Well, as of 9:30 AM central time 24.224.174.18 isn't accepting connections, so it's either been slashdotted or taken down.
My web logs show some attempts yesterday, but with the wget going to 24.224.174.18. I haven't read up on this worm, but I'm guessing that's an address of someone upstream who was infected. E.g. if my machine infected a new one, the wget for those new infectees would point to my machine.
Yes, I'm aware of the applications of linear algebra to circuits and structural analysis. I also do a simplified version of google page-rank using eigenvalues in my class, and adjacency matrices, error-correcting codes, applications to biological models, economic models, traffic flow through networks, and Markov chains. But I'm still curious what kind of job murr has.
On the other hand, I wouldn't be able to do my current job without Linear Algebra, and there are many days I wish I'd have paid more attention in all my math classes.
As someone who teaches Linear Algebra, I'm very curious -- what kind of job do you have? (I figure it may be another motivator for my students.) And don't say you are a Linear Algebra teacher.;-)
Back in the old days, read early 90's, we had keyboards with a decent sized return key. I miss them.
If by "decent sized return key" you mean the double-height return keys, I always make sure the keyboards I buy do NOT have those giant return keys. I want a single-height return key, with an easy-to-reach backslash key right above it. The giant-sized return-key keyboards usually move the backslash key way down into the corner, which is a royal pain for those of us who write LaTeX code which needs a million backslashes.
...I have been practically keyboard-free for a couple of years now (on both Linux and Windows). I use one for typing in my login password, and for text-mode console emergencies, but that's about it.
It's quite possible to be fully productive without a keyboard.
That's funny, I do my very best to avoid using my mouse (and I don't have other input devices), so basically I try as hard as possible to do everything via keyboard, since it's much faster than mouse for me. Although I often wish I had a couple of foot pedals which I could program as control keys or page up / page down or other things (does anyone make such a thing?).
What other input devices do you use? (And by OOS, do you mean Occupational Overuse Syndrome, i.e. Repetitive Stress Syndrome?)
After getting my bachelor's degree in CS, I worked for an interesting computer company for almost 2 years. But I simply didn't enjoy the job. Unlike what others are saying here about insane overtime, the hours usually weren't bad at all. My evenings and weekends were usually my own, but the days at work just weren't all that interesting.
I eventually went back to grad school and got a PhD, and am now on the tenure track. It's totally the opposite. Now I'm always insanely busy, evenings and weekends I'm just trying to keep up with teaching and research, and there's always something new coming up to deal with. But I absolutely love the job. Even though there's plenty of mundane stuff to deal with, in a very fundamental way I am mostly free to work on what I want to work on. I'm earning less money than I could, but I can't imagine switching to any other job. Also, even though I'm always busy, in some ways my schedule is way more flexible. Lots of times, if I feel I need to take an hour off in the afternoon, I can go ahead and do it, unless I have any meetings or office hours or class at that time; I don't need to check with anyone or tell anyone. Just as long as I do a good job with my teaching and research, I dictate my schedule, within some constraints.
Also, my research often consists of me just thinking/working on my own, in my own little world. But then I get my recommended daily allowance of social contact by teaching (I do really enjoy interacting with the students), and also by talking about my research with my graduate advisees or other faculty or students, etc.
When I switched from industry/CS to academia/Applied Math, I wasn't sure I'd succeed. I'm still not sure (I haven't reached tenure yet), but so far I don't at all regret the decision to try to do what I really want to do.
Right now I'm in the final throes of preparing a grant proposal, so just coming out of one of the high-stress periods. Now I move on to work on some papers, while waiting the 6+ months to hear back on the proposal...
There are some nasty academic politics to deal with, and other crap too (e.g. plagiarizing students, I hate it when they do that). Every job has its ups and downs. You just need to figure out which one has the downs you can tolerate, and the ups which make it all worthwhile. I'm lucky, I managed to find one for me. And fortunately, we're all different, otherwise we'd all be fighting over the same job!
