I've like to get outdoors -- the GPS is a fun way to do something geeky while hiking, and often makes a good excuse to go for a hike (to get a specific point). However, just being outside is great.
I like to juggle -- I've been doing it for about 15 years now, and recently have put a lot more time into it, so that I can juggle 5 balls. Juggling is really fun with a partner.
One thing I enjoy which I haven't seen other people mention is knitting & crocheting. It's very algorithmic and you can give things that you make to your friends and they can be duly impressed. There's a lot of room to mess around with patterns and make up your own stuff.
I *used* to be very into radio controlled planes. I don't have time anymore, which is sad. That's really fun.
Playing music and recording it... but that's not terribly original. Anyway, thanks for the question.
I took the Computer Science GRE straight out of undergrad (from Cal Poly San Luis Obispo), and it was fairly hard. However, I did get into a good PhD program (UCSD).
When preparing, the only material I had were the "worst-reviewed book" (and it truly was bad), my friend who I studied with, and my textbooks from classes.
Even though the red book was bad, I still got two things out of it: (1) an appreciation for what the real test would be like (to some degree), and (2) we were forced to evaluate each question for correctness (since there are incorrect answers), which made us study the material more critically, and understand it better. So there is some benefit to using the red book.
If I had to do it over again, I would still use the red book (but with the knowledge that it is poorly written). I would *definitely* find another person (of my same caliber) to study with again. The only thing I would do differently is I would review my classes & old textbooks more, guided by topics found in the red book.
I disagree with the posters who say that you need a master's first to do well on this exam. I didn't ace it, but I did well enough to get into a good program.
Finally, for the GREs (all of them), nothing beats taking practice tests for getting prepared. I took one practice test a week for about 6 weeks, and that helped a lot (for the general GREs).
Hey, I believe this name had been taken (maybe not in a legal sense) for a similar application that I've had on linux for a long time: http://www.ensta.fr/internet/unix/utilities/ical .h tml
I picked up "Seeing in the Dark" (the subject of the book review in the article) from a local bookshop a few months ago on a whim, since I dabble in astronomy, and recently built my first telescope.
I _loved_ this book. Reading it was all I wanted to do until I finished it. I heartily recommend it to people who are like-minded (interested in amateur astronomy). Timothy Ferris is a superb writer. He interviews very interesting astronomers for this book, and visits some great observatories, and has a lot of inspiring stories and interesting facts. I learned a lot.
And that's all I have to say about that. Read it for yourself, and enjoy.
Since naive Bayes gives probabilities, this is easy to get out of what Mozilla (and Paul Graham, and others) are trying to do. However, it is well-known that the probabilities that naive Bayes classifiers give are typically exaggerated (too close to either 0 or 1). This is partly because of the naive assumption (conditional independence of features).
However, while the probabilities themselves may be exaggerated, they are also usually found to be ranked correctly, which would give you what you want here -- a ranked list of possible spams.
This approach is more commonly called "Naive Bayes" classification in the field of machine learning. It is naive because it considers each word to be a feature (dimension), but it also considers each word in an email to be conditionally independent of all other words in the document (which is not true, but really useful in practice).
The author of the web page on using this technique to classify spam (Paul Graham) has a better explanation of Naive Bayes on this web page.
I've written my own naive Bayes classifier to identify spam, with less positive results than he reports. However, naive Bayes can be a very effective technique, and I can believe his results.
The two things you have to beware of when using it are "smoothing" probabilities of words you've never seen (you don't want them to always be zero, as straight naive Bayes will give you), and you need LOTS of training data for naive Bayes to work well. That means that you need to already have a fair amount of spam to identify spam well.
You can see a paper I wrote on using naive Bayes to classify hard drive failures here, or look for more stuff on naive Bayes on Google. Also, don't reinvent the wheel: Andrew McCallum has written a very good toolkit for doing these sorts of things in Bow.
They're not saying that "make xconfig" is new, just that they have changed what it means into a frontend for this new configuration system.
Know where satellites are at all times
on
ISS Flashing Earth
·
· Score: 2, Informative
Heavens Above is a great site if you want to keep up with the ISS and many other satellites orbiting Earth. It explains when and where you can see them in the night sky, and their brightness. It is all based on your location. I have used it to see several satellites, including a very spectacular showing of the ISS last year.
