Chronicles of a Human Programmer

More things I probably should have been aware of, but wasn't

Sun, 06 May 2007 08:15:37 GMT

From an ad in The Economist:

Understand the Fun and Beauty in Mathematical Concepts

Humans have been having fun and games with mathematics for thousands of years. Along the way, they've discovered the amazing utility of this field--in science, engineering, finance, games of chance, and many other aspects of life.

This course of 24 half-hour lectures celebrates the sheer joy of mathematics, taught by a mathematician who is literally a magician with numbers. Professor Arthur T. Benjamin of Harvey Mudd College is renowned for his feats of mental calculation performed before audiences at schools, theaters, museums, conferences, and other venues.

Throughout these lectures, Professor Benjamin shows how everything in mathematics is magically connected--how the beautiful and often imposing edifice that has given us algebra, geometry, trigonometry, calculus, probability, and so much else is based on nothing more than fooling around with numbers.

This course is one of The Great Courses, noncredit recorded college lecture series from The Teaching Company. Award-winning professors of a wide array of subjects in the sciences and the liberal arts have made more than 200 college-level courses that are available.

Being a sucker for such things, I bought it.

Small World (Angband Edition)

Wed, 22 Nov 2006 04:11:47 GMT

Chris Weisiger (author of the Angband Newbie Guide, among other things) also works at Amazon. My user icons have never been more appropriate.

Returning to Meme

Fri, 17 Nov 2006 04:18:13 GMT

Sometimes, the attraction of SF is just too great. Especially since I'm currently engaged in rereading one of the series on the list.

The Lord of the Rings, J.R.R. Tolkien *
The Foundation Trilogy, Isaac Asimov *
Dune, Frank Herbert
Stranger in a Strange Land, Robert A. Heinlein
A Wizard of Earthsea, Ursula K. Le Guin *
Neuromancer, William Gibson
Childhood's End, Arthur C. Clarke
Do Androids Dream of Electric Sheep?, Philip K. Dick
The Mists of Avalon, Marion Zimmer Bradley
Fahrenheit 451, Ray Bradbury
The Book of the New Sun, Gene Wolfe ** (currently re-reading)
A Canticle for Leibowitz, Walter M. Miller, Jr. **
The Caves of Steel, Isaac Asimov
Children of the Atom, Wilmar Shiras
Cities in Flight, James Blish
The Colour of Magic, Terry Pratchett
Dangerous Visions, edited by Harlan Ellison
Deathbird Stories, Harlan Ellison
The Demolished Man, Alfred Bester
~~Dhalgren, Samuel R. Delany~~
Dragonflight, Anne McCaffrey
Ender's Game, Orson Scott Card
The First Chronicles of Thomas Covenant the Unbeliever, Stephen R. Donaldson
The Forever War, Joe Haldeman *
Gateway, Frederik Pohl
Harry Potter and the Philosopher's Stone, J.K. Rowling [Waiting for the series to be finished before I start it.]
The Hitchhiker's Guide to the Galaxy, Douglas Adams
I Am Legend, Richard Matheson
Interview with the Vampire, Anne Rice
The Left Hand of Darkness, Ursula K. Le Guin
Little, Big, John Crowley
Lord of Light, Roger Zelazny
The Man in the High Castle, Philip K. Dick
Mission of Gravity, Hal Clement
More Than Human, Theodore Sturgeon
The Rediscovery of Man, Cordwainer Smith [On the list to read, but moderately far down.]
On the Beach, Nevil Shute
Rendezvous with Rama, Arthur C. Clarke
Ringworld, Larry Niven
Rogue Moon, Algis Budrys
The Silmarillion, J.R.R. Tolkien **
~~Slaughterhouse-5, Kurt Vonnegut~~
Snow Crash, Neal Stephenson
Stand on Zanzibar, John Brunner *
The Stars My Destination, Alfred Bester *
Starship Troopers, Robert A. Heinlein
Stormbringer, Michael Moorcock
The Sword of Shannara, Terry Brooks
Timescape, Gregory Benford
To Your Scattered Bodies Go, Philip Jose Farmer

Also, today was the first sunny day in Seattle this month. Sometimes I like Seattle, and occasionally it reciprocates.

Tilting at Windmills

Fri, 10 Feb 2006 06:21:26 GMT

I'm currently repressing the urge to rewrite everything in Common Lisp. Even if it _would_ solve hunger.

Vacation from my vacation

Mon, 30 Jan 2006 07:07:05 GMT

Amazon's post-holiday party was Saturday night. After being in the presence of so many people for so long, I'm ready for my post-post-holiday-party book. Something quiet, at least.