No, I don't know of any job openings at the moment. I'm a (relatively new) faculty, and if I knew of some job openings, I'd probably hoard the info for my students.:-) I'll tell you what I tell my students, though -- any chance to get involved in any kind of project, for pay or not, is really important. I did tons of projects in my spare time as an undergrad and grad; some were research assistant jobs (even as an undergrad), some were just my own things, but done well enough that I could show them to other people. Summer jobs on some kind of research project really help. My second year as an undergrad, I started knocking on prof's doors until I found some willing to give me some work, which then led to more and more work. And so began the long path to my current job, which will be permanent if I can just make it through tenure.
I did a good dose of measure theory in grad school, and found it very interesting, but haven't really used it since then. And I've taken plenty of applied stats and mathematical statistics, but again I forget most stuff I don't use. Although I do teach elementary stats now and then, and a course in deterministic and stochastic modeling and simulation which involves a lot of Poisson processes.
Hmm, it's true that the exponential distribution has its mode at 0, so in some sense you're "most likely to be infected the moment you connect". But e.g. for an exponential with a mean of 17.3 like we were talking about, you have a 25% chance of first being infected in the first 5 minutes, but still a 19% chance of first being infected within the second 5 minutes, and a 14% chance in the third 5 minutes. So it's not all bunched up at 0 as much as you may imagine.
I think Poisson processes are pretty cool. I like putting them in my modeling class because I can use Poisson processes to tie together the following probability distributions and show relations between them all: continuous uniform, binomial, normal, exponential, and Poisson. Even the students that learned about them in basic probability/stats never realized they were all linked together. Sheldon Ross' book on "Intro to Probability Models" (up to about 9th edition or so now) is a pretty readable book which talks about them quite a bit. They're used to model e.g. failing parts in complex machines, incoming phone calls on a busy phone line, automobile traffic, etc.
Hey, if there's one thing you can count on on/., it's the spelling/grammar nazis and the math police.
I generally don't bother responding to basic math errors. But this one seemed more interesting. And a guy bragging about being a brilliant mathematician in his sig is just asking for extra scrutiny.:-)
But seriously, being able to totally geek out without limit in a discussion is what this place is all about, right? And after the other response, if you look at my other long reply, I think it's actually not so obvious -- the mean and median really are different. Although it's not over yet, I admit I may still be proved wrong.
I saw your sig, that's why I figured you should know better.:-) I am a mathematician (with a job), but generally don't flaunt it.
First, the AC didn't say "large number of samples", he/she said "large number" and wasn't very clear about the exact meaning. Yes of course if you compute a sample mean from a large sample, its sampling distribution converges to a normal distribution. You were also not so clear, because when you said "I was thinking of continuous, you were thinking of discrete," you made it sound like those two things are opposites.
Second, of course this distribution can't be truly normal, because it's truncated at 0 on the left. Although I guess you already know, if you want to talk about sample means of large samples, you can generally ignore that since the variance becomes small enough that the probability in the truncated tail is negligible.
Finally, none of the above really matters anyway, as the proper distribution for the time until infection would be an exponential distribution in this case, since there are a very large number of infected machines out there on the network, each with a very small chance of infecting any given target within a reasonably small time interval, and so this system should be fit extremely well by a Poisson process. An exponential distribution with a mean of 12 has a median of about 8.3. Or, if the first line of the article is written correctly and there really is a 50% chance of getting infected within 12 minutes, it means that the median is 12, in which case the mean time to infection is about 17.3 minutes. And these are theoretical means and medians, which are independent of sample size.
And finally finally, I will point out that, given the above information, if you took a really large sample of PCs and measured the sample mean time to infection in that group, the sampling distribution of mean would be normally distributed around 17.3, not around 12.
Oh, [Code Red] is still around. LogWatch complains to me nightly of many 404 Not Founds in my Apache access logs for files that Code Red [II] tries to access on IIS web servers.
See, now that's pretty impressive. I wonder when will the last "active/live" instance of Code Red die? Will it still be around after 10 years, in the year 2011? Eventually there won't be enough susceptible machines around to sustain it, when enough people have moved on from the versions of Windows which are vulnerable. But that could be a very long time. Some people still put Windows 98 machines on the Internet.
