I think there is some sort of bad voodoo regarding robots just before a competition. The last time I participated in a robotics contest, NASA dropped it off the back of a truck as it was heading off to florida. That sure taught us not to put "fragile! delicate space equipment!" stickers on the box, though.
1. Like another poster said, get em checked out, if you haven't already. There are a lot of diet and drug things you can do about migranes, which won't help for other kinds of headaches. Amazing things can trigger migranes, and avoiding those foods helps me amazingly. (On that note, it's truly amazing how many foods contain MSG.)
2. Caffine fluxuations can cause really impressive migranes, so I've found it better to just avoid the stuff, with one exception. I would really, really reccomend that you don't drink a lot of the stuff, it will screw up your system, and give you a lot of migranes that you didn't have to have.
3. Caffine is also a really good way of treating migranes during the aura stage. Saving it for this purpose is the best way to keep it effective, and will avoid a lot of extra migranes. They suck, you don't need any more of them.
Or instead you could play one of the other games out there which is much, much, much better.
In terms of Lord of the Rings games specifically some very good choices are:
Lord of the Rings (the cooperative game) -- you play across different scenarios to try to throw the ring into mount doom. an expansion lets someone play as sauron and try to kill everyone else, even though the game is pretty good at doing that already (2-5 players in the base game, can be easily adapted for solo play, if you're into that)
Lord of the Rings -- the Confrontation this game is Stratego-like, where one player plays the light side, the other plays dark, and has the appropriate characters and victory conditions. I haven't played this one, but it's supposed to be very good. (2 players)
There are a lot of amazingly good games out there. I was stunned when I found out just how much better games could be than what I was playing. There are funny ones, chaotic ones, deep-thought ones, cooperative ones, ones where you build your own empire, abstract ones, beautiful ones... fun ones!
Seriously, the point of PhD courses is for background and interesting research problems. It makes it much more interesting when people aren't concerned about the grade so much as the material. We are also under a lot of pressure from different sources, and having to put off or skimp on other work when you have a paper deadline is just (a frustrating) part of life.
I did my undergrad at berkeley, myself, so I might be able to address a few of your points.
1. Berkeley has some extremely good theory profs. Just like/every/ other top research school, they don't get tenure for teaching (in some schools profs are punished for teaching well), they get tenure for research. It is natural, though regrettable, that this is the case. Not as many students apply for a PhD program as you might think, especially as admissions are quite competetive, especially after the dot-com bust.
2. The vast majority of CS grads from berkeley are from EECS, not L&S CS. As you said, this implies a BS, not a BA. The requirements in EECS are structured so that you need only take one "project" class, the three lower-div CS classes, the two lower-div EE classes, and some other assorted requirements to graduate. I took basically all theory and math classes, I know others who went differnt routes. Personally, I liked the flexibility. Classmates were very likely to be interested in that particular class and motivated by interest in the subject matter. So no, you don't need to do much hardware at all (in fact, until quite recently, it was LESS than in L&S CS) to graduate with an EECS degree.
3. The professors at any top school are not there to create programmers. They are there to develop your critical thinking skills, ability to clearly communicate ideas, and give you strong background skills which are transferrable to many areas. Finance companies hire quite a few physics majors and CS majors (with a PhD by preference). They have no background in the work that they're asked to do, but they have the critical and mathematical skills required. Math is an extremely useful beast, in all areas of computer science. Code-monkey-ing is something which is going to be extremely easy to pick up with the skills you learn, and hopefully you'll have enough view of the bigger picture so that you can be more useful.
Even as grad students, we're very much pressured to be able to speak well, write well, do brilliant research, and teach adequately. TAing is referred to as an infinite time-suck, which it is, instead of being a priority, which it should be.
But hey, I'm just working on my bitterness quotient. If I'm going to graduate in 5 years, I've got to get it up there!
How well does it save to PP? If slides with equation editor stuff in em save to good.ppt, I'll buy it tomorrow. (Especially since I REALLY do not want to convert my linux server to windows, and there's no powerpoint latex port yet, according to Necula's page)
I'm still conflicted on PP. My advisor requires that I use it (instead of SliTex) for one simple reason: she can "borrow" my slides for her own presentations. However, I find that if you have any real sort of equations, it's going to get kinda scary when you move it from one computer to another.