Sagging shelves

Sat, 21 Jan 2006 06:25:35 GMT

It's a sad, sad day when you have to weigh your books to make certain your bookcases can actually support their weight. Hopefully my next set of bookshelves won't have this problem. I'm going to stop by Ikea this weekend, I think, and pick up some of these. Assuming I can find my way there (my sense of direction isn't exactly what would be called 'reliable').

Working at Amazon has, so far, been pretty nice. The hours are a bit long, but the work is varied, and there's a little bit of everything to do. Plus I've been taking advantage of the employee discount. My group is quite new (all joined Amazon within the past six months), so we're all getting used to the Amazon way of doing things together. My first piece of code went live on the website last week, which is somewhat of a rush. So far most of the difficulty hasn't been in the problems themselves, but rather in determining how pieces of code are supposed to fit together and which of the seven steps the documentation says to follow are correct and mandatory, which are merely optional, and which are downright wrong because the package/operating system/framework/library changed fifteen months ago and nobody bothered to update everything which discusses it.

On the plus side, I'm reading more again, aided along by my bus rides to and from work. Having twenty-five minutes each way of dedicated time assigned to a task is definitely conducive to getting it done.

Back?

Tue, 20 Dec 2005 05:58:19 GMT

New location (Seattle). New job (Amazon). Maybe a new life as well; we'll see. It's odd to be back.

Vanity searching gone horribly wrong

Thu, 29 Jan 2004 03:39:19 GMT

And yes, my middle inital is 'H'.

Other notes: Installed Afterstep 2.0.0 beta 3. Except for the fact that my Angband windows no longer autoposition correctly, I like it a lot. It's now much easier to customize. I don't like what they did with the WinList module (it sits in the middle of the screen and prevents any window from opening on top of or abov it), but since I don't use that module any more it's not that big of a deal. My virtual destops have also changed from 3 × 2 × 8 to 4 × 3 × 4 for easier navigation.

Hmmm... the screenshot doesn't seem to have captured the pager module. Oh well.

I also finally got around to sending in the rebates for my computer, with about a week left on the 30 day allowable period. Hopefully they will be honored. There shouldn't be any problem, but...

And I really need to automate the portion of my music ripping which is 'make a playlist and stick into the database.' Though it seems like my solution to nearly any data-collection or data-analysis problem these days is to stick it into a database.

Infrequent Update

Thu, 15 Jan 2004 06:00:17 GMT

Angband and comics and dark sun and computational fluid dynamics and databases and research and computers and angband and food and ellipses without end.

First Entry of the Year

Wed, 07 Jan 2004 07:13:36 GMT

(for me, at least).

Vacation was good. I had jury duty the week of Christmas, but never actually had to report to the courthouse. I also got to see my parents again and my sister for the first extended period since last Christmas.

I also got a new computer (Sony laptop) and am now involved in the process of installing on it everything I wish, so that it becomes a newer, better, more transportable version of this computer.

Primarily, though, right now I'm rather tired since I didn't get enough sleep last night.

XSLT is also about having a big penis

Wed, 10 Dec 2003 05:55:58 GMT

Well, that's not quite correct. Really, it's about having a big Xpenis.

Sorry, I just felt the need to share that.

For some reason.

Yeah.

Various Updates

Fri, 21 Nov 2003 09:39:26 GMT

The container security paper got rejected by Science, so we're trimming it down to submit to PNAS. Of course, the appendix is already around four times the size of the main paper, and this will just make it worse. It's almost as if the paper is merely a teaser.

Finished up work on the paper I working on before this one, so that should be submitted somewhere too. Most of the work was done by Professor Wein's student at MIT (he was at MIT before coming to Stanford), so I'll only be third author on that one.

Now I'm working on yet another paper, which will (hopefully) be an analytical analysis of the model used in the container security paper. We're simplifying the model somewhat, however, so that it is analytically more tractable--smoothing out the step functions, assuming one queue with varying speed rather than a variable number of queues each with fixed speed, etc. Professor Wein and I both hope that this will form at least a portion of my eventual dissertation.

Is this a good thing?

Tue, 11 Nov 2003 14:42:22 GMT

I realised when composing an email to my advisor that the problems I had been having were because I forgot one of the basic premises of the model while looking at the equations.