The fact that you still see it so often means it's not just a handful of machines out there infected by it too; there must be quite a lot in order for you to be seeing them hitting your machine practically every day. (Or, just a bunch "near you" in IP address space, since Code Red II had a kind of mixed movement strategy which include a good chunk of localized scanning.)
To you and the other person who both said "RedHat 9 is obsolete, use something newer and you'll be safe" -- newer isn't always necessary if you secure your machine well.
My firewall at home is an old Pentium with 32MB of memory, running RedHat 6.1. Last time I rebooted it (to plug it into its new UPS) was Sept 9, 2003; it's been running smoothly since then. No one has ever broken into it. Before that I used an older PC for at least 3 years, again with no break-ins.
I don't see any need to update that machine from RedHat 6. All it does is protect my other machines, and it still does that quite well. I use Adelphia cable for internet access, so there are plenty of break-in attempts (although granted, most of the attempts are looking for Windows vulnerabilities).
Anyone can take an unpatched Windows host and put it on the network with no firewall and say "Look! It got owned in X minutes!"
I totally understand the point you're making, but I believe that the information they conveyed (paraphrased above) is still very interesting. It shows that there is still a very large number of infected machines out there, trying to infect new machines. The USA Today story from last November mentioned that within 12 minutes, an unpatched machine was getting infected via vulnerabilities which had been exploited by worms 6 months earlier.
Usually we only hear about the "latest-and-greatest" new worms coming out. Stories like this shows that the old ones aren't gone; there are still vast populations of machines out there infected by very old worms, and as a result, there is an ever-present background noise of things trying to infect new machines.
E.g. consider this -- do you think an unpatched machine put on the net right now could get infected by Code Red II? That came out just about 4 years ago. I honestly don't know the answer to this question, but I'd really like to. Is Code Red II still "alive" out there? Or has it completely died off yet?
But I had continuous distributions in mind when I wrote that, and apparently you were thinking of finite distributions.
Now you seem to be confusing "finite" with "discrete" by saying it's one versus the other. A distribution can be discrete but (countably) infinite, e.g. the Poisson distribution.
Although what this has to do with mean vs median, I don't know. The mean and median are defined for both discrete and continuous distributions, so the fact that the other guy said median instead of mean doesn't mean he was thinking of a discrete distribution.
The mouse pad cannot be operated on any metal surface. Since the warnings about this are printed on the box, in the manual, and on the mouse pad itself, I figured it was in my best interest to avoid finding out why metal and power-over-RFID don't mesh.
Well there's a sentence that sure didn't end the way I wanted it to... Where's his sense of adventure?
There are some blogs in China, I found a bunch of them I think last year, although I seem to have lost my links to them. I'm not a big blog-reader anyway.
I think people in China are more free to talk about many topics than many Americans think, although you are right, if you talk about the wrong thing, you can end up in deep trouble. But I get the impression that many people in the US think if you even whisper something about governments, the police will come breaking in the door and take you away. I've had many conversations with many people in various places about "forbidden" topics. But I also wouldn't dare publish my ideas on those topics publicly in China.
I'm actually about to head back to China for 2 months; guess I'll soon see what's hot there these days.
I think it was about 2 or 3 years ago, there was an incredibly popular blog in China, written by a woman, about her sexual experiences. (And from what I read, yes, it really was written by a woman.) I'm blanking on her name, and can't even remember if she was in Hong Kong, or maybe Guangzhou. Anyway, it was incredibly popular, among people in China, and overseas. I think that put the idea of blogging on the map, in people's minds in China. But it seems to have faded from the public consciousness since then.
I'm sure someone here knows what I'm talking about and can post a link to more info.
Agreed. I really hate it when I receive a conference announcement, and after reading it, have absolutely no idea what the conference is about. In those cases, I'd think the people publicizing the conference would want people to know just what is being advertised. This is a little different, being a story about the conference rather than an announcement trying to attract people to it, but it's still the same problem...
Today I just submitted a research grant proposal to the US Department of Energy, in their "Multiscale Mathematics in Research and Education" program. If you look at their proposal guidelines, near the end under "The evaluation under item 2", they talk about making materials available to the public as open source. I was happy to see it, as I have made some software available from my previous research available in this way, and plan to do so again in the future on my new projects. I'll have to wait many months to see if they fund my current project proposal, though...