But this is exactly why people use it for teaching. If you make a good set of slides (and they get changed some every year) you can use them to teach again and again. It saves work over the long run. (use white/blackboard for writing on as well)
One thing I do know for sure is that these people aren't dumb, they just have better things to do than to redo slides all the time.
Lea
Re:Vegetarians and Vegans and Atkins
on
Hackers On Atkins
·
· Score: 1
I'm ovo-lacto as well, and it's surprisingly managable in european countries (even Germany). I'm more worried about going to a conference in China -- sure, I can say "I eat like a Bhuddist", but that doesn't cover MSG, and that'll take me down for longer than meat will...
I look at the Atkins diet and I'm baffled. Granted, I don't like the taste of meat at all, but people who eat meat don't smell good, and people on Atkins smell downright terrible. I suppose it's like smoking -- you just don't notice it.
I'm in the lucky minority that starts feeling ill if I eat too much, too often. I get really sick, I gain a bit of weight, and then I compensate, so perhaps I just don't understand the strength of the motivation to cut out just about everything I eat...
I haven't cycled through a snow-laden winter yet (though I may break down and buy a junker bike so I can do so -- my road bike doesn't deserve that kind of treatment), but a major problem with cycle commuting, especially in hilly areas, is injury. When I get more sick than just having a cold or a minor muscle pull, I'm pretty screwed, bike-wise. I spent nearly 8 months on crutches (non bike-related injury), and the bus system was a lifesaver.
I'm not sure I could make it through that cold, either (I've heard some bad things about clipless in that cold), becasue of the wind chill on big hills. But hey, I'm from CA.
Point being, biking in the winter or summer isn't for everyone. The humidity around here, even, is like sucking soup, and getting a cold, or being at all delicate is going to make it rather impossible. Winter illness can be rather incompatible with strenous biking as well.
I've had good experiences with Volvo, as well. It might depend on the model -- I have one of the last of the Boxy Car ones. I'm only 5'9", ~34" inseam, but my dad is 6'5" and all legs, and he likes it too. YMMV:)
It is also possible that the universe periodically bangs/crunches, or is "created" out of a black hole in a different universe. There are lots of possibilities.
I've struggled with this one for a while, and this is where I am:
1. erasable ballpoint is my eternal enemy 2. bic smears, but not as badly 3. if you get decent ink for your fountain pen, and use the finest nib you can (I'm using a Rotring EF (extrafine)), fountain pens are great. make sure to get a left-handed or straight-ground nib, otherwise you are pretty much out of luck. pretty much everything will be stright, especially if you use tiny nibs. 4. rotring pencils feel very nice, very heavy, well balanced, but mine broke after only a few years. I'm going to try to have it fixed, along with my rotring fountain pen. 5. never let anyone borrow your fountain pen. fine nibs are easier for people to bend in bizzare ways than I would have thought possible.
I'm a PhD-type, and neither do I use a calculator, nor am I a libarts type (except in my Copious Free Time). These things certainly do not form a partition of the disciplines available!
1. Bruce Schneier wrote applied cryptography. It's a compliation of other people's work, in very large part. There are a staggering number of errors in the book, which have led people who don't know any better to implement some rather insecure things. the book is great as an overview book, and as a crypto for non-cryptographers -- IF they double-check that there are no bugs.
He did not invent most of the things you seem to think he did. He in fact does credit the people who did invent them. Naor, Rabin, and Blum, for example, I would call much bigger names in cryptography. So while he has certainly made a valuable contribution, keep it in perspective.
Not all compsci students are sysadmins, or want to be sysadmins. I sure as hell do NOT want to run a network. I think it would make me quite insane.
Then again, I don't run windows of any flavour either, so making a note on my Permanent Record (OH NO!) that I don't know anything about it is just fine with me.
Oh, btw, I wouldn't trust a personal firewall, myself. There are certainly better options.
Well, I actually do crypto, but you don't want to call me in to compare code. Most of the work I've seen in that area has had to do with comparing parse trees; it's not really my area of expertise. This isn't really a pattern-matching problem per se, unless you abstract it out a bit, and it's not clear to me how to do this. Perhaps it is to the compilers or software engineering types, but not to me, so it's likely you'd want those sorts of people to be involved as well. Mathematicians alone are probably not as useful in this situation.