Basically, we're using a Gaussian plume model to simulate a release of anthrax spores, and I was trying to calculate the amount of spores that will reach a sensor at a given position. Now, if this position is downwind of the release, there are no problems. If it is upwind, you end up trying to calculate a negative number to the 2 * d power, where d = 0.78. This, however, tells you that you get both real and imaginary spores at your sensor, which I realised couldn't be correct. So I instead took the negative number, squared it, and took it to the dth power. Which is perfectly reasonable, except that the madel doesn't support that interpretation, and in fact the number of spores at any location upwind of the release is zero.

Oops.

(For what it's worth, if there is no wind, the whole model just goes kaput.)

But it's better now. I hope.

This is what you get when you come back to a project after several months of working on something else.

Sloth and Indolence

Sat, 08 Nov 2003 17:00:54 GMT

For those of you who are curious, I have attached the results of the matlab run of earlier in the week.

bytes used=326690544480, alloc=1330923488, time=106159.25
bytes used=326695008676, alloc=1330923488, time=106164.70
bytes used=326699017364, alloc=1330923488, time=106169.07
bytes used=326703193956, alloc=1330923488, time=106174.14
bytes used=326707656524, alloc=1330923488, time=106179.16
bytes used=326712311756, alloc=1330923488, time=106184.62
bytes used=326716955284, alloc=1330923488, time=106190.11
bytes used=326746458780, alloc=1331120060, time=106194.98
bytes used=326751018456, alloc=1331120060, time=106200.68
bytes used=326755714696, alloc=1331120060, time=106206.34
bytes used=326760309388, alloc=1331120060, time=106211.75
bytes used=326764914232, alloc=1331120060, time=106217.16
bytes used=326768915700, alloc=1331120060, time=106221.45
bytes used=326773383764, alloc=1331120060, time=106227.07
bytes used=326778221844, alloc=1331120060, time=106232.73
bytes used=326782813216, alloc=1331120060, time=106238.16
bytes used=326787478124, alloc=1331120060, time=106243.60
bytes used=326791659080, alloc=1331120060, time=106247.94
bytes used=326796349552, alloc=1331120060, time=106254.01
bytes used=326801312444, alloc=1331120060, time=106260.05
bytes used=326806107040, alloc=1331120060, time=106265.69
bytes used=326810772488, alloc=1331120060, time=106271.33
bytes used=326814953420, alloc=1331120060, time=106275.68
bytes used=326819633872, alloc=1331120060, time=106281.88
bytes used=326824680208, alloc=1331120060, time=106288.05
bytes used=326829462188, alloc=1331120060, time=106293.72
bytes used=326834238596, alloc=1331120060, time=106299.46
bytes used=326838625980, alloc=1331120060, time=106303.88
bytes used=326865528812, alloc=1331316632, time=106308.59
bytes used=326870094568, alloc=1331316632, time=106314.35
bytes used=326874692680, alloc=1331316632, time=106320.02
bytes used=326879274548, alloc=1331316632, time=106325.49
bytes used=326883756800, alloc=1331316632, time=106330.93
bytes used=326887758680, alloc=1331316632, time=106335.29
bytes used=326892267156, alloc=1331316632, time=106341.01
bytes used=326897044088, alloc=1331316632, time=106346.71
bytes used=326901653724, alloc=1331316632, time=106352.34
bytes used=326906260828, alloc=1331316632, time=106357.92
bytes used=326910265632, alloc=1331316632, time=106362.36
bytes used=326914889588, alloc=1331316632, time=106368.57
bytes used=326919715432, alloc=1331316632, time=106374.71
bytes used=326924483848, alloc=1331316632, time=106380.35
bytes used=326929036732, alloc=1331316632, time=106386.06
bytes used=326933221480, alloc=1331316632, time=106390.50
bytes used=326937983984, alloc=1331316632, time=106396.68
bytes used=326943064260, alloc=1331316632, time=106402.81
bytes used=326947859228, alloc=1331316632, time=106408.54
bytes used=326952599788, alloc=1331316632, time=106414.42
bytes used=326956720384, alloc=1331316632, time=106419.11
bytes used=326983549816, alloc=1331513204, time=106424.00
Execution stopped: Memory allocation failed.
bash-2.04$

I joined Project Gutenberg's distributed proofreading project and am now in the 84th percentile. I have a reasonable suspicion that the total number of pages done by a person is exponentially distributed, but I'd have to take a closer look at the data to make certain.

Fustian is not faustian.