Yes, it was my fault -- bugs are the fault of the programmer!
But thank you for bringing more light on the topic I was raising -- that bugs in release mode may not be replicable in debug mode, which leaves you with a very strange situation.
I disagree that debug builds should be disavowed altogether; using a debugger can still be very useful.
Definitely; I agree (now, having learned my lesson) that putting state-changing statements in code that is also testing the state is a bad idea.
Again, it was stupidity that caused the bug -- my stupidity. However, I thought it was interesting that the release and debug builds produce different behavior, and the problem in release cannot be verified in debug. That was the point of my post.
Yes, yes, it was my own fault for using an assert like this. I should not blame MS for this, you're right, but that was not the point of this post.
THE POINT OF MY POST WAS: this bug presented an interesting problem (difference in debug builds versus release builds), and the fact that sometimes you cannot reproduce bugs in debug mode that occur in release mode.
My favorite bug was allocating memory inside of an assert() using VisualC++ (I hate MS tools; I had to use it for work).
So the gist of the code went something like this:
... 0. int array[]; 1. assert(array = new int[SIZE]); 2. for (int i = 0; i < SIZE; i++) { 3. array[i] = i; ... and the code would segfault on line 3. So I brought it into debug mode, and stepped through. But it worked fine. Back to release mode, and it segfaults.
To restate, here we have the classic example of something you don't want: it works fine in debug mode, but it bombs in release mode.
Of course, since I have simplified the code the answer should be obvious -- in release mode, VisualC++'s compiler was stripping out the assert(), and the allocation inside. In debug mode, it left the assert() in, so the allocation worked fine. I had never changed a flag that said I wanted it to strip them, so I assumed it wouldn't. Never trust M$...
The reason they run on old hardware is because it has been proved reliable, and the legacy software that runs on it has also been proved reliable. When you have an investment like that, you can't just upgrade every time Intel makes a new chip. If you did, you could have major glitches due to oversights in new software/hardware combinations. There's something to be said for a well-developed system, even if it is running on old hardware.
To toot my own horn, I helped them get started with a personal project of mine, which was mapping out the wireless access points at UCSD. See here: http://www.cs.ucsd.edu/~ghamerly/wireless.html
I'm not directly involved with the active campus project, but they started by using some of my data. I got data by collecting it with my gps and laptop as I rode my bike around campus.
No one seems to have commented on the fact that when you are indoors, GPS doesn't work. Perhaps they'd still have something with their pager & 911 dialer, but if you were evil enough you could disable the GPS location on this wristwatch by simply placing the unit indoors, or covering it with something thick. I'm not suggesting this, just playing devil's advocate from a geocacher who has lost gps signals under even heavy tree cover.
First off, I'm doubtful that there are any open-source programs that do this well, as it's a very difficult problem! It has to do with understanding a document, which computers really can't do.
So I'd like to take a moment to point out a good resource for some existing summaries, at bookaminute.
Seems that everyone is ignoring the fact that Linux or Windows is irrelevant if the hardware isn't that fast. The poster said that the system works, but isn't great because the hardware isn't great. Once you get new, better hardware, THAT will be the time to evaluate windows vs. linux. However, I suspect that the linux solution will still be the winner. Until then, the important thing is to get better hardware.
I've been interested in this since I found out about weather underground, which lets you feed your own weather information (gathered from your instruments) to their site for others to look at. They have thousands of people feeding them information. Besides this, they carry National Weather Service and they have a lot of other information that other sites don't seem to have. However, still being a graduate student, I gasped at the prices for a mid-range personal weather station (~$500, from my brief search).
See wunderground's page on personal weather stations here .
I think that when he said "gui", the first poster was referring to "typeset mathematics". Mathematica and Maple have this, but Matlab does not (but matlab does have a maple plugin). Yes, though, the latest version of matlab has a windowed user interface.
Slashdot Mirror
Your moral compass is spinning wildly... it's making me dizzy. What a rationalization.
I've like to get outdoors -- the GPS is a fun way to do something geeky while hiking, and often makes a good excuse to go for a hike (to get a specific point). However, just being outside is great.
