then back to the start.
The game begins when the first batter comes up to bat, and the pot is passed to the next person whenever a new player comes up to bat for either team. The player at bat when a person is holding the pot is “his planer” (or “her player”).
What happens to the pot depends on what happens to your player in his time at bat:
1.If your player goes out – e.g., strikes out, flies out, tagged out on the play his hit initiated – you add $1 to the pot and pass it to the next person. That player’s being tagged out later – due, say, to being picked off by the pitcher or tagged out in a play initiated by someone else – doesn’t matter.
2.If your player walks, you just pass the pot, neither adding nor removing anything from it: “If he walks, it walks.”
3.If your player hits a single, you take $1 from the pot and pass it on.
4.If your player hits a double, you take $2 from the pot and pass it on.
5.If your player hits a triple, you take $3 from the pot and pass it on.
6.If your player hits a home run, you take the entire pot. Everybody antes up again, and the new pot goes to the next person in order.
7.The person holding the pot when the game ends keeps the entire pot.
That’s all there is to it. The pot size can get reasonably large in a low-scoring game. In a high-scoring game, it rises and falls a lot.
I generally don’t like betting, but I found this enhanced my enjoyment of the game. Be sure you arrive with a bunch of $1 bills, though.
I’ve been wondering if the same principles could be applied to other sports. Football (American) – maybe. Something like taking from the pot on positive yardage, and putting more in on negative. Maybe $1 for each 5 yard interval, starting with $1 for 1-5 yards? Needs a cap, though, or 90-yeard plays would get too expensive for a minor fun game.
Soccer? Can’t see it. Ditto Hockey. Cricket? I don’t know enough about it to say.
The Blogger guys have added really simple-to-use statistics gathering in their "Blogger in Draft" beta page, so I've been trying it, comparing it to StatCounter and Google Analytics, both of which I've embedded in Perils of Parallel.
The interesting thing is that all three give different numbers. For example, Google Analytics always gives smaller hit counts than StatCounter, and both are lower than Blogger Stats. So obviously I prefer Blogger Stats. The differences are major, like a factor of two at each stage (so Blogger Stats is 4X Google Analytics).
I guess it's not horribly unexpected that there are hit count differences, although those magnitudes are a bit worrying. But I wouldn't have expected such differences in their capturing what browsers and OSs are used.
All three say Firefox is #1: Analytics says 49%, Stats says 42%, and StatCounter says 46%.
But Analytics and StatCounter say Chrome is #2, at 25% and 23% respectively; while Stats says #2 is IE, at 27%. #3 for Analytics & StatCounter is IE, at 9% and 15% respectively, with Stats saying Chrome is #3 at 13%.
I was thinking the difference is in duration over which the numbers are gathered, but it doesn't change much between last week and last month.
There's a similar inversion in operating systems: Windows is #1 all around, but StatCounter says MacOSX is #2 while Analytics & Stats give #2 to Linux. (Stats does so only indirectly, just saying the "OS" is X11; I wonder how many people stare at that wondering what X11 is.)
So, my take is that nothing's very reliable. But it's fun to fool around with. I'd be more concerned if I were getting any significant revenue from AdSense, since it's based on Analytics. But I'm not.
I am bothered by the type of explanation exemplified by the NY Times article Why So Blue?, which purports to answer the question “Why do ‘Martha Stewart’ chickens produce blue and green eggs?”, “Martha Stewart chickens” are the Arucana breed she favors. Why does she favor them? Not mentioned.
It says that the breed produces “a pigment called oocyanin, a byproduct of bile production,” which apparently permeates the shell. The more usual mottled or brown colors apparently just rub off on the way down the oviduct. The normal stuff is from the breakdown of red blood cells. Hmm. Yuck?
This is not an explanation. It merely says the eggs are blue because they have a blue pigment, a statement that is not spectacularly explanatory. If I saw a blue egg, I would know it already. It just puts a meaningless language tag (oocyanin) on the pigment. You might as well say those chickens have a blue-producing property. It explains nothing.
I’m obviously not searching for something like “It was willed by God / Allah / &your_personal_deity.” Then I’d want to know why she or he willed it, and the answer to goes into the dead end of deities being ineffable.