Because it's been a while and I already have one lj-cut here, my current music rankings.

name	song	score	count
Peter Tchaikovsky	1812 Overture	23172	50
Jean Sibelius	Violin Concerto in D minor, Op. 47	20757	19
Nikolai Rimsky-Korsakov	Sheherazade	20454	47
Blind Guardian	The Maiden and the Minstrel Knight	18749	51
Blind Guardian	Time Stands Still	14142	64
Igor Stravinsky	Le Sacre du printemps	13446	23
Final Fantasy VI	Ending Theme	13081	45
Final Fantasy IV	The Lunarians	12663	36
Final Fantasy IV	Epilogue	12247	44
Peter Tchaikovsky	Marche Slave	11613	42

They seem more stable now, except when I add in new CDs. My four newest CDs of Sibelius and three of Bruckner are not yet fully integrated (not that you can tell from the above data, since the only thing from any of them is Sibelius' Violin Concerto in D Minor). I also finally went in and turned off playing of duplicate performances of various songs--I had three copies of Beethoven's 6th, for example.

My FreeCiv 'Pave the World' game finally broke 7 billion in population. Apparently, the newest versions of the client also use much more memory than they should, since while playing I ran out of swap space; civclient had been using 370 MB of memory (out of 512 MB total) and 1.7 GB of swap (of 2 GB). Probably some memory leaks, too. Or maybe I just shouldn't build civilizations that large.

In other news, my paper (well, I'm second author out of four, meaning I did the majority of the grunt work) has been submitted to Science. I should know by Monday whether or not they'll bother send it out for review. Apparently the lead editor was once associated with Stanford when my advisor was a student here, so that's a plus. In the meantime, I'm working on the next project (asymptotic analysis of an anthrax infection model). Or rather, not working on it. But that's a technicality.

And for those of you taking GREs: in my experience, they were just like the SATs, except that the fact that since you can take them on a computer, if you are your own ride then you can leave two hours early with no problems. Or maybe that's just me.

The other users must hate me

Tue, 04 Nov 2003 17:09:27 GMT

bytes used=123742121132, alloc=89636832, time=16469.98
bytes used=123746720464, alloc=89636832, time=16470.66
bytes used=123750797532, alloc=89636832, time=16471.16
bytes used=123755042348, alloc=89636832, time=16471.88
bytes used=123759258692, alloc=89636832, time=16472.50
bytes used=123763341908, alloc=89636832, time=16473.31
bytes used=123767532600, alloc=89636832, time=16473.98
bytes used=123771706912, alloc=89636832, time=16474.71
bytes used=123775861124, alloc=89636832, time=16475.76
bytes used=123779938660, alloc=89636832, time=16476.51
bytes used=123784136356, alloc=89636832, time=16477.37
bytes used=123788211628, alloc=89636832, time=16477.85
bytes used=123792677892, alloc=89636832, time=16478.56
bytes used=123796697880, alloc=89636832, time=16479.37
bytes used=123800804368, alloc=89636832, time=16480.14
bytes used=123805292296, alloc=89636832, time=16481.02
bytes used=123809435636, alloc=89636832, time=16482.06
bytes used=123814627264, alloc=89636832, time=16482.60
bytes used=123818809796, alloc=89636832, time=16483.60
bytes used=123823237080, alloc=89636832, time=16484.44
bytes used=123827323972, alloc=89636832, time=16485.43
bytes used=123831651784, alloc=89636832, time=16486.27
bytes used=123836245596, alloc=89636832, time=16487.16
bytes used=123840445528, alloc=89636832, time=16488.12
bytes used=123845058960, alloc=89636832, time=16488.82
bytes used=123849220424, alloc=89636832, time=16489.69
bytes used=123853463936, alloc=89636832, time=16490.42
bytes used=123857478468, alloc=89636832, time=16490.85
bytes used=123861591068, alloc=89636832, time=16491.83
bytes used=123865822912, alloc=89636832, time=16492.63
bytes used=123870171140, alloc=89636832, time=16493.67
bytes used=123874190368, alloc=89636832, time=16494.43
bytes used=123878682268, alloc=89636832, time=16495.30
bytes used=123882870436, alloc=89636832, time=16496.27
bytes used=123886928096, alloc=89636832, time=16496.89
bytes used=123891073048, alloc=89636832, time=16497.70
bytes used=123896199456, alloc=89636832, time=16498.60
bytes used=123904126288, alloc=89636832, time=16498.95
bytes used=123908303852, alloc=89636832, time=16499.73
bytes used=123912520536, alloc=89636832, time=16500.78
bytes used=123916563408, alloc=89636832, time=16501.52
bytes used=123920946376, alloc=89636832, time=16502.61
bytes used=123925118636, alloc=89636832, time=16503.38
bytes used=123929526104, alloc=89636832, time=16504.43
bytes used=123935384408, alloc=89636832, time=16505.26
bytes used=123939388908, alloc=89636832, time=16506.35
bytes used=123944242172, alloc=89636832, time=16507.40
bytes used=123952743936, alloc=89636832, time=16507.87
bytes used=123956932184, alloc=89636832, time=16509.01
bytes used=123961300180, alloc=89636832, time=16509.74
bytes used=123965610424, alloc=89636832, time=16510.90
bytes used=123969737976, alloc=89636832, time=16511.63
bytes used=123974047024, alloc=89636832, time=16512.76
bytes used=123978670688, alloc=89636832, time=16513.69
bytes used=123982810880, alloc=89636832, time=16514.33
bytes used=123987074412, alloc=89636832, time=16515.35
bytes used=123991816228, alloc=89636832, time=16516.05
bytes used=123996664924, alloc=89636832, time=16516.34
bytes used=124000919168, alloc=89636832, time=16517.20
bytes used=124005327772, alloc=89636832, time=16518.16
bytes used=124009642080, alloc=89636832, time=16518.84