A dairy farmer could probably right a book on cows, cow technology, cow behaviour, and how all that relates to his philosophy of life and why his kid is studying rocket science at university.
First, I think you mean "write", not "right".
Second, I have a book I inherited from my great-grandfather (a farmer). The book was published in 1944, and is called "Cowphilosophy". No joke. The subtitle is "The Art of Practical Dairy Practice". Inside the front cover, before the title page, is a page with some pictures of cows, and the text:
We Are Your Cows We have to eat what you provide. Drink what you give us. Live where you put us. We may be good cows or we may not. We may be healthy, or we may not. We may be comfortable or we may not. We may be profitable, or we may not --- So much depends on you, the dairyman. THE FIRST REQUISITE OF A PROFITABLE DAIRY BUSINESS IS A GOOD DAIRYMAN!
(Strange but true. I love having that book on my shelf. I didn't grow up to be a farmer, though. I ended up getting a PhD in applied math. But I can always consult my Cowphilosophy book when I need some real wisdom.)
I'm a third-year assistant prof in applied mathematics now, just for reference.
I've reviewed papers relevant to my current research, but nothing I felt I really needed to cite in my own papers. But there really isn't a way to cite a paper you anonymously reviewed, without breaking anonymity (unless you simply wait for it to be published). That essentially happened to me once. A guy reviewed my paper; he said he liked it, but didn't think it belonged in the journal I'd submitted it to, which caused it to be rejected by that journal. I happened to give a talk in his dept around that time, and he told me he was reviewing my paper, and what his review would say. He then proceeded to cite my paper a few times in his own papers over the next couple of years, citing it as "A. Name, unpublished manuscript" or something along those lines. It was annoying, since it would have been published if not for his review.:-) So it was good enough for him to cite, but not for him to recommend that that journal publish it. I finally did get it published elsewhere, but it took some time.
So you could identify yourself to the author of the paper, i.e. sign your review, which is something I think you're always allowed to do if you so choose. Except in your case if the paper was sent to your advisor, then you'd be revealing that your advisor gave you a copy, which some people including the author and editor may not be happy about. Your advisor could sign the review, and then ask the author if it's OK to circulate copies to his/her students. Although if your advisor gave the paper a negative review, that may not be nice.:-)
I think the best option is to pretend you never saw it, until it gets published. That's what I try to do.
Good point about arXiv; it's a good example to bring up.
When I was a grad student at a couple of schools, my advisors never asked me for help reviewing papers. And as I mentioned, I wanted to show my students some papers I'm reviewing, but my colleagues suggested I shouldn't.
What field are your experiences in? Mine is in applied math / mathematical biology / ecology. I know some things vary a lot between disciplines, schools, depts, etc.
But to be honest most people publish when they are *active* researchers and we all know that active *researchers* have to get governmen/business grants. It's the only way to survive.
I mostly agree. Although I hope we aren't heading to a catch-22 situation. When applying for research grants, the proposal reviewers look at the background of the applicant, e.g. how much have they published recently, as one indicator of the person's qualifications. So you need to publish, in order to get grants. If we then end up in a world where you need grants in order to publish, something is wrong.
In practice what happens a lot is that people take post-doc positions paid for by someone else's research grant, and that other person's grant can cover publication costs for the post-doc. This would probably become even more common/necessary under the new system.
Already it's getting harder and harder to get academic jobs. 30 years ago you could become a professor without even having a PhD. My dept is hiring right now, and we are requiring our applicants to have post-doc experience. Personally, I was against that requirement, but got outvoted. But I can see it did give us a higher-quality applicant pool. I bet 10-20 years down the road, many academic job openings will require people to have already gotten $X worth of grants. We're already moving towards that; applicants who have gotten grants are favored over other candidates who haven't, if all else is even close to equal.
Slashdot Mirror
Well, as of 9:30 AM central time 24.224.174.18 isn't accepting connections, so it's either been slashdotted or taken down.