I guess I’m looking for something like the following paragraph. I know we (probably?) can’t do this yet, but at least a nod in its direction would be nice.
Arucanas have key genome differences from other chicken breeds that define them as Arucanas; list them, if its' not too boring. Those key differences were produced by selective breeding or environmental pressure or random genetic drift; specify how much of each was involved and what it consisted of if appropriate. Those key differences cause oocyanin production because (a) the organization of the genome is such that having this [specify] key difference must imply oocyanin production – the two are coupled. Or (b) there’s no connection to the key differences, they just happened at the same time by chance. Specify one of (a) or (b). The oocyanin produced gets into the eggshells by such-and-such a process.
That would tell me why the dang eggs are blue.
Explaining why the sky is blue is a whole lot simpler.
So, this friend of mine tweeted a link to someone’s prescriptions of how to write more: A Chapter a Day… I responded “Gregor advice - not my cup of tea. Especially the mornings part.” And one thing lead to another, and since I’ve written a book, what do I have to say about it? So here we are.
First, I’ve got nothing against what Gregor has written. He made an honorable new year’s resolution, and is trying to get himself to keep it. His advice is probably good and suitable for many people. Stop procrastinating. Use mornings. Write something each day.
One thing I will fault him on: He didn’t emphasize that these are things that work for him (presumably), and your own mileage will probably vary, if mileage you have and any desire to increase it.
Again, no reflection on Gregor; what he’s said is roughly normal, I assume, and it’s probably all good. But for me it’s damn Mary Poppins, cheerfully but firmly telling me how to do things so nice and right. Practically perfect. Cloying. Saccharine. Smiling church ladies happily living the straight and narrow, anxious for you to join them in their sweet, blissful surrender to the true path.
Makes me ill.
For me, writing is a much darker proposition. It involves intense fantasizing and serious mental momentum issues, along with obsessive categorizing and polishing.
And I hate mornings.
With that buildup, I obviously must now delve those dark spaces.
I have a very hard time getting myself going. On anything. My task-switching time is huge. I like to tell myself that’s because I have an over-average amount of mental state, keeping more things in mind at the same time than most people. But on reflection, that’s probably not true; otherwise I wouldn’t need the categorizing.
If I manage to pull myself away from whatever I was doing before, sometimes (not always) guilting myself into starting some actual work, I usually end up initially playing solitaire. I’ve played 1000s of games of solitaire. (And I win 18%. I think that’s high.) (And it makes no statistical difference whether you pick things from the deepest stack – I checked. Over 1000s of games.)
Once I tear myself away from that, and actually open a file of stuff being written, I usually start by doing small edits, like fiddling with commas, or splitting/joining sentences. This gradually progresses to larger and larger edits: rewriting phrases, moving paragraphs, changing inline lists to bullets or vice versa, and so on, until finally I imperceptibly segue into writing brand new stuff.
By then, I’m engaged. Hooked. My head’s embedded in what I’m doing. The only thing that interrupts me is the need for bathroom breaks. Hours on end I go, often deep into the night, until I can’t see straight any more – and even then, some of my best (well, funniest) writing happens when I’m groggy; apparently the filters are further down then. I just keep going – unless I get interrupted by something scheduled at some specific time that I really want to do, a Person from Prolock, as it were. Like right now: I was just told by my phone that it’s time to go to a Tai Chi class.
OK, I’m back.
Anyway, you can see why starting right up in the morning before phone calls, all chipper and perky, is not my cup of tea. I am not one of those people who can write between conference calls. I can’t even read between conference calls. My wife reads during commercials on TV. I can’t do that, either. I play solitaire.
I rewrite / edit / tweak everything. Always. Often. Continuously.
I have this basic rule: Did I enjoy reading that? If not, fix it. Is the content as correct as I can make it? What questions would I have about it that would need clarification? Beyond those, and more often than not, it’s something like: Is there something odd about the sentence structure that’s awkward or inhibits understanding? Something about the cadence, causing it not to fit with the content? Something about the paragraph structure?
Paragraph breaks are a great emphasis tool.
As illustrated. Anyway, I don’t necessarily think about any of this consciously, though. I just read it and usually tear the crap out of it.