I'm sure that Dwork, Goldberg, and Naor are really happy to know this. Their scheme requires interaction (as do all of them I've seen) and has a quite reasonable complexity assumption.
As far as I know, NO ONE has implemented any of the reasonable schemes that I've seen float around the crypto community. You can, however, find the paper and slides from talks on google:
http://www.google.com/search?q=On+Memory-Bound+F un ctions+for+Fighting+Spam&ie=UTF-8&oe=UTF-8
If you actually do have a way of breaking any of these family of schemes, I'd be very interested to know how. But "get a sample, run some numbers and bam: you have an algorithm" isn't very descriptive. The point that those numbers have special relationships which are believed to be difficult to compute without knowing a special piece of information (called the trapdoor information) may be slipping by you. If you send a response to a query which wasn't given out recently by the server, it's not going to be accepted. If you give out a wrong response, it's not going to be accepted. The probability that one of a reasonable (polynomial) number of queries was given recently is quite small (negligable).
In any case, I'm very interested if you can break any of these schemes, since most of them reduce to useful complexity assumptions, which I'd prefer to avoid if they were false.
google claims you're untenured, which does technically mean this could blow up in your face. there have been a good number of people denied tenure becasue they were not sufficiently "congenial" -- shit-stirring isn't really "go along to get along". obviously depends on your situation and your school, of course.
Lea
Re:some interesting applications
on
Powered by Blood
·
· Score: 1
I'd be more worried about getting a perscription for it and for the syringes, especially somewhere I didn't speak the language...
It depends heavily on where you are. PA has some very weird liquor laws that make wine VERY expensive. Before this I lived in CA, where wine is quite inexpensive and good. I don't actually drink the stuff, techincally, but the difference is obvious even to me.
That no one can compete with the PA liquor store monopoly also means that you have to be lucky to get decent wines, since they don't have much, from what I hear.
Slashdot Mirror
>Care to point me towards a major battle that we lost in Vietnam?
Tet offensive?
Lea
(hi tom)
I think there is some sort of bad voodoo regarding robots just before a competition. The last time I participated in a robotics contest, NASA dropped it off the back of a truck as it was heading off to florida. That sure taught us not to put "fragile! delicate space equipment!" stickers on the box, though.
Lea
This is illegal in many locales (such as PA, but not NY). Check before you do it...
Lea
1. Like another poster said, get em checked out, if you haven't already. There are a lot of diet and drug things you can do about migranes, which won't help for other kinds of headaches. Amazing things can trigger migranes, and avoiding those foods helps me amazingly. (On that note, it's truly amazing how many foods contain MSG.)
2. Caffine fluxuations can cause really impressive migranes, so I've found it better to just avoid the stuff, with one exception. I would really, really reccomend that you don't drink a lot of the stuff, it will screw up your system, and give you a lot of migranes that you didn't have to have.
3. Caffine is also a really good way of treating migranes during the aura stage. Saving it for this purpose is the best way to keep it effective, and will avoid a lot of extra migranes. They suck, you don't need any more of them.
Lea
are you saying that when people /do/ understand you, they want to kill you?
In terms of Lord of the Rings games specifically some very good choices are:
Lord of the Rings (the cooperative game) -- you play across different scenarios to try to throw the ring into mount doom. an expansion lets someone play as sauron and try to kill everyone else, even though the game is pretty good at doing that already (2-5 players in the base game, can be easily adapted for solo play, if you're into that)
Lord of the Rings -- the Confrontation this game is Stratego-like, where one player plays the light side, the other plays dark, and has the appropriate characters and victory conditions. I haven't played this one, but it's supposed to be very good. (2 players)
There are a lot of amazingly good games out there. I was stunned when I found out just how much better games could be than what I was playing. There are funny ones, chaotic ones, deep-thought ones, cooperative ones, ones where you build your own empire, abstract ones, beautiful ones... fun ones!
Lea
It's not a terrible thing, either *wink*
Seriously, the point of PhD courses is for background and interesting research problems. It makes it much more interesting when people aren't concerned about the grade so much as the material. We are also under a lot of pressure from different sources, and having to put off or skimp on other work when you have a paper deadline is just (a frustrating) part of life.
I did my undergrad at berkeley, myself, so I might be able to address a few of your points.
