As a grad student - in Tommy Gold's department, no less (-: - I have to say that I'm a firm believer in peer review. Like all systems, it has its failings: I could tell you about the referees who lost papers, the ones who sat on papers until their versions of the same theory were published, the ones who are ignorant, biased, narrow minded or just plain stupid...
But: on average, it works.
True, it is biased towards incremental progress rather than revolutions, but that is the way science works most of the time. Most of us do not recreate whole systems of thought like Feynman (another iconoclast idol) - the times that require that are few and far between. (Though for physics, the current impasse with GUTs might be one of them.)
The reason we have to submit to the tyranny of peer review is simple: no one is an expert on everything. With our increasingly narrow specialisations, I know next to nothing about topic A at wavelength B, though I'm the world's expert on topic C. So if someone says A affects C in some way, I have much less chance of judging his claim correctly than another expert in A. But I should hear about, not have that view supressed by others, right?
So now we have preprint servers. One little 386 (yes!) at LANL archives all the submitted preprints in astronomy and physics on a daily basis - some people submit them after peer review and acceptance at a major journal (to prevent embarrassing retractions), some people submit them as soon as they send in a paper (to establish priority) and some people just publish papers on the preprint servers (and we know them well, as kooks of various kinds).
In this day, when results are shared at conferences and off webservers, journals are having an increasingly hard time justifying subscriptions. (I read preprint abstracts daily, and never use dead tree journals...) So they are evolving into keepers of standards - if its published in the Fancy Journal of UnGnomon News, it must be good stuff on Gnus!
That's quite enough raving - but as for this comment in the article: 'He [..] migrated to a "much more livable" environment at Cornell' - let me just add: "yeah, right!"
Okay, so this is the obvious question, but it has to be asked by someone (shades of first post here...):
What made you choose Linux over a standard so-called off-the-shelf solution like Solaris or NT? Or easy-ware like Macs? If you were already a convert to Open/Free software, did you consider the *BSDs? And have you had occasion to regret your decision? In terms of scalability, stability, price, TCO, or your favorite marketroid jargon term?
That's all one question. Really! Just ignore the man behind the curtain:-)
I'd like to second (third?) the idea of a Slashdot User Forum. Or maybe an "About Slashdot" section like "Ask Slashdot" - though I like the regular thread better.
That way, when I notice a little something like "Insightful" spelt "Insghtful", or a grammatical error that bothers the obsessive-compulsive in me, I can post it to the vast faceless group running slashdot (-: without the pangs of guilt that prevent me from adding minor quibbles to CmdrTaco's overflowing (I'm sure) mailbox.
Oh, and this comment also demonstrates the first +5 post gravity effect - I think shuffling the few top +5 postings is a neat idea.
Okay, so the rebuttal above should really get moderated up, since it makes much of this discussion hypothetical.
But while we're speaking hypothetically, combine the coke machine with a finger daemon (like so - or try finger coke@l.gp.cs.cmu.edu) and what do you get? The Coca Cola weather service!
Even more fun - maybe you could telnet in from your Playstation to the coke machine and start a multiplayer contest... the possibilities are endless!:)
Oh, and yes, I forgot to put TM and (R), but Coke, Coca Cola and Playstation are trademarks of... wait a minute, this is fair use of trademarks, for dummies, even. Nyah, nyah, nyah!
And this is the real pity - with NASA's new "faster better cheaper" program, the missions do exactly what was in the initial mission, no more (though sometimes much less - look at the Climate Orbiter fiasco).
So something like the Sojourner Rover died in a month (its minimal design goal was a week) - if this had been one of the older Cadillac-style overengineered no expense spared missions, it would still be chugging away and getting good science done, instead of having provided us with tantalising glimpses of stuff