Then I switch it all to another font, so the line breaks are in a different place and it just looks different, and go through it again. It’s amazing how many things show up when you do that.
(Posts to this blog tend to have less of this than my other writing.)
I am always imagining presenting what I’m writing to an audience. My writing is in no small part a transcript of what I would like to be saying – a perfect transcript, since it’s a presentation where, if I like, I can rewind time and change the way I said something.
I fantasize about such talks frequently. I lay in bed thinking about them. I rehearse them standing in line, sitting in an airport waiting for a plane, everywhere. In fact: The one thing that can get me going writing, fast, in the morning is having such a reverie and coming up with a particularly great way to get something across, a super analogy, a simple but clear diagram, something really good. Then I can’t wait to get it captured.
If such a brainstorm happens late at night, the only way I can get to sleep is to at least capture the essence on paper. Otherwise I’ll just lay there going through it, over and over, sleepless.
I simply can’t keep track of long lists of stuff. My mind isn’t built for it. This definitely includes masses of technical detail. Oh, I can remember bunches of stuff, but it takes ages for me to memorize it; it’s not a natural act.
A result of this deficiency – and I definitely count it as such; it makes me rotten remembering names, for example – I’ve learned to compensate by getting pretty good at boiling things down to something that captures the correct technical essence in a very few concepts. This may result in many fewer words, or it may result in more; the additional words may be needed to express an analogy, for example.
One result of continually having to boil things down is absolute frustration with most marketing materials. They're obviously written by people who really want (or should want) to do this, but don’t fathom that you actually do need to understand the technical part before you can express it simply.
I built this into one of the chapters in In Search of Clusters, by the way, where I talk about two kinds of people: People who can remember lists, and me. Or at least people like me. I spent a while polishing that explanation (of course), so I invite you to take a look there.
So ends this tale of mental deficiency, obsession, and fantasizing, a tale far from the sunny realms of most “how to get things written” soliloquies. Again, I’m not saying such things are bad. They’re good. Too good for me. Too good for me to take seriously, anyway. They doesn’t match my mental constitution.
Well, it looks like I'm connecting to a Tai Chi (taijichuan) class and group again.
I went for the first time to a class run at the Broomfield Rec center by Paul and Elizabeth (don't know their last names yet). Thanks to Kaiser, it's free for me -- not the major consideration, but a nice thing.
They also teach the same style I've been doing -- Chen Man Ching's short form. Their lineage connects more directly to "The Professor" (which is how they always refer to him) than what I've had before, as it runs through Ben Lo, a student of The Professor when he had a school on Taiwan, before going to NYC. I'm not sure how many levels are between Ben Lo and them. It may not matter much, since Ben Lo apparently comes up from his home in CA somewhere to give a seminar in the area, usually in Boulder (not surprising).
Their form is very close to what I've done. They have different names for some of the moves, and there are some variations in how they do some of them. I have a bit of worry there -- one of the most aesthetically pleasing moves seems to have gotten mushed -- but we'll see how it works out.
Anyway, Paul and Elizabeth seem to be OK folks, and they're connected in to the local Tai Chi community, something I value a lot.
I'm of two minds about the recent Supreme Court decision to eliminate all current and historical constraints on corporate and union funding of political campaigns.
On the one hand, legislating against it has never really been fully successful; ways around the prohibitions have always been found. I think there's an analogy here to the outlawing of any behavior perceived as bad, like owning guns, drinking booze, or smoking marijuana. It's always done anyway, at least in the absence of truly draconian measures that are more than people can stomach in the long term.
On the other hand, all constraints? Wow.
What about sponsorships? I can see it now: "This mayoral election is brought to you by AIG, helping to insure freedom to sell job-friendly stock derivatives!"
Or even naming rights: "Welcome to our coverage of the Swiffer Sweeper Congressional Elections, dedicated to cleaning house!"
Prohibiting any of this would be depriving the companies of a form of free speech, right? And don't forget, elections are expensive. I'm sure there are lots of cash-strapped municipalities who would welcome some help covering the cost.