My web logs show some attempts yesterday, but with the wget going to 24.224.174.18. I haven't read up on this worm, but I'm guessing that's an address of someone upstream who was infected. E.g. if my machine infected a new one, the wget for those new infectees would point to my machine.
Yes, I'm aware of the applications of linear algebra to circuits and structural analysis. I also do a simplified version of google page-rank using eigenvalues in my class, and adjacency matrices, error-correcting codes, applications to biological models, economic models, traffic flow through networks, and Markov chains. But I'm still curious what kind of job murr has.
On the other hand, I wouldn't be able to do my current job without Linear Algebra, and there are many days I wish I'd have paid more attention in all my math classes.
;-)
As someone who teaches Linear Algebra, I'm very curious -- what kind of job do you have? (I figure it may be another motivator for my students.) And don't say you are a Linear Algebra teacher.
Back in the old days, read early 90's, we had keyboards with a decent sized return key. I miss them.
If by "decent sized return key" you mean the double-height return keys, I always make sure the keyboards I buy do NOT have those giant return keys. I want a single-height return key, with an easy-to-reach backslash key right above it. The giant-sized return-key keyboards usually move the backslash key way down into the corner, which is a royal pain for those of us who write LaTeX code which needs a million backslashes.
Everybody's got their own preferences...
It's quite possible to be fully productive without a keyboard.
That's funny, I do my very best to avoid using my mouse (and I don't have other input devices), so basically I try as hard as possible to do everything via keyboard, since it's much faster than mouse for me. Although I often wish I had a couple of foot pedals which I could program as control keys or page up / page down or other things (does anyone make such a thing?).
What other input devices do you use? (And by OOS, do you mean Occupational Overuse Syndrome, i.e. Repetitive Stress Syndrome?)
The slashdot article and the engadget blurb both keep using worm and virus interchangeably. Which is it?
After getting my bachelor's degree in CS, I worked for an interesting computer company for almost 2 years. But I simply didn't enjoy the job. Unlike what others are saying here about insane overtime, the hours usually weren't bad at all. My evenings and weekends were usually my own, but the days at work just weren't all that interesting.
I eventually went back to grad school and got a PhD, and am now on the tenure track. It's totally the opposite. Now I'm always insanely busy, evenings and weekends I'm just trying to keep up with teaching and research, and there's always something new coming up to deal with. But I absolutely love the job. Even though there's plenty of mundane stuff to deal with, in a very fundamental way I am mostly free to work on what I want to work on. I'm earning less money than I could, but I can't imagine switching to any other job. Also, even though I'm always busy, in some ways my schedule is way more flexible. Lots of times, if I feel I need to take an hour off in the afternoon, I can go ahead and do it, unless I have any meetings or office hours or class at that time; I don't need to check with anyone or tell anyone. Just as long as I do a good job with my teaching and research, I dictate my schedule, within some constraints.
Also, my research often consists of me just thinking/working on my own, in my own little world. But then I get my recommended daily allowance of social contact by teaching (I do really enjoy interacting with the students), and also by talking about my research with my graduate advisees or other faculty or students, etc.
When I switched from industry/CS to academia/Applied Math, I wasn't sure I'd succeed. I'm still not sure (I haven't reached tenure yet), but so far I don't at all regret the decision to try to do what I really want to do.
Right now I'm in the final throes of preparing a grant proposal, so just coming out of one of the high-stress periods. Now I move on to work on some papers, while waiting the 6+ months to hear back on the proposal...
There are some nasty academic politics to deal with, and other crap too (e.g. plagiarizing students, I hate it when they do that). Every job has its ups and downs. You just need to figure out which one has the downs you can tolerate, and the ups which make it all worthwhile. I'm lucky, I managed to find one for me. And fortunately, we're all different, otherwise we'd all be fighting over the same job!
No, I don't know of any job openings at the moment. I'm a (relatively new) faculty, and if I knew of some job openings, I'd probably hoard the info for my students. :-) I'll tell you what I tell my students, though -- any chance to get involved in any kind of project, for pay or not, is really important. I did tons of projects in my spare time as an undergrad and grad; some were research assistant jobs (even as an undergrad), some were just my own things, but done well enough that I could show them to other people. Summer jobs on some kind of research project really help. My second year as an undergrad, I started knocking on prof's doors until I found some willing to give me some work, which then led to more and more work. And so began the long path to my current job, which will be permanent if I can just make it through tenure.