I like to juggle -- I've been doing it for about 15 years now, and recently have put a lot more time into it, so that I can juggle 5 balls. Juggling is really fun with a partner.
One thing I enjoy which I haven't seen other people mention is knitting & crocheting. It's very algorithmic and you can give things that you make to your friends and they can be duly impressed. There's a lot of room to mess around with patterns and make up your own stuff.
I *used* to be very into radio controlled planes. I don't have time anymore, which is sad. That's really fun.
Playing music and recording it... but that's not terribly original. Anyway, thanks for the question.
I took the Computer Science GRE straight out of undergrad (from Cal Poly San Luis Obispo), and it was fairly hard. However, I did get into a good PhD program (UCSD).
When preparing, the only material I had were the "worst-reviewed book" (and it truly was bad), my friend who I studied with, and my textbooks from classes.
Even though the red book was bad, I still got two things out of it: (1) an appreciation for what the real test would be like (to some degree), and (2) we were forced to evaluate each question for correctness (since there are incorrect answers), which made us study the material more critically, and understand it better. So there is some benefit to using the red book.
If I had to do it over again, I would still use the red book (but with the knowledge that it is poorly written). I would *definitely* find another person (of my same caliber) to study with again. The only thing I would do differently is I would review my classes & old textbooks more, guided by topics found in the red book.
I disagree with the posters who say that you need a master's first to do well on this exam. I didn't ace it, but I did well enough to get into a good program.
Finally, for the GREs (all of them), nothing beats taking practice tests for getting prepared. I took one practice test a week for about 6 weeks, and that helped a lot (for the general GREs).
Ah hah! Now we know who has been posting as AC for all these years!
Check out the international dark-sky association if you're really interested in doing something about this.
Hey, I believe this name had been taken (maybe not in a legal sense) for a similar application that I've had on linux for a long time:l .h tml
http://www.ensta.fr/internet/unix/utilities/ica
I don't think Apple had "iCal" back in 1995.
I picked up "Seeing in the Dark" (the subject of the book review in the article) from a local bookshop a few months ago on a whim, since I dabble in astronomy, and recently built my first telescope.
I _loved_ this book. Reading it was all I wanted to do until I finished it. I heartily recommend it to people who are like-minded (interested in amateur astronomy). Timothy Ferris is a superb writer. He interviews very interesting astronomers for this book, and visits some great observatories, and has a lot of inspiring stories and interesting facts. I learned a lot.
And that's all I have to say about that. Read it for yourself, and enjoy.
Since naive Bayes gives probabilities, this is easy to get out of what Mozilla (and Paul Graham, and others) are trying to do. However, it is well-known that the probabilities that naive Bayes classifiers give are typically exaggerated (too close to either 0 or 1). This is partly because of the naive assumption (conditional independence of features).
However, while the probabilities themselves may be exaggerated, they are also usually found to be ranked correctly, which would give you what you want here -- a ranked list of possible spams.
This approach is more commonly called "Naive Bayes" classification in the field of machine learning. It is naive because it considers each word to be a feature (dimension), but it also considers each word in an email to be conditionally independent of all other words in the document (which is not true, but really useful in practice).
The author of the web page on using this technique to classify spam (Paul Graham) has a better explanation of Naive Bayes on this web page.
I've written my own naive Bayes classifier to identify spam, with less positive results than he reports. However, naive Bayes can be a very effective technique, and I can believe his results.
The two things you have to beware of when using it are "smoothing" probabilities of words you've never seen (you don't want them to always be zero, as straight naive Bayes will give you), and you need LOTS of training data for naive Bayes to work well. That means that you need to already have a fair amount of spam to identify spam well.
You can see a paper I wrote on using naive Bayes to classify hard drive failures here, or look for more stuff on naive Bayes on Google. Also, don't reinvent the wheel: Andrew McCallum has written a very good toolkit for doing these sorts of things in Bow.
They're not saying that "make xconfig" is new, just that they have changed what it means into a frontend for this new configuration system.