I've been furnishing the office in my new house, and one thing I wanted to do was provide an adequate level of lighting (read: bright!), while not roasting the roof or using more electrons than necessary. Hence: compact fluorescent lights (CFLs).
A neat thing about CFLs is that you can get a light fixture that is officially rated as only taking, say, a 50W bulb; and stuff it with "100W" CFLs, since they only use about 23W of electricity.
More good news is that CFL prices are diving faster than an peregrine falcon. Home Depot has "100W" CFLs from Phillips for $1.74 each. Wow! A far cry from the days of puny "40W" CFLs costing $7.
Now comes the bad part. Be sure to read the fine print. No two manufacturers seem to have agreed on basic things like how much light a "100W" CFL emits, or what actual color a "soft white" CFL emits. Or whether it works at all.
I found CFLs labeled "100W" rated as emitting anywhere from 1400 to 1650 lumens. That's nearly a 20% difference. All are comfortably higher than the light emission from some 100W incandescents I had hanging around the house, but egad.
And having bought two of the Philips ones (1600 lumens), I found one only lit up half of the length of tubing. So, I returned it. Figuring that was a fluke, while in the store I picked up a four-pack of the same. But when I tried using them, not one worked. They didn't turn on at all. I was blaming the lamp, since it was new, too -- how could all four be bad? -- but then plugged in an incandescent bulb, which worked.
What the heck?
Tried a different brand -- "GreenLite Mini" -- and it worked, too.
I think it may be a slightly different socket shape on the Philips bulbs. But why the singletons I bought worked, and the four-pack didn't, who knows.
Anyway, I'm hoping to get a bunch of the GreenLites now that I know they work. I also prefer their color -- they actually specify a color temperature: 2700K, which is rather on the yellow, "warm," side, but not as yellow as the Philips. Unfortunately, I only found the GreenLites at a specialty lighting store, but the price was still fine: About $5 for a four-pack.
I really wish someone would make a CFL that's wide rather than tall. Then that ugly tubing wouldn't stick out of fixtures so much.
There are in English several situations where the words for animals differ from the words for their meat: Cow vs. beef, calf vs. veal, and so on. I've heard, repeatedly, that this came from the Normal Conquest: The Anglo-Saxon serfs raised the animals, calling them by more Germanic words; but their Norman French lords ate the meat, and used the French words for what they ate. People seem to like to trot out this explanation to display their erudition.
Well, it turns out that it's wrong.
Stephen Pinker, in The Stuff of Thought, says that this theory originated in Sir Walter Scott's Ivanhoe, when Wamba the jester, explained it to a swineherd ("Mynheer Calf... becomes Monsieur le Veau..."). Pinker references Burchfield's The English Language as stating that "Anglo-Saxon and French words didn't sort themselves out until centuries later."
I don't have a copy of Burchfield's book to check this (I'll be fixing that), but frankly I don't understand what Pinker's "sorting" statement means or how it relates to the issue.
A quick google, however, came up with a discussion on AskMetaFilter where others pointed out that many languages refer to animals with different words than those used for food, for example: Japanese ushi (cow) vs. gyuuniku (meat from a cow); Spanish gallina (chicken) vs. pollo (chicken meat).
Since there was nothing analogous to a Normal Conquest for those other languages, I think the French-origin legend is toast without getting Burchfield involved. But this does raise a question: Why is there a general, cross-language tendency to use a different word for an animal than for its meat? One possible very direct answer: Because they're different things. Duh.
So apparently Slate has picked up on the Domino's ad where they quote allegedly real customer comments slamming their pizza, like "Domino's pizza crust to me is like cardboard" and "The sauce tastes like ketchup."
A shame nobody mentioned that Domino's pizza really does have crust like cardboard, sauce like ketchup, and, unmentioned, little hamburger balls like soft marbles.
Ugh. Horrible stuff.
I've no clue whether they actually did change it in any discernible way.
I found myself wanting to post things longer than a tweet or a FB entry, but not within the professional confines of my other blog, The Perils of Parallel (which is a really good technical blog about multicore, clouds, accelerators, virtual worlds, and similar stuff that, if you're into such stuff, you really should look at).
Hence this blog.
My family, at least, will recognize the reference to "gorp."