I did a good dose of measure theory in grad school, and found it very interesting, but haven't really used it since then. And I've taken plenty of applied stats and mathematical statistics, but again I forget most stuff I don't use. Although I do teach elementary stats now and then, and a course in deterministic and stochastic modeling and simulation which involves a lot of Poisson processes.
Hmm, it's true that the exponential distribution has its mode at 0, so in some sense you're "most likely to be infected the moment you connect". But e.g. for an exponential with a mean of 17.3 like we were talking about, you have a 25% chance of first being infected in the first 5 minutes, but still a 19% chance of first being infected within the second 5 minutes, and a 14% chance in the third 5 minutes. So it's not all bunched up at 0 as much as you may imagine.
I think Poisson processes are pretty cool. I like putting them in my modeling class because I can use Poisson processes to tie together the following probability distributions and show relations between them all: continuous uniform, binomial, normal, exponential, and Poisson. Even the students that learned about them in basic probability/stats never realized they were all linked together. Sheldon Ross' book on "Intro to Probability Models" (up to about 9th edition or so now) is a pretty readable book which talks about them quite a bit. They're used to model e.g. failing parts in complex machines, incoming phone calls on a busy phone line, automobile traffic, etc.
Hey, if there's one thing you can count on on /., it's the spelling/grammar nazis and the math police.
:-)
I generally don't bother responding to basic math errors. But this one seemed more interesting. And a guy bragging about being a brilliant mathematician in his sig is just asking for extra scrutiny.
But seriously, being able to totally geek out without limit in a discussion is what this place is all about, right? And after the other response, if you look at my other long reply, I think it's actually not so obvious -- the mean and median really are different. Although it's not over yet, I admit I may still be proved wrong.
I saw your sig, that's why I figured you should know better. :-) I am a mathematician (with a job), but generally don't flaunt it.
:-)
First, the AC didn't say "large number of samples", he/she said "large number" and wasn't very clear about the exact meaning. Yes of course if you compute a sample mean from a large sample, its sampling distribution converges to a normal distribution. You were also not so clear, because when you said "I was thinking of continuous, you were thinking of discrete," you made it sound like those two things are opposites.
Second, of course this distribution can't be truly normal, because it's truncated at 0 on the left. Although I guess you already know, if you want to talk about sample means of large samples, you can generally ignore that since the variance becomes small enough that the probability in the truncated tail is negligible.
Finally, none of the above really matters anyway, as the proper distribution for the time until infection would be an exponential distribution in this case, since there are a very large number of infected machines out there on the network, each with a very small chance of infecting any given target within a reasonably small time interval, and so this system should be fit extremely well by a Poisson process. An exponential distribution with a mean of 12 has a median of about 8.3. Or, if the first line of the article is written correctly and there really is a 50% chance of getting infected within 12 minutes, it means that the median is 12, in which case the mean time to infection is about 17.3 minutes. And these are theoretical means and medians, which are independent of sample size.
And finally finally, I will point out that, given the above information, if you took a really large sample of PCs and measured the sample mean time to infection in that group, the sampling distribution of mean would be normally distributed around 17.3, not around 12.
OK, your turn.
Oh, [Code Red] is still around. LogWatch complains to me nightly of many 404 Not Founds in my Apache access logs for files that Code Red [II] tries to access on IIS web servers.
See, now that's pretty impressive. I wonder when will the last "active/live" instance of Code Red die? Will it still be around after 10 years, in the year 2011? Eventually there won't be enough susceptible machines around to sustain it, when enough people have moved on from the versions of Windows which are vulnerable. But that could be a very long time. Some people still put Windows 98 machines on the Internet.
The fact that you still see it so often means it's not just a handful of machines out there infected by it too; there must be quite a lot in order for you to be seeing them hitting your machine practically every day. (Or, just a bunch "near you" in IP address space, since Code Red II had a kind of mixed movement strategy which include a good chunk of localized scanning.)