Heavens Above is a great site if you want to keep up with the ISS and many other satellites orbiting Earth. It explains when and where you can see them in the night sky, and their brightness. It is all based on your location. I have used it to see several satellites, including a very spectacular showing of the ISS last year.
Yes, it was my fault -- bugs are the fault of the programmer!
But thank you for bringing more light on the topic I was raising -- that bugs in release mode may not be replicable in debug mode, which leaves you with a very strange situation.
I disagree that debug builds should be disavowed altogether; using a debugger can still be very useful.
Definitely; I agree (now, having learned my lesson) that putting state-changing statements in code that is also testing the state is a bad idea.
Again, it was stupidity that caused the bug -- my stupidity. However, I thought it was interesting that the release and debug builds produce different behavior, and the problem in release cannot be verified in debug. That was the point of my post.
Yes, yes, it was my own fault for using an assert like this. I should not blame MS for this, you're right, but that was not the point of this post.
THE POINT OF MY POST WAS: this bug presented an interesting problem (difference in debug builds versus release builds), and the fact that sometimes you cannot reproduce bugs in debug mode that occur in release mode.
My favorite bug was allocating memory inside of an assert() using VisualC++ (I hate MS tools; I had to use it for work).
...
...
So the gist of the code went something like this:
0. int array[];
1. assert(array = new int[SIZE]);
2. for (int i = 0; i < SIZE; i++) {
3. array[i] = i;
and the code would segfault on line 3. So I brought it into debug mode, and stepped through. But it worked fine. Back to release mode, and it segfaults.
To restate, here we have the classic example of something you don't want: it works fine in debug mode, but it bombs in release mode.
Of course, since I have simplified the code the answer should be obvious -- in release mode, VisualC++'s compiler was stripping out the assert(), and the allocation inside. In debug mode, it left the assert() in, so the allocation worked fine. I had never changed a flag that said I wanted it to strip them, so I assumed it wouldn't. Never trust M$...
The reason they run on old hardware is because it has been proved reliable, and the legacy software that runs on it has also been proved reliable. When you have an investment like that, you can't just upgrade every time Intel makes a new chip. If you did, you could have major glitches due to oversights in new software/hardware combinations. There's something to be said for a well-developed system, even if it is running on old hardware.
This is not just the work of one individual; see http://activecampus.ucsd.edu
To toot my own horn, I helped them get started with a personal project of mine, which was mapping out the wireless access points at UCSD. See here: http://www.cs.ucsd.edu/~ghamerly/wireless.html
I'm not directly involved with the active campus project, but they started by using some of my data. I got data by collecting it with my gps and laptop as I rode my bike around campus.
Here is a link to the yahoo copy of the same reuters story:
n m/ 20020409/en_nm/media_piracy_dc_7
http://story.news.yahoo.com/news?tmpl=story&u=/
No one seems to have commented on the fact that when you are indoors, GPS doesn't work. Perhaps they'd still have something with their pager & 911 dialer, but if you were evil enough you could disable the GPS location on this wristwatch by simply placing the unit indoors, or covering it with something thick. I'm not suggesting this, just playing devil's advocate from a geocacher who has lost gps signals under even heavy tree cover.
First off, I'm doubtful that there are any open-source programs that do this well, as it's a very difficult problem! It has to do with understanding a document, which computers really can't do.
So I'd like to take a moment to point out a good resource for some existing summaries, at bookaminute.
Seems that everyone is ignoring the fact that Linux or Windows is irrelevant if the hardware isn't that fast. The poster said that the system works, but isn't great because the hardware isn't great. Once you get new, better hardware, THAT will be the time to evaluate windows vs. linux. However, I suspect that the linux solution will still be the winner. Until then, the important thing is to get better hardware.
Last year i took my laptop & gps & a few perl scripts and mapped out the wireless access at my campus (UCSD). I made some maps too. Pretty fun!
http://www.cs.ucsd.edu/~ghamerly/wireless.html
See wunderground's page on personal weather stations here .
Well, it appears that the debris is cooling in Egypt. See this story on yahoo.
I think that when he said "gui", the first poster was referring to "typeset mathematics". Mathematica and Maple have this, but Matlab does not (but matlab does have a maple plugin). Yes, though, the latest version of matlab has a windowed user interface.