Way too much CPU time to be used in computing a single integral. I can't believe I stayed up for this. Ugh. At least it won't be nearly as cold for my trip back to the apartment.

Passive Selfdestruction

Sat, 25 Oct 2003 12:00:10 GMT

I read A Deepness in the Sky yesterday and today, which is one reason why I'm still up, even though I woke at a reasonable hour.

I would think that, given the choice, I might become Focused, except that I'm effectively declining that option by my actions here in grad school. Well, putting off the decision, really, but it amounts to the same thing. Still, it would make things easier.

I checked, and it appears I've only read a little under 18K pages so far this year, which means that I'm unlikely to make 25K for the entirety. Too many months of net-reading only, I guess, to be counteracted by a last futile sprint, but you never know. If I can keep up 100 pages a day, I should be in good shape, and I do generally read more when at home, so Christmas break should be productive (for certain values of).

I had to fetch the apartement manager and get a part of my deadbolt replaced Thursday night. It's rather annoying to be locked out when you have your key in your hands. Certainly I hope the problem doesn't recur tonight, as it's almost certainly too late to do anything about until morning. Of course, I may not be heading back until morning, either.

Even fifteen minutes of separation between work and home hasn't stopped my problems with late nights, obviously. It's just altered them somewhat.

Bunch of Stuff

Fri, 17 Oct 2003 10:57:46 GMT

Went to the Ren Faire on Sunday. I think it takes someone rather more outgoing than I am to make the most of one, but it was fun nonetheless. Traffic, though, was atrocious. And I picked up a pound of pecans afterwords, which make a good early morning/late night snack.

My first paper is close to being submitted; I'll be generating what is probably the last of the real data for it tonight. It'll be nice to have some more time again.

I've turned to ToME in an attempt to get my first win at Angband, but all that has really changed is that I now get annoying high level deaths rather than annoying mid-level deaths.

I'm also looking through the Project Gutenberg archives to add some of their books to my list. I do need a better method of estimating page length, though--lynx says that part I of Spenser's Fairie Queen is around 4600 pages. I also probably need a better method of reading them than 'less', as plain ASCII (although eminently portable) is hard on the eyes. I also really don't want to be printing them out, since I would feel guilty using that much of my department's ink and paper.

My music rankings seem to have settled down. On the logarithmic scale for scores, the songs seem distributed in a pear shape, with the majority below the mean and a select few above it. Setting the probability of a song being played to be proportional to its current score over the total number of times it has been played means that every song on a new CD gets played several times at least before being consigned to the bit bucket.

I probably need to go to bed, but I'm not at all sure that I will.

Proposal for an Angband Variant

Sat, 04 Oct 2003 08:49:12 GMT

Serialband:

The basic premise would be that an adventurer's actions can and do have an impact upon the way the dungeon in laid out for later characters. Some examples:

Some carryover of equipment/gold from character to character. Only that which is given away before death, though.

Uniques are unique, even across characters. That is, you can only kill Maggot (or Ferny, or whatever) once per save file, and after that, he's dead for good.

Random uniques, generated either when a character is killed (the killer becomes a unique) or when a new character is started and too few uniques remain.

Similarly, artifacts are also unique across characters. For the most part, this won't matter too much, except that disenchantment now becomes even more of an annoyance, and scrolls of curse weapon/armor may actually be useful (as the only way to permanently get rid of artifacts you don't want).