To you and the other person who both said "RedHat 9 is obsolete, use something newer and you'll be safe" -- newer isn't always necessary if you secure your machine well.
My firewall at home is an old Pentium with 32MB of memory, running RedHat 6.1. Last time I rebooted it (to plug it into its new UPS) was Sept 9, 2003; it's been running smoothly since then. No one has ever broken into it. Before that I used an older PC for at least 3 years, again with no break-ins.
I don't see any need to update that machine from RedHat 6. All it does is protect my other machines, and it still does that quite well. I use Adelphia cable for internet access, so there are plenty of break-in attempts (although granted, most of the attempts are looking for Windows vulnerabilities).
Anyone can take an unpatched Windows host and put it on the network with no firewall and say "Look! It got owned in X minutes!"
I totally understand the point you're making, but I believe that the information they conveyed (paraphrased above) is still very interesting. It shows that there is still a very large number of infected machines out there, trying to infect new machines. The USA Today story from last November mentioned that within 12 minutes, an unpatched machine was getting infected via vulnerabilities which had been exploited by worms 6 months earlier.
Usually we only hear about the "latest-and-greatest" new worms coming out. Stories like this shows that the old ones aren't gone; there are still vast populations of machines out there infected by very old worms, and as a result, there is an ever-present background noise of things trying to infect new machines.
E.g. consider this -- do you think an unpatched machine put on the net right now could get infected by Code Red II? That came out just about 4 years ago. I honestly don't know the answer to this question, but I'd really like to. Is Code Red II still "alive" out there? Or has it completely died off yet?
But I had continuous distributions in mind when I wrote that, and apparently you were thinking of finite distributions.
Now you seem to be confusing "finite" with "discrete" by saying it's one versus the other. A distribution can be discrete but (countably) infinite, e.g. the Poisson distribution.
Although what this has to do with mean vs median, I don't know. The mean and median are defined for both discrete and continuous distributions, so the fact that the other guy said median instead of mean doesn't mean he was thinking of a discrete distribution.
Remember when editors actually read submissions?
No.
I like to trot out this link in cases like this.
Slashdot editors used to verify stories
The mouse pad cannot be operated on any metal surface. Since the warnings about this are printed on the box, in the manual, and on the mouse pad itself, I figured it was in my best interest to avoid finding out why metal and power-over-RFID don't mesh.
Well there's a sentence that sure didn't end the way I wanted it to... Where's his sense of adventure?
There are some blogs in China, I found a bunch of them I think last year, although I seem to have lost my links to them. I'm not a big blog-reader anyway.
I think people in China are more free to talk about many topics than many Americans think, although you are right, if you talk about the wrong thing, you can end up in deep trouble. But I get the impression that many people in the US think if you even whisper something about governments, the police will come breaking in the door and take you away. I've had many conversations with many people in various places about "forbidden" topics. But I also wouldn't dare publish my ideas on those topics publicly in China.
I'm actually about to head back to China for 2 months; guess I'll soon see what's hot there these days.
I think it was about 2 or 3 years ago, there was an incredibly popular blog in China, written by a woman, about her sexual experiences. (And from what I read, yes, it really was written by a woman.) I'm blanking on her name, and can't even remember if she was in Hong Kong, or maybe Guangzhou. Anyway, it was incredibly popular, among people in China, and overseas. I think that put the idea of blogging on the map, in people's minds in China. But it seems to have faded from the public consciousness since then.
I'm sure someone here knows what I'm talking about and can post a link to more info.
Agreed. I really hate it when I receive a conference announcement, and after reading it, have absolutely no idea what the conference is about. In those cases, I'd think the people publicizing the conference would want people to know just what is being advertised. This is a little different, being a story about the conference rather than an announcement trying to attract people to it, but it's still the same problem...
Today I just submitted a research grant proposal to the US Department of Energy, in their "Multiscale Mathematics in Research and Education" program. If you look at their proposal guidelines, near the end under "The evaluation under item 2", they talk about making materials available to the public as open source. I was happy to see it, as I have made some software available from my previous research available in this way, and plan to do so again in the future on my new projects. I'll have to wait many months to see if they fund my current project proposal, though...
