Two related points... First, if your institution feels like coughing up a wad of cash anyway, you get a discount on publication fees. For the equivalent of a typical institutional journal subscription ($2k/yr), you get 10% off.
Second, I know that BMC waives publication fees for authors that have recently done reviews for them. I haven't reviewed for the PLOS journals yet, but I'd bet they do something similar.
I grabbed the 1.7.2 source tarball and configured as usual (Linux/gtk2, disabled everything but the browser), but there appear to be some files missing. The build first bombs when trying to create nsBuildID.h (missing the Perl module Moz/Milestone.pm, which I assume is part of the developers' environment). If I hand-copy nsBuildID.h.in to nsBuildID.h, it then bombs later because a variety of the needed Makefile.in files are missing (there are a bunch of warnings about this during configure).
I tried grabbing the 1.7.2 release directly from CVS, but the same problems occur. Anyone have any better luck? Was there an FM that I forgot to R? I had no problems building 1.7 from source by the same procedure.
In that case, the editors got the tense of the headline wrong... I think it should be something like "How Google Wioll Haven be Achievening the Semantic Web".
Finally, a university has built a weirder-looking CS building than the one at my undergrad institution. MIT's new building makes good old Duncan Hall look positively conservative.
Disclosure: I am a CS professor (in comp bio, not networking), and my department (WUSTL) is well-known for networking-related research.
If you want to get into stuff like protocol design, routing architecture and algorithms, and the other "guts" of designing and building high-speed networks, consider a CS or Computer Engineering degree. Besides the obvious courses in networking per se, you will have the opportunity to study algorithms relevant to the area (e.g. minimum spanning tree, network flows, suffix trees and other fast string matching methods for routing tables); hardware design (for building gigabit-and-up routers and other cool network gear); and design of large software systems, including the principles behind distributed systems like the DNS and peer-to-peer networks.
Depending on where you go and what your interests are, you can also bone up on the underlying math (e.g. queueing theory for protocol design) and maybe even some physics/EE (signal propagation, etc).
Yes, you can probably work your way to a strong practical knowledge of how to build a network, and getting your hands dirty is essential to success. However, the point of a CS degree is (1) breadth in computing fundamentals, which I hope I've persuaded you are relevant to networking, and (2) quickly getting up to speed on how to *think* about networks, independent of any particular protocol or hardware standard.
Oh, and speaking of getting hands dirty... if you go to the right place, you might be able to get on board a networking-related research project!
I think the poster was misremembering a classic stock scam. What he *really* meant was this:
(1) Identify 2^n gullible people as "marks."
(2) Send each mark an email predicting the outcomes of the next n successive events (I'm assuming the outcomes are all yes/no). Each mark gets one of the 2^n possible sets of predictions.
(3) Wait for the n events to unfold.
(4) Identify the one mark who got an email with all correct predictions, and persuade him to subscribe to your "amazingly accurate" prediction service for a hefty fee.
(5) Abscond.
In the original version of the scam, the events were that a particular stock would go up or down on each of a succession of days.
Why does a language with only 120 words and one tense need at least two irregular verbs and two exceptions to the rule on subject-verb linkage?
In any case, what good is a language where you can't even tell a joke whose punchline is "Mister, that's the first time I've ever heard that verb used in the pluperfect subjunctive!"?
"So why do universities and research labs continue to judge hackers by publications? For the same reason that "scholastic aptitude" gets measured by simple-minded standardized tests, or the productivity of programmers gets measured in lines of code. These tests are easy to apply, and there is nothing so tempting as an easy test that kind of works."
Did anyone else find this rather dyspeptic comment ironic given the author's views on hacking?
Some subway systems have ads on their tunnel walls that are meant to be viewed at the speed of a moving train. In the future, perhaps advertisers targeting Tivo users will buy an extra-long commercial slot to play a greatly slowed-down version of their regular ad that appears normal when fast-forwarded.