New random artifacts created at character death to replace those (if any) destroyed during that game. These should be generated based upon the dead character's non-artifact equipment, if possible.

If enough of a single type of monster are killed in a short enough period of time, that monster goes extinct (a la Nethack).

There are some other things I'd like to put in as well (shopkeepers who actually retire when they say they're going too, for example), but those are the basics.

In other news, I am still alive, even though at times I feel I'm not.

And school has started again, though I'm not taking classes (just doing research).

And I seem to be temporarily thrust into a semi-leadership role at my eating club, due to seniority.

Oh well. At least I've collected my back pay for the summer.

Another day, another entry

Mon, 01 Sep 2003 06:26:13 GMT

I read through Shiras' Children of the Atom on the plane yesterday. It was a short story I had already read, expanded into a novel by adding four additional chapters. The basic premise is that a nuclear accident causes the children of the victims to develop vastly superior intelligences, and these children must learn how to hide themselves in 'normal' society. It's okay--little conflict or directed plot, and the author's own biases are astonishingly clear (all the well-adjusted children are or become Thomists, for example), but still an easy enough read. I also had finally read Storm of Swords over two nights earlier in the week. Really, I think I shouldn't read series unless the individual volumes are relatively self-contained or the series has has all of its elements already published in paperback, because the delay between volumes is extremely annoying. Another reason to read more Guy Gavriel Kay, I guess, except that I'm pretty sure I've already seen everything he's written. But at least I'm not anxiously awaiting the next volume in an epic of his.

I will finish Ulysses tonight, unless something goes horribly wrong. Only six pages remain, and while they comprise the ending of a sentence begun 40 pages ago, I CAN force myself through. Really. After this, the longest time since starting a book I've begun reading will be only around two months, for a collection of Checkov's plays.

One good thing about moving my computer from my apartment to my office is that my default evening activity is now reading rather than net surfing, which should be at least marginally more intellectually stimulating, even with my preferred choice of reading material. I have enough Russian classics in my upcoming book schedule that I should get something out of them (even if only an intense hatred of all translators).

My mother flew up to Santa Rosa today to start scheduling the funeral for her mother. My dad and I will follow once we know the date and time, which gives my at least some time to placate my advisor before leaving. I do need to make some optimizations in my computations before I go, though, because currently my programs are spending too much time on the redundant or the impossible and then calling themselves completed too early on the cases on which they should be spending the most time.

The past two weeks

Sun, 31 Aug 2003 06:18:26 GMT

So I moved fully into my new apartment; then, a week later, finally got some furniture (a bed, dresser, bookshelf, table, and five chairs, to go with the two couches and coffee table that my roommate already had). My roommate's name is Paul, and he seems to be a nice enough guy, but I expect that my relationship with him will be similar to my relationship with Peter (my freshman year roommate): we get along well enough, but we hardly ever interact. I mean, either he leaves before I get up or I leave before he gets up, depending on whether I set my alarm. At night, neither of us typically gets back before 10:00, and even then we tend to closet ourselves in our respective rooms before heading to bed.

After a couple of weeks of living in it, the apartment seems nice enough, although it has some noticable problems (the fact that the fan wiring in my bathroom is a fire hazard, for one, or the extensive stains on the carpet). Given that I've moved my computer into my office and am mainly using the apartment as a place to sleep and store stuff, it should work well enough, though. I just wish the Bay Area housing market would cool substantially.

My current research project is almost done--the model appears to be finalized and I'm onw generating data. I just wish I could generate data faster--I've been computing for more than a day now, and I'm not sure when it will finish. Hoepfully soon.

I just flew home today to start two weeks of vacation, which should have been nice.

My grandmother died tonight.

It's not as though it's entirely unexpected (she's been on the decline since her first stroke more than three years ago), but stilll...

I'm going to be flying back up to the Bay Area with my parents to go to the funeral either tomorrow or Monday (depending on when it will be held).

Moving

Sun, 17 Aug 2003 06:05:10 GMT

With my family's help, I moved from my dorm room to an off-campus apartment yesterday. I'm not completely moved in yet--I still don't have furniture, for example--but the majority of my stuff has been transported. One problem I've run into, though, is htat the apartment only has two-prong outlets, rather that three-prong ones, so I don't have anywhere to plug my computer. I bought some grounding adapters earlier this evening, so hopefully this will fix the problem. However, as I'm composing this from the dorm computers, I don't know whether the outlets in my new apartment have the differently-sized slots needed by the grounding adapter. If they don't, I'm going to have to move my computer into my office, which will mean I can't access it after I leave school for the night (since I still don't have a car and really don't wish to purchase one, I become much less mobile after dark).

