Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
96 lines (55 sloc) 12.4 KB
> "I will repeat this point again until I get hoarse: a mistake is not something to be determined after the fact, but in the light of the information until that point."^[[Nicholas Nassim Taleb](!Wikipedia)]
It's a good idea to [analyze our successes]( to avoid getting big heads and forget how fortunate we were, but then again, it's even more important to analyze your failures. If you aren't doing the latter, it's pointless to think about the former. I've already analyzed my *intellectual* mistakes in [Mistakes](), so, it's worth taking time to think about where I have screwed up in *life*.
This is about my own mistakes. Arguably, I have been more severely damaged by my hearing impairment than by any mistake discussed here, but since that was congenital and there is nothing whatsoever I could have done differently to prevent it, and apparently nothing I can do to restore my hearing (I once asked my mother whether cochlear implants would be better for me; she said that they had considered them, but there was a good chance they would not be better, and would destroy my existing hearing, which was fairly substantial as these things go) short of spending many thousands of dollars for very high-end hearing aids, and it's unclear whether there is any way to make up for some neurological audio processing deficits. (I *could* probably repair my near-sightedness with [LASIK](!Wikipedia) or something similar, but glasses are only an annoyance, not a genuine problem like being hard of hearing.)
> "The keenest sorrow is to recognize ourselves as the sole cause of all our adversities."^[More literally translated as, "The greatest griefs are those we cause ourselves." Line 1184, _[Oedipus Rex](!Wikipedia)_, Sophocles]
When I graduated high school, computer science was the obvious thing to do. I am almost stereotypically nerdy, and had adapted well to using Linux and learning the CLI & shell scripting, so I had dipped my toe in the waters and not gotten it bit off. I had no real reason to pick CS over the other majors RIT offered like IT and software engineering (SE), so I went with Paul Graham's [advice]( to start with the hardest major you are interested in. (Math was *too* hard to major in, I wasn't interested in statistics, and CS seemed - and other people generally agreed - harder than IT or SE.)
I was accepted, IIRC, into every school I applied to; but I hadn't applied to very many and hadn't bothered with the really impressive CS schools like Stanford or MIT. So my choice boiled down to [SUNY Geneseo](!Wikipedia) & [RIT](!Wikipedia). The costs were roughly comparable - Geneseo didn't offer me much, but as a SUNY school they were cheap; RIT on the other hand is a very expensive private school but I am hard of hearing and so was accepted into [NTID](!Wikipedia) which came with considerable financial aid. (I've never been very aware of the deaf community or whether my life is above or below average given my hearing impairment. I have the vague impression that I'm considerably above average, since [Gallaudet University](!Wikipedia) paid for me to fly down to Washington DC, and at the one deaf conference I went to in Boston in high school, a lady who was a recruiter or something mentioned that my SAT scores & grades were very high and she was particularly impressed by my 800 Verbal given the hearing impairment; as well, I was told that very few NTID students were enrolled for technical majors and I would be one of the only deaf or hearing impaired CS majors.)
I was uncertain which to go to. Both had good reputations, good extracurriculars, price was similar, locations were similar, etc. Ultimately, my parents said they preferred RIT because NTID provided many services for the deaf or hard of hearing, and SUNY Geneseo didn't. So I went along.
What then happened? I discovered that RIT's CS program was heavily focused on Java, and then I discovered that I hated Java, and was positively nauseated. I had been learning Emacs and found Emacs & Common Lisp much more interesting, and then I began learning Haskell, which I liked even more than the Lisps. I was so repulsed that I basically stopped CS classes and switched to philosophy & psychology & required courses.
At this point, my choice for RIT bit me. RIT did not have a philosophy major and I exhausted the available classes by the end of year 2. If I had been at Geneseo, I could have done a philosophy major and that would keep me sane while I did a CS major as well. But all RIT had was a philosophy minor. So I decided to transfer. I couldn't get into anywhere else on such short notice, so I went back home, enrolled in Suffolk Community College, and waited until I could get into SUNY Stony Brook, where I got my philosophy degree.
In retrospect, I can identify so many mistakes in this short history.
1. I knew in high school I might want to do both CS & philosophy, but didn't appreciate that I was forfeiting options by going to RIT, and that forfeiting options is expensive
2. I didn't buckle down and just suffer RIT's Java material in the hope that later classes would be better
3. I didn't seriously look into transferring to Geneseo
4. I didn't seriously look into taking classes at Geneseo
5. I didn't take the design-your-own-major option
# Germany internship
In 2010, in a spasm of akrasia fighting, I signed up for a federal jobs database for the disabled. After a long interview, the only offer I got was a sort of internship in a library on a US military base in Germany. After pondering it, I decided to turn it down, emailing them the following analysis:
## Rejection email
> It has been a week now, with no reply from any of you three. It's possible that my emails went into a spam folder, but more likely that y'all are overworked or I am not a priority or the slots have been filled. Any of those three render this less attractive. after thinking about it, unless something has changed, I think I will turn down this offer. (Sorry.)
> Going abroad at the last minute, my friends advise me, is a dicey proposition to begin with, and if one doesn't have support, likely to end poorly. But the money issues were very real. I always hate it when someone makes a decision and justifies it only vaguely with 'money', so I'm 'showing my work', if you care.