Of course, Tivo will immediately counter with a fast-fast-forward mode for such ads, which will be met with even more slowed-down spots, and so on...
At first, I was surprised that the comments for this article hadn't degenerated into a flame-fest about IP models and renting vs buying books. After a bit more thought, however, I think there are two key factors that make this model appealing.
Firstly, many if not most O'Reilly technical books have a fairly limited useful life. They must evolve to track tool version upgrades, and many potential users don't need the content of a particular book for longer than it takes to complete a project. Secondly, the service offered is not rental of a specific book, but rather online access to a broad library of titles. You're paying for the convenience of not having to find a physical lending library that carries the book.
What other books would be appropriate for such a rental model? If you could access Springer-Verlag's yellow books online under a similar model, would you do it? How about a library of major American literature, 1920-1970? How about a library of (mostly pot-boiler) mystery or scifi titles? If the usability issues can be worked out, I would seriously consider buying a monthly subscription to an up-to-date online scifi library (think of a much expanded version of what Baen now offers for free). It would probably be a better value than the tripe I get from my cable TV subscription.
Two last items for thought, however. First, if this model becomes as successful as I think it could, will we be in danger of losing the option of purchasing a book outright? Second, what will be the fate of public libraries?
Slashdot Mirror
Here's a link with actual content on what the algorithm does:
http://www.hpcwire.com/features/IBM-Invents-Short-Cut-to-Assessing-Data-Quality-85427987.html
Now that the thieves can't simply sneak in and make off with the torahs, they will presumably resort to armed robbery, or "torah-jacking".
Just imagine a masked gunman barging into the sanctuary in the middle of the service and shouting, "Gimme loot, chasidim!"
... All those moments will be lost in time, like tears in the rain. Time to die.
Two related points... First, if your institution feels like coughing up a wad of cash anyway, you get a discount on publication fees. For the equivalent of a typical institutional journal subscription ($2k/yr), you get 10% off.
Second, I know that BMC waives publication fees for authors that have recently done reviews for them. I haven't reviewed for the PLOS journals yet, but I'd bet they do something similar.
I grabbed the 1.7.2 source tarball and configured as usual (Linux/gtk2, disabled everything but the browser), but there appear to be some files missing. The build first bombs when trying to create nsBuildID.h (missing the Perl module Moz/Milestone.pm, which I assume is part of the developers' environment). If I hand-copy nsBuildID.h.in to nsBuildID.h, it then bombs later because a variety of the needed Makefile.in files are missing (there are a bunch of warnings about this during configure).
I tried grabbing the 1.7.2 release directly from CVS, but the same problems occur. Anyone have any better luck? Was there an FM that I forgot to R? I had no problems building 1.7 from source by the same procedure.
In that case, the editors got the tense of the headline wrong... I think it should be something like "How Google Wioll Haven be Achievening the Semantic Web".
(Dr. Streetmentioner, please call your office.)
Thank you, Captain Obvious.
Hath smaller fleas that on him prey;
And these have smaller still to bite 'em;
And so proceed ad infinitum.
- Swift
Sorry, why are we bringing Italy into this?
Finally, a university has built a weirder-looking CS building than the one at my undergrad institution. MIT's new building makes good old Duncan Hall look positively conservative.
Clearly, you never had a class from my undergrad real analysis prof.
:-)).
(For that matter, you've never taken my automata theory class
Well, I myself was hoping for "Yuggoth."
Disclosure: I am a CS professor (in comp bio, not networking), and my department (WUSTL) is well-known for networking-related research.
If you want to get into stuff like protocol design, routing architecture and algorithms, and the other "guts" of designing and building high-speed networks, consider a CS or Computer Engineering degree. Besides the obvious courses in networking per se, you will have the opportunity to study algorithms relevant to the area (e.g. minimum spanning tree, network flows, suffix trees and other fast string matching methods for routing tables); hardware design (for building gigabit-and-up routers and other cool network gear); and design of large software systems, including the principles behind distributed systems like the DNS and peer-to-peer networks.