Theoretically, I alos have a rooommate in the apartment, but given that last night and this I will be sleeping in my dorm room (as I don't have a bed in the apartment yet), I haven't met him yet. I do know he is there though; sometimes, a door that I know was open when I left will be closed upon my return. The thing is that this roommate is supposedly moving out sometime between now and the start of the next semester, so I will be getting another new roommate after that. On the other hand, this will hopefully give me a chance to move all my stuff to the master bedroom before the new roommate comes.

Other than that, not too much is happening. I'm still working on my research, and hopefully the underlying model has undergone its final changes, so that I no longer have a moving target against which to code. I do think we are close to have actual results, though, and it will be nice to be done with this paper.

But I want my own computer back, not this one-button Mac with an absurdly slow repeat rate on the keyboard and a pointer that won't follow where I attempt to lead!

LJ is quiet. Too quiet.

Sun, 03 Aug 2003 11:58:56 GMT

Given that the last entry on my LJ friends page is over a day ago, I figured I should write something. Or add more people to my friends page. Or both.

Anyway, I think I've finally figured out (after 8 years!) where my science fiction book list comes from. Unfortunately, because I figured it out at 2100 yesterday, I can't check until the library opens (1300). And given that I might not still be up then, that means I may not find out for certain until Monday. Grrrr...

Relearning PHP. It seems to be like perl, except where it's not. I haven't tried regular expressions yet, though.

I think that's about it. I do need to figure out the best way to force my research code not to keep venturing into infeasible regions, though. You can't have an inflow rate greater than the outflow rate and expect to have a finite steady state waiting time! Stupid optimizer.

There was something I wanted to write, but I don't remember what it is, so I won't.

Genericness

Thu, 31 Jul 2003 05:45:59 GMT

Making progress on research. Damn calculus--it the keeping track of constants that is really annoying, not the actual integration.

Changed journal to use the S2 style system because I noticed Mary's comments page matched her journal. I would like to customize everything so that it looks like it used to, but it apperas that that will have to wait. At least I now get all my various user pics on my main journal page, though.

Installed php for the first time in two major versions so that I could use phpPgAdmin. I also have an idea for another programming projet i'd like to do using it (a very specific interface to a database), but we'll see if that actually goes anywhere. One benefit of phpPgAdmin: cut and pasting my current top ten music list:

name	song	score	count
Peter Tchaikovsky	1812 Overture	24819	37
Antonin Dvorak	Symphony #9	19035	36
Jean Sibelius	Violin Concerto in D minor	17707	34
Blind Guardian	The Maiden and the Minstrel Knight	14513	35
Blind Guardian	Time Stands Still	14443	53
Jean Sibelius	Concerto for Violin and Orchestra in D minor	12066	21
Simon and Garfunkel	The Boxer	11159	36
Edvard Grieg	Peer Gynt	10700	38
Gustav Mahler	Symphony #1	10317	28
Hector Berlioz	Requiem	10083	27

The biggest change (at least to my eyes) since the last time I posted is the rampant score inflation. It seems to have leveled off now, though.

I really need to start updating this journal more often than three times a month. That doesn't mean that it's going to happen, though.

No, I'm not dead (yet)

Mon, 28 Jul 2003 03:55:31 GMT

My mother asked me the same question.

Anyway, research is continuing. An additional student was also assigned (temporarily) to the project, so he can check to see where I've made mistakes. And given that I'm integrating something that purports to be a probability distribution and the result is 1.57, there is a mistake somewhere. Hopefully, though, once this problem is resolved we will be able to generate some actual results, since everything else seems to be working.

We finished the basic 14 quests in HeroQuest and started on Kellar's Keep. No notable developments; my only recent kill was of the dwarf in quest 10. And unfortunately it looks like my dreams of crushing all four heros with a boulder probably won't come true, as they have come to the conclusion that plate mail (and the resultant reduced movement rate) is a bad idea and therefore will likely be able to outrun it. Oh well.

I'm making more progress into Ulysses: 79% done, as of today. If I can keep reading, I could be done by the end of August. (Of course, if I could keep reading I would have been done by April. Of last year.) In better news, it appears that Solzhenitsyn (Gulag Archipelago, specifically) is much more readable than Joyce. My supply of fluff books is running rather low, though. Which reminds me: I should post the science-fiction book list I'm working off of on Wiki at some point.