A dairy farmer could probably right a book on cows, cow technology, cow behaviour, and how all that relates to his philosophy of life and why his kid is studying rocket science at university.
First, I think you mean "write", not "right".
Second, I have a book I inherited from my great-grandfather (a farmer). The book was published in 1944, and is called "Cowphilosophy". No joke. The subtitle is "The Art of Practical Dairy Practice". Inside the front cover, before the title page, is a page with some pictures of cows, and the text:
We Are Your Cows
We have to eat what you provide.
Drink what you give us.
Live where you put us.
We may be good cows or we may not.
We may be healthy, or we may not.
We may be comfortable or we may not.
We may be profitable, or we may not ---
So much depends on you, the dairyman.
THE FIRST REQUISITE OF A PROFITABLE DAIRY BUSINESS IS A GOOD DAIRYMAN!
(Strange but true. I love having that book on my shelf. I didn't grow up to be a farmer, though. I ended up getting a PhD in applied math. But I can always consult my Cowphilosophy book when I need some real wisdom.)
Kraft had suggested that Mrs Budimir launch a new website under the domain name www.milkacouture.fr...
Mrs. Budimir should suggest that Kraft re-launch their own website, under the name www.butthead-astronomers-chocolate.fr.
I'm a third-year assistant prof in applied mathematics now, just for reference.
:-) So it was good enough for him to cite, but not for him to recommend that that journal publish it. I finally did get it published elsewhere, but it took some time.
:-)
I've reviewed papers relevant to my current research, but nothing I felt I really needed to cite in my own papers. But there really isn't a way to cite a paper you anonymously reviewed, without breaking anonymity (unless you simply wait for it to be published). That essentially happened to me once. A guy reviewed my paper; he said he liked it, but didn't think it belonged in the journal I'd submitted it to, which caused it to be rejected by that journal. I happened to give a talk in his dept around that time, and he told me he was reviewing my paper, and what his review would say. He then proceeded to cite my paper a few times in his own papers over the next couple of years, citing it as "A. Name, unpublished manuscript" or something along those lines. It was annoying, since it would have been published if not for his review.
So you could identify yourself to the author of the paper, i.e. sign your review, which is something I think you're always allowed to do if you so choose. Except in your case if the paper was sent to your advisor, then you'd be revealing that your advisor gave you a copy, which some people including the author and editor may not be happy about. Your advisor could sign the review, and then ask the author if it's OK to circulate copies to his/her students. Although if your advisor gave the paper a negative review, that may not be nice.
I think the best option is to pretend you never saw it, until it gets published. That's what I try to do.
Good point about arXiv; it's a good example to bring up.
When I was a grad student at a couple of schools, my advisors never asked me for help reviewing papers. And as I mentioned, I wanted to show my students some papers I'm reviewing, but my colleagues suggested I shouldn't.
What field are your experiences in? Mine is in applied math / mathematical biology / ecology. I know some things vary a lot between disciplines, schools, depts, etc.
But to be honest most people publish when they are *active* researchers and we all know that active *researchers* have to get governmen/business grants. It's the only way to survive.
I mostly agree. Although I hope we aren't heading to a catch-22 situation. When applying for research grants, the proposal reviewers look at the background of the applicant, e.g. how much have they published recently, as one indicator of the person's qualifications. So you need to publish, in order to get grants. If we then end up in a world where you need grants in order to publish, something is wrong.
In practice what happens a lot is that people take post-doc positions paid for by someone else's research grant, and that other person's grant can cover publication costs for the post-doc. This would probably become even more common/necessary under the new system.
Already it's getting harder and harder to get academic jobs. 30 years ago you could become a professor without even having a PhD. My dept is hiring right now, and we are requiring our applicants to have post-doc experience. Personally, I was against that requirement, but got outvoted. But I can see it did give us a higher-quality applicant pool. I bet 10-20 years down the road, many academic job openings will require people to have already gotten $X worth of grants. We're already moving towards that; applicants who have gotten grants are favored over other candidates who haven't, if all else is even close to equal.