interstice: (Default)
To say nothing of this outspoken designer, what the hell was this "graduate student" thinking? (And what is she studying that requires so little power of discernment?)

http://ny.metro.us/metro/local/article/Obama_slave_shirt_sparks_lawsuit_threat/13001.html
interstice: (Default)
http://www.youtube.com/watch?v=D6IlGXhCUHo

Well, umm... the question is vague... You don't say what kind of birth control it is. Uhh... One thing is certain! I would not hurt the woman.
interstice: (Default)
Where is the eccentric billionaire to press this in court, when you need one? Certainly Mr. Soros wouldn't mind pitching in?

http://excesscopyright.blogspot.com/2008/07/is-torture-by-music-performance-in.html
interstice: (Default)
I just learned that the "spade" in "call a spade a spade" isn't the racist spade: http://en.wikipedia.org/wiki/Call_a_spade_a_spade

It still doesn't mean you should say it in public though - it's important to be stingy with regard to others' connotations.
interstice: (Default)
Another 5-4 decision: http://www.supremecourtus.gov/opinions/07pdf/07-290.pdf

In short: "The Second Amendment protects an individual right to possess a firearm unconnected with service in a militia, and to use that arm for traditionally lawful purposes, such as self-defense within the home."

---

In similar Amendment-validating news, I'm glad I'm not the only one making a connection between abstract-idea patents and the First:

http://www.aclu.org/freespeech/gen/34784prs20080404.html
interstice: (Default)
Q: You bought a peach & a nectarine at a grocery store. Which do you eat first?

A: The peach - you'll need the nectarine to get rid of its nasty taste. Why the hell did you bother with a peach that wasn't from a greenmarket anyway? Did you really think the experience would differ from the past ten times?
interstice: (Default)
I saw this advertisement the other day: http://animalnewyork.com/news/2008/05/dear-keyel-one-maker-take-your.php

They missed a chance to make Douglas Hofstedter happy: "Dear Ketel One Drinker / One Thousand Eight Words".

Or maybe better: "Dear Ketel One Drinker / One Thousand Seven Eight Nine Eleven Words".

ingenious

Apr. 17th, 2008 12:44 pm
interstice: (Default)
Simply ingenious. This has "Lifetime channel special" written all over it.

http://news.bbc.co.uk/2/hi/americas/7351810.stm

clever...

Mar. 29th, 2008 05:21 pm
interstice: (Default)
This is a very "clever" thing to do to people: http://www.wired.com/politics/security/news/2008/03/epilepsy

Of course I went googling for these images and found a couple. I also found this interesting tool: http://tools.webaccessibile.org/test/check.aspx which failed to detect the danger in the few images I found.

I did not have an attack, so I suppose I'm probably not epileptic. But I also noticed that the image was stuttering and unsync'ed due to my cruddy i810 video non-card. Anyway, it occurred to me that you could make custom video drivers for epileptics to use which wouldn't even detect dangerous flicker per se, but rather just control updating of any region of the screen so as to not fall into the danger zone (i.e. if there are significant oscillatory changes occurring at 4-55 Hz, throttle them down to 2 Hz). It'd be easy with these new killer graphics cards, but also sort of a wasteful way to use the processing from a technical and economic standpoint. I wonder how efficient you can make it.
interstice: (Default)
...were each at one point research assistants for Alan Dershowitz.

Weird, or not?

neatokeen!

Mar. 10th, 2008 04:57 pm
interstice: (Default)
It's such a fantastic feeling when

1) Your simulation doesn't work;
2) you do an integral by hand to surmise why (biased MCMC samples due to a naive lognormal jump kernel);
3) you do 5 more pages of not-completely obvious integrals and approximation
4) (having a happily large number of terms cancel out!);
5) to solve for a reparametrized correction;
6) and change your code to use this method;
(here's where it gets good:)
7) and finally, it actually solves (or at least greatly mitigates - it's not clear whether a proper Solution even exists) the problem!

You know, I don't get nearly enough of step 7) in my life. Sadly step eight remains elusive:

8) Someone authoritative gives a damn about it.
interstice: (Default)
OK, now this is another weird bit from the bbc.

http://news.bbc.co.uk/1/hi/health/7266687.stm

First, how can this "health boost" be unexpected? Since the effect is at the moment purely speculative, it is predicted by a model. That's pretty much the definition of "expected" to me... What would be unexpected is if the model were violated, which will of course be nigh-impossible to evaluate.

But even so, the benefit is a total of 1888 years of life amongst 7M residents since 2003. Uhhh... Crunching this down we get a spectacular benefit of 0.59 hours of life per person, per year of living in the city. Since living in a city in the first place has undoubtedly a much greater negative effect, this is straining at gnats.

The other reasons for the traffic tax are probably reasonable (I support one for nyc for many reasons), but supposing that I were to spend 30 years commuting in London, I think trading off 0.59h*30=about-half-a-day of my life to save about £2000*30*($2/£1) = $120K... you'd be insane not to do it.

I wish that we could come up with consensus measures for "happiness" (which would account for unjust-feeling adversities such as cancer and depression but not some ridiculous mean effect). The traffic tax would certainly help with happiness (hopefully), at least in the big picture. And if it didn't, we could scrap it. Then we wouldn't have to bullshit ourselves about living an expected 16 extra hours.

It reminds me of this, also British; there would be American examples but we don't give even an illusory damn about our wellbeing: http://www.stat.columbia.edu/~cook/movabletype/archives/2005/12/the_ministry_of.html
interstice: (Default)
How did I miss this one? News travels slowly out of the Tir.

http://news.bbc.co.uk/1/hi/northern_ireland/6430683.stm
interstice: (Default)
http://www.rozhlas.cz/d-dur/download_eng

or:

wget -r -l 1 -nH -nd -Pfreemusicdirectory --accept mp3 http://www.rozhlas.cz/d-dur/download_eng

or for .flacs just s/mp3/flac/

teaching

Feb. 25th, 2008 02:05 pm
interstice: (Default)
I've noticed that, at least subjectively, my classes alternate almost day-to-day from better to worse to better and so forth. I teach twice/week, which might be very relevant if it were actually day-to-day alternation (i.e. stationary with period 2). However, it's not. So I have to wonder whether there is any real periodicity at all, or if there is a confounder which I think now is more likely.

My candidates for explanatory variable are [using the computer] and [reviewing material] which are themselves correlated. My reviews often involve some simple 'rithmetic and algebra on the blackboard which serves to focus my attention and also give the students something to fixate on. I'm a little sad about it though, for this reason: When I'm not reviewing these simple procedures, I do put some effort into graphical demos via the laptop, on "real" datasets. However I don't have a very streamlined way to integrate the demos yet, and they aren't as interactive as I'd like. I wish that I knew whether they were doing more harm and good for the class in their current form. I may ditch them anyway.

I think maybe the way to do it is to introduce the simple computation first just to have a concrete foothold for the students and then add the computer example later. Often the computer examples (being "real" datasets and also kinda small) are not as idealized as the textbook examples, thus requiring more familiarity to not get distracted by details. This is at odds to how I would look at them: a reality-grounded first-step into the problem domain (they are that, but only to me and I don't need an introduction anyways - it's incredible how self-negating the process of teaching needs to be).

This may not have a lot of general validity; it's all based on a reaction to today's lecture, which I had fun with and got some good interaction from the students for. We covered regression lines (3rd day or so, of 4) and specifically, after half the class reviewing the concept of a regression line, we computed the associated conditional standard error (given X) using the shortcut form $\sqrt{1-r^2}$. Throughout I used a semi-original example of someone retaking the SAT: they do well the first time (720/800); pay a lot of money for a tutor; and then perform the same the second time (with the population distribution the same too). The moral is that even though he did the same, that's still pretty impressive since he had to match his already strong performance from the first time around - the regression we used predicted he would go down from 720 to 640 whereas he kept it up and got another 720. This was a nice introduction if I say so myself, to computing the SD of the regression line, in order to informally test for significance of his second 720. Note, without the correct conditioning, it is significant at the 0.02-level whereas properly it is only at the 0.12-level. [Note, I used a correlation of about r=0.5 which I suspect is way too low, and was honest with my students about that as well as noting some of the difficulty of computing a correlation across not-entirely-like samples.]

And not a computer or dataset in sight; we only used the summary statistics since we're all comfortable with mean, SD, correlation coëfficient, &c. As a result I was able to talk about the concepts (binormal; conditional normality; hints of homoscedasticity; I even got to relate the regression fallacy to my SAT example) with a bit of fluency. I suspect I may have been doing things backwards before...

markts

Feb. 19th, 2008 04:03 pm
interstice: (Default)
So this company "Laptop Heaven" will give you quotes for laptops instantly after providing model name and a quick summary of damage (no adapter, no battery, broken screen, bad HD, can't power up) and optical drive option.

My thoroughly shattered last-model iBook G4 would go for $100; with a working LCD, it'd be about $200. This was about in line with my expectations. So I decided to look up my old thinkpad 600 (just a bad HD). I expected around $20, but no. They would pay $2 (and shipping), including the adapter! Two bucks?! That was a bit crazy, so I got a quote for a pristine macbook pro: $550. !!! What kind of idiot?!

Anyway, it reminded me of why these specialized and low-information sites end up becoming exclusively junk shops. Basically, so that they don't get systemically fleeced, they have to low-ball everyone by assuming that anything they don't know, is minimized (least RAM; scratches; missing/stuck keys; &c.). Then again since they are low-balling everyone, most people who resort to their service will have smashed systems. The end result is that you implicitly end up with a trade exclusively in total junkers, which is a small and presumably undesirable chunk of the market.

Here is something interesting, though. When I claim my thinkpad has a broken LCD, the value goes UP to $5. This suggests that they're using an automated valuation system, and not experts. But maybe it does make sense in terms of risk management - a lappy with a broken LCD is more of a "known quantity" than a lappy with none of the 5 damages indicated.

Or maybe it's a mistake. It doesn't matter anyway, I'm just curious how they do their pricing. Maybe someone should mine their website and reveng their pricing.
interstice: (Default)
Since I'm sure that everyone cares deeply about prepaid mobile phone plans, I am posting one more time.

If you do talk moderately and thus try to optimize for minutes-per-dollar then for the price of a 450-minute standard (not prepaid) Verizon plan you can get about 400-500 minutes of tracfone time per month (taxes make it hard to get an exact value - tracfone doesn't have any visible ones for now, whereas Verizon tacks them on).

So the perk is that your talk time is effectively prorated. The big downside is of course that you don't get nights and weekends, or within-network minutes. It might be an unintended side effect that tracfone is thus a decent plan for occasional business consultants who are willing to give up expensive "insurance" against a high-volume month. Supposedly you can also call many international landlines for no extra charge.

This is all I will post concerning this topic.
interstice: (Default)
Having recently cancelled my Verizon account, I've been doing some analysis of pay-as-you-go cellular phone services in the US. My friends currently in Europe may begin laughing at this point.

The major carriers (AT&T, Verizon, T-Mobile, Virgin) all seem to have variants on the following strategies. Pay $X per day that you actually use the phone; and an additional $Y per minute following. Furthermore, you have to pay a (often quite high) minimum $ every month. I'll look at Verizon (hereafter: V) just for kicks.

V has the three options:
  • sporadic @ X=$1, Y=$0.10 (plus free minutes for mobile calls to V phones);
  • occasional @ X=$2, Y=$0.05 (plus free nights & weekends in addition to free V minutes);
  • long-distance-relationship @ X=$3, Y=$0.02 (with the same perks as occasional)


On the other hand, there is another "tier" of providers exemplified by Tracfone. Since you can buy Tracfone phones at Target and Radioshack, and you can buy minutes at my bodega (!!) as well as online, I will assume that they are geared toward the poor and/or immigrants used to more liberal cellphone policies. Tracfone sells its service differently. Since they are a bulk reseller of spare bandwidth (minutes?) from the major carriers, this is not unexpected - for one, they don't have any notion of "nights and weekends" for I am guessing this reason.

With tracfone, rather than having a hard minimum per month, each purchase for $Z adds I minutes of talk-time and J days of subscription. If your subscription runs out, you have 60 days to resubscribe to keep your minutes but will likely lose your phone number. I was curious how this worked out in terms of X and Y (like in the other plans). Surprisingly, it is easy to translate if you make a few assumptions - read on! Tracfone has basically the following plans (it gets a little more complicated since they started a new offer which I'll detail later - it does not much change the essential result):

$Z I J
20 60 90
30 120 90
40 200 90
100 400 365
140 800 365

A simple (multiple) linear regression shows that this is consistent with 1) no start-up fee (\beta_0=0); 2) $0.10 per minute; 3) $1 per week of use. Further, the R-squared is >= 0.996 (details available) so the linear regression isn't hiding much. Indeed, the residual error is around or under |$5|/year (!) for each plan. I am surprised by the consistency here, and again it suggests a very conservative strategy on tracfone's part.

Thus, assuming that 1) you are interested in a "sporadic" type plan; but 2) you still use your phone at least once a week, tracfone is better than the alternative. If you didn't think that the "sporadic" plan was for you, I don't think tracfone would be either... (Also assuming that 3) there are no big qualitative differences between the carriers. Apparently, at least in major metropolitan areas, tracfone does just fine. YMMV.)

One problem with this analysis, is that I don't consider running out of minutes. This is where the new offer comes in. If you have a "doubling card", you get double the minutes with any refill, for the life of your current phone. The doubler costs $50, or comes automatically with the $140-plan above. Assuming you have a doubler, you can get minutes as low as $0.125 each if you buy 800 minutes for $100. This will also extend your subscription by a year, which can't hurt. Without a doubler, your best bet for minutes is to get them for $0.20 each with the $40-refill.

One caveat: if you are "roaming", your minutes are used up at a 2:1 rate - since they are a reseller of bandwidth, I don't know what "roaming" means. If you aren't in a MMA beware.

Anyways, it looks like tracfone fits me pretty well. Since I can get a basic phone for $15, I'll try out their service for $35 total. If this works OK, I'll get the $140 plan and see how that does me. That adds up to just about 3 months of Verizon service, and I get a "new" phone. Plus I can buy cell phone-time along with my six of Brooklyn Double Chocolate Stout while waiting for the frazzled guy in front of me to fish out enough coins for a 40. Awesome!
interstice: (Default)
I wound up with some red tea (rooibos) with vanilla the other day. I've been enjoying it in the mornings with Knob Creek in the evenings, when I got an idea. It worked well:

1. Brew a small to medium mug of rooibos. Consider adding a drop of vanilla extract unless it is preflavored.
2. Add at most one teaspoon or so of dark brown sugar. I think a tablespoon is far too much. Bourbon is actually quite sweet and will do the trick almost on its own. However, the optimal amount of sugar is probably inversely proportional to the quality of the bourbon.
3. Add a shot of decent bourbon.
4. Enjoy while very hot (optionally make a toddy by adding a pat of (unsalted) butter - I haven't done this.).

I'm sure there are many variants of this. It also deserves a good name.
Page generated Aug. 19th, 2025 10:05 pm
Powered by Dreamwidth Studios