Depending on where you go and what your interests are, you can also bone up on the underlying math (e.g. queueing theory for protocol design) and maybe even some physics/EE (signal propagation, etc).
Yes, you can probably work your way to a strong practical knowledge of how to build a network, and getting your hands dirty is essential to success. However, the point of a CS degree is (1) breadth in computing fundamentals, which I hope I've persuaded you are relevant to networking, and (2) quickly getting up to speed on how to *think* about networks, independent of any particular protocol or hardware standard.
Oh, and speaking of getting hands dirty... if you go to the right place, you might be able to get on board a networking-related research project!
Can we take up a collection and buy these guys some consonants? Sheesh!
I think the poster was misremembering a classic stock scam. What he *really* meant was this:
(1) Identify 2^n gullible people as "marks."
(2) Send each mark an email predicting the outcomes of the next n successive events (I'm assuming the outcomes are all yes/no). Each mark gets one of the 2^n possible sets of predictions.
(3) Wait for the n events to unfold.
(4) Identify the one mark who got an email with all correct predictions, and persuade him to subscribe to your "amazingly accurate" prediction service for a hefty fee.
(5) Abscond.
In the original version of the scam, the events were that a particular stock would go up or down on each of a succession of days.
Guilty as charged.
"Traven" tapes? As your experience indicates, their motto must be:
Backups? We ain't got no backups! We don' need no stinking backups!
(OK, OK, -1 obscure)
He must be considering alternative splicing.
Why does a language with only 120 words and one tense need at least two irregular verbs and two exceptions to the rule on subject-verb linkage?
In any case, what good is a language where you can't even tell a joke whose punchline is "Mister, that's the first time I've ever heard that verb used in the pluperfect subjunctive!"?
"So why do universities and research labs continue to judge hackers by publications? For the same reason that "scholastic aptitude" gets measured by simple-minded standardized tests, or the productivity of programmers gets measured in lines of code. These tests are easy to apply, and there is nothing so tempting as an easy test that kind of works."
Did anyone else find this rather dyspeptic comment ironic given the author's views on hacking?
If you think the Mortar Ballistic Computer is too big, just compare it to this
earlier model.
Some subway systems have ads on their tunnel walls that are meant to be viewed at the speed of a moving train. In the future, perhaps advertisers targeting Tivo users will buy an extra-long commercial slot to play a greatly slowed-down version of their regular ad that appears normal when fast-forwarded.
Of course, Tivo will immediately counter with a fast-fast-forward mode for such ads, which will be met with even more slowed-down spots, and so on...
Yes, and the screen adaptation made it a whole 45 minutes (including commercials) before confusing "nova" with "supernova." Sigh.
At first, I was surprised that the comments for this article hadn't degenerated into a flame-fest about IP models and renting vs buying books. After a bit more thought, however, I think there are two key factors that make this model appealing.
Firstly, many if not most O'Reilly technical books have a fairly limited useful life. They must evolve to track tool version upgrades, and many potential users don't need the content of a particular book for longer than it takes to complete a project. Secondly, the service offered is not rental of a specific book, but rather online access to a broad library of titles. You're paying for the convenience of not having to find a physical lending library that carries the book.
What other books would be appropriate for such a rental model? If you could access Springer-Verlag's yellow books online under a similar model, would you do it? How about a library of major American literature, 1920-1970? How about a library of (mostly pot-boiler) mystery or scifi titles? If the usability issues can be worked out, I would seriously consider buying a monthly subscription to an up-to-date online scifi library (think of a much expanded version of what Baen now offers for free). It would probably be a better value than the tripe I get from my cable TV subscription.
Two last items for thought, however. First, if this model becomes as successful as I think it could, will we be in danger of losing the option of purchasing a book outright? Second, what will be the fate of public libraries?
It's been said before, and much better.