/every/ other top research school, they don't get tenure for teaching (in some schools profs are punished for teaching well), they get tenure for research. It is natural, though regrettable, that this is the case. Not as many students apply for a PhD program as you might think, especially as admissions are quite competetive, especially after the dot-com bust.
1. Berkeley has some extremely good theory profs. Just like
2. The vast majority of CS grads from berkeley are from EECS, not L&S CS. As you said, this implies a BS, not a BA. The requirements in EECS are structured so that you need only take one "project" class, the three lower-div CS classes, the two lower-div EE classes, and some other assorted requirements to graduate. I took basically all theory and math classes, I know others who went differnt routes. Personally, I liked the flexibility. Classmates were very likely to be interested in that particular class and motivated by interest in the subject matter. So no, you don't need to do much hardware at all (in fact, until quite recently, it was LESS than in L&S CS) to graduate with an EECS degree.
3. The professors at any top school are not there to create programmers. They are there to develop your critical thinking skills, ability to clearly communicate ideas, and give you strong background skills which are transferrable to many areas. Finance companies hire quite a few physics majors and CS majors (with a PhD by preference). They have no background in the work that they're asked to do, but they have the critical and mathematical skills required. Math is an extremely useful beast, in all areas of computer science. Code-monkey-ing is something which is going to be extremely easy to pick up with the skills you learn, and hopefully you'll have enough view of the bigger picture so that you can be more useful.
Lea
Ah, someone who makes sense. :)
Even as grad students, we're very much pressured to be able to speak well, write well, do brilliant research, and teach adequately. TAing is referred to as an infinite time-suck, which it is, instead of being a priority, which it should be.
But hey, I'm just working on my bitterness quotient. If I'm going to graduate in 5 years, I've got to get it up there!
Lea
I've just used SliTex, which has been more than good enough for me. However, my advisor can't grab my slides, which makes this a no-go.
Lea
How well does it save to PP? If slides with equation editor stuff in em save to good .ppt, I'll buy it tomorrow. (Especially since I REALLY do not want to convert my linux server to windows, and there's no powerpoint latex port yet, according to Necula's page)
Lea
I'm still conflicted on PP. My advisor requires that I use it (instead of SliTex) for one simple reason: she can "borrow" my slides for her own presentations. However, I find that if you have any real sort of equations, it's going to get kinda scary when you move it from one computer to another.
But this is exactly why people use it for teaching. If you make a good set of slides (and they get changed some every year) you can use them to teach again and again. It saves work over the long run. (use white/blackboard for writing on as well)
One thing I do know for sure is that these people aren't dumb, they just have better things to do than to redo slides all the time.
Lea
I'm ovo-lacto as well, and it's surprisingly managable in european countries (even Germany). I'm more worried about going to a conference in China -- sure, I can say "I eat like a Bhuddist", but that doesn't cover MSG, and that'll take me down for longer than meat will...
I look at the Atkins diet and I'm baffled. Granted, I don't like the taste of meat at all, but people who eat meat don't smell good, and people on Atkins smell downright terrible. I suppose it's like smoking -- you just don't notice it.
I'm in the lucky minority that starts feeling ill if I eat too much, too often. I get really sick, I gain a bit of weight, and then I compensate, so perhaps I just don't understand the strength of the motivation to cut out just about everything I eat...
Lea
I haven't cycled through a snow-laden winter yet (though I may break down and buy a junker bike so I can do so -- my road bike doesn't deserve that kind of treatment), but a major problem with cycle commuting, especially in hilly areas, is injury. When I get more sick than just having a cold or a minor muscle pull, I'm pretty screwed, bike-wise. I spent nearly 8 months on crutches (non bike-related injury), and the bus system was a lifesaver.
I'm not sure I could make it through that cold, either (I've heard some bad things about clipless in that cold), becasue of the wind chill on big hills. But hey, I'm from CA.
Point being, biking in the winter or summer isn't for everyone. The humidity around here, even, is like sucking soup, and getting a cold, or being at all delicate is going to make it rather impossible. Winter illness can be rather incompatible with strenous biking as well.