I also bought Diceland: Deep White Sea from James Earnest Games (the division of Cheapass Games which sells things which cost more than 7 dollars). Haven't had a chance to really play it yet, though. Too much HeroQuest. Only got to play one game against Anna, which I won handily through greater knowledge of the rules, and then Jon sat on and crushed one of the pieces. Grrrrr. So the blue team now has a leader who throws oddly.

I think that's about eveything that I'm not attempting to block off from memory.

Sequel

Sat, 05 Jul 2003 12:51:20 GMT

Updated to Logjam 4.2.2. Hopefully, this version handles umlauts in song titles better than the 4.1.x series did. After all, eveyone needs their USRDA of umlauts. Mmmm ümläüt. Äñd mörë ümläüts, jüst becäüsë.

Well, I was going to include a summary of more of the HeroQuest stuff in here, but I'm apparently heading off to a party in about twenty minutes (edit--fourty minutes. Jonathan ran out of gas.), so I probably won't have time to finish (edit--And I didn't. So this is being posted now rather than say ten hours ago). Well--there's always tomorrow.

Anyway,

Quest 1, Attempt 3

Before this attempt, I made two suggestions, since I was getting rather sick of running the same quest over and over again. The first was that since they obviously knew the location of the gargoyle (given the way that they conspicuously did not open the room in which it laired, one of only two rooms that had been unexplored during the previous attempt), they should kill it first, before going in and exploring everything else. That way, if something disasterous happens and they ended up running for the stairs, even if only a few survived they would still have successfully completed the quest and I wouldn't have to run it again. Second, they should limit the random treasure searching to once per room, rather than once per character per room. Given how they had been savaged by wandering monsters in the past, this should hopefully improve their survivability (while, not coincedentally, reducing the amount of treasure they would find--he he he).They apparently took my suggestions (commands for on high) to heart.

Anyway, this time Brad played the barbarian, Anna the dwarf, and Jon the elf and wizard. Upon beginning the quest, three heros went off to kill the gargoyle, while the dwarf headed toward the other unexplored room. Luckily, she didn't get too severely mauled in there, as it only contained a goblin and an orc. She then headed back toward the rest of the party, which seemed to have grasped the concept of tactics for the first time: they actually had the barbarian open the room, giving them four turns before I got a chance to retaliate, rather than having it opened by whoever reached the door first. This made the battle fairly easy, and the gargoyle and minions were dispatched and treasure looted with little fuss. The party then continued to clear out the rest of the dungeon, utilizing their new-found tactical sense to sweep through the remainder of the game board while taking little damage. They even had a healing spell left over at the completion of the quest.

Quest 2: The Rescue of Sir Ragnar

This quest has several notable features, mostly related to its difficulty (or lack thereof). First, the quest enemies consist almost entirely of goblins and orcs, with a single fimir the toughest opponent--nothing like the previous quest, which had a gargoyle, three chaos warriors, and an enhanced mummy, among others. Second, this quest, unlike the previous one, also has traps and secret doors. Well, trap and secret door, but who's counting? Its hallways are also populated by several goblins, so you can't be quite as carefree in exploring the corridors as in the previous quest. However, it's still probably the easiest of the first fourteen quests.

For this quest, since they had actually survived the previous quest, the same heros were played by the same people. They first proceeded to clear out the northeast quadrant of the quest board, using the tactics they had developed in the previous quest. Because the enemies were more easier, however, I rarely even got the chance to attack, since most of the monsters tended to die before my turn came around. As a consequence, neither the wizard nor the elf cast a spell until the very end of the quest, as they were saving them either for a case of necessity or for a stronger opponent (neither of which ever appeared). After clearing the northeast, they headed to the southwest (as the northwest and southeast were solid rock) and found: a 20' by 20' room with a goblin guarding a chest. The goblin was predictably dispatched even before he had a chance to plead for his pitiful life, and the chest was opened. At which point the party discovered that there were, indeed, traps in this quest. Also that single goblins can't accumulate much in the way of treasure. At this point, the party decided to search for traps (in vain, since the only trap in the quest was the one that had just been sprung). Then they opened the last two remaining doors on the board, exterminated the monsters inhabiting those rooms, and then realized that, given that they had as of yet seen no sign of Sir Ragnar, there must be a secret door somewhere that they had missed. Fortunately for them, they started searching from their current room, which just happened to be the location of the lone secret door. After disposing of the final two monsters, they then proceeded to locate Sir Ragnar in room adjoining the one entered from the secret door and then escort him back to the stairs without further incident (given that all the monsters in the quest had already been defeated).

More to come, probably; we finished playing quest 5 this night.