### Money
> I tried looking for airfare. I searched [Travelocity](!Wikipedia) JFK->Frankfurt, round trip, August 15 to October 1st (+- a day), and the *cheapest* round-trip is a cool 1k: <>
#### Germany
##### Costs
> Figure that there are miscellaneous fees and whatnot, and the real price will be \$1100. Room & utilities are $150/month, and August 15 is pretty much exactly half way through August, so figure that will be 1 and 1 half months: $1.5 \times 150 = 225$. So room, utilities, and travel are \$1325. Board is not mentioned. I don't know what that will be. Judging from college, I think I can guess that lunch & dinner would run me about \$15/day. The 7 weeks cover 48 days (minus the day I fly back), so $7 \times 48 = 720$. $1325 + 720 = 2045$ in expenses. Traveling around and entertainment naturally cost more. How much? I have no idea. Let's leave it at a cool 2k in fixed, obvious, foreseeable expenses. Murphy's law guarantees the actual bill will be worse.
> Not to mention, I would have to actually work. As compared to the nightwatch where I have 8 hours to study as I please. If I valued the 8 hours of non-working time at the pool at $5 a piece and I value the hours at the library at $0 an hour (as opposed to actively disliking them and regarding them as a loss of -$5 or something), that's a loss of $280 \times 5 = 1400$.
##### Benefits
> If I got there August 15, Sunday, and the week starts then (for a full week), working from August 15 to September 30, that is 7 full weeks. $5 \text{days} \times 7 \text{weeks} \times 8 \text{hours}$, or $40 \text{hours} \times 7 \text{weeks} = 280$. The quoted salary is 11.75/hr, but I expect that to be lower, and there are taxes. (I'm a civilian, so I don't get any military tax breaks.) So call it \$11/hr. $11 \times 280 = 3080$.
##### Net
> $3080 - 2045 = 1035$ profit
#### United States
##### Costs
> I refuel my car every 2 or 3 weeks. Pessimistically, I will refuel 4 times during the period in question, at \$20 per refuel, or \$80. I don't foresee any other expenses - I'm not traveling anywhere except NYC, room/board/utilities don't apply, and so on.
##### Benefits
> Looking at my most recent check from the town, my biweekly payment was 621.64, round to 622. $\frac{622}{2} = 311$ per week. It's 7 weeks, so $7 \times 311 = 2177$.
##### Net
> $2177 - 80 = 2097$
#### Conclusion
> So. Assuming I can get to Germany cheap, assuming there are no significant expenses there, and so on, then at a minimum, taking this internship will cost me $1000.
> Valuing leisure time as above, the cost will be $3500.
## Why was this a mistake?
There are at least 3 reasons this was a mistake:
1. I did not work 7 weeks for the town. In fact, I worked until September 12, not September 30. My salary expectation was off by 3 weeks, or $3 \ times 622 = 1866$. So I would actually have made more in salary had I gone.
2. I overestimated the value of my leisure time. Examining what I do in practice, my leisure time during nightwatch is better estimated at $2-3 (nightwatch wreaks havoc on one's energy & intellectual ability), so I should have estimated the cost of the internship at around $700 less.
3. I underestimated the value of time overseas.
In retrospect, it's clear to me that I did so largely out of fear of spending time & getting lost in a place I didn't speak the language; I have procrastinated severely on trips in to NYC, which objectively are far less risky & more straightforward than trips in Germany, so it's unsurprising that my judgement would be powerfully warped by a even more extreme trip.
# Quantified Health Prize
On 2 December 2011, the [Quantified Health Prize]( was announced; the challenge was to write up research on the mental benefits of dietary supplementation of various minerals, with the prizes being \$5000 for first place and $500 for 2-5^th^ place. It was [immediately assumed]( that I'd enter due to my well-known interest in [nootropics](Nootropics) and since I had won [an earlier contest](, on [spaced repetition](Spaced repetition) (for \$385). After thinking about it and looking at the list of minerals on Wikipedia and poking around a little, I [declined to](, saying there was little evidence of big effects from minerals.
Somewhat to my surprise, there didn't seem to be too many entries to the contest. I did not think my entry would have beaten [Kevin's team](, which he described as "...a great entry, that took me around 100 hours of work + 20 hours from a partner. It's over 9000 words." But there were 4 other prizes than first place. Given my straitened financial circumstances, even if I had spent all my time in December working on it, the \$500 and result would have been worthwhile.
Why didn't I enter? [*Really*?]( I could use the money. Even at the time, I explicitly knew the aforegoing reasoning ("...I suppose I could package up the research on those 3 and hope for one of the smaller \$500 prizes if it looks like, as last time, there will be few participants.") I think the reason is that I really do think minerals will do little, and hence it is *immoral* to try to profit off research on it. ('Impolite' might be a better word.) The [results]( bore out my expectations: there were only 5 entries, the top prize had sections similar to work I had *already* done, Kevin's entry turned out to be an abdication & polemic, and the other 3 didn't impress me very much (one focusing on the paleo diet!). Not participating was a clear mistake.
This is foolish. They felt it was worthwhile and it's their money, they set up the contest. It would be a lot of work for me, so it is not as if I am scamming them. This is an ethical hangup I must overcome. Refusing to participate was a mistake.