Lea
I've had good experiences with Volvo, as well. It might depend on the model -- I have one of the last of the Boxy Car ones. I'm only 5'9", ~34" inseam, but my dad is 6'5" and all legs, and he likes it too. YMMV :)
Lea
It is also possible that the universe periodically bangs/crunches, or is "created" out of a black hole in a different universe. There are lots of possibilities.
Lea
I've struggled with this one for a while, and this is where I am:
1. erasable ballpoint is my eternal enemy
2. bic smears, but not as badly
3. if you get decent ink for your fountain pen, and use the finest nib you can (I'm using a Rotring EF (extrafine)), fountain pens are great. make sure to get a left-handed or straight-ground nib, otherwise you are pretty much out of luck. pretty much everything will be stright, especially if you use tiny nibs.
4. rotring pencils feel very nice, very heavy, well balanced, but mine broke after only a few years. I'm going to try to have it fixed, along with my rotring fountain pen.
5. never let anyone borrow your fountain pen. fine nibs are easier for people to bend in bizzare ways than I would have thought possible.
Lea
I'm a PhD-type, and neither do I use a calculator, nor am I a libarts type (except in my Copious Free Time). These things certainly do not form a partition of the disciplines available!
Lea
(see, I talk like an academic, at least!)
1. Bruce Schneier wrote applied cryptography. It's a compliation of other people's work, in very large part. There are a staggering number of errors in the book, which have led people who don't know any better to implement some rather insecure things. the book is great as an overview book, and as a crypto for non-cryptographers -- IF they double-check that there are no bugs.
He did not invent most of the things you seem to think he did. He in fact does credit the people who did invent them. Naor, Rabin, and Blum, for example, I would call much bigger names in cryptography. So while he has certainly made a valuable contribution, keep it in perspective.
2. "a lot" is two words, not one
Lea
Not all compsci students are sysadmins, or want to be sysadmins. I sure as hell do NOT want to run a network. I think it would make me quite insane.
Then again, I don't run windows of any flavour either, so making a note on my Permanent Record (OH NO!) that I don't know anything about it is just fine with me.
Oh, btw, I wouldn't trust a personal firewall, myself. There are certainly better options.
Lea
Well, I actually do crypto, but you don't want to call me in to compare code. Most of the work I've seen in that area has had to do with comparing parse trees; it's not really my area of expertise. This isn't really a pattern-matching problem per se, unless you abstract it out a bit, and it's not clear to me how to do this. Perhaps it is to the compilers or software engineering types, but not to me, so it's likely you'd want those sorts of people to be involved as well. Mathematicians alone are probably not as useful in this situation.
Lea
I'm sure that Dwork, Goldberg, and Naor are really happy to know this. Their scheme requires interaction (as do all of them I've seen) and has a quite reasonable complexity assumption.
F un ctions+for+Fighting+Spam&ie=UTF-8&oe=UTF-8
As far as I know, NO ONE has implemented any of the reasonable schemes that I've seen float around the crypto community. You can, however, find the paper and slides from talks on google:
http://www.google.com/search?q=On+Memory-Bound+
If you actually do have a way of breaking any of these family of schemes, I'd be very interested to know how. But "get a sample, run some numbers and bam: you have an algorithm" isn't very descriptive. The point that those numbers have special relationships which are believed to be difficult to compute without knowing a special piece of information (called the trapdoor information) may be slipping by you. If you send a response to a query which wasn't given out recently by the server, it's not going to be accepted. If you give out a wrong response, it's not going to be accepted. The probability that one of a reasonable (polynomial) number of queries was given recently is quite small (negligable).
In any case, I'm very interested if you can break any of these schemes, since most of them reduce to useful complexity assumptions, which I'd prefer to avoid if they were false.
Lea
google claims you're untenured, which does technically mean this could blow up in your face. there have been a good number of people denied tenure becasue they were not sufficiently "congenial" -- shit-stirring isn't really "go along to get along". obviously depends on your situation and your school, of course.
Lea
I'd be more worried about getting a perscription for it and for the syringes, especially somewhere I didn't speak the language...
Lea
It depends heavily on where you are. PA has some very weird liquor laws that make wine VERY expensive. Before this I lived in CA, where wine is quite inexpensive and good. I don't actually drink the stuff, techincally, but the difference is obvious even to me.
That no one can compete with the PA liquor store monopoly also means that you have to be lucky to get decent wines, since they don't have much, from what I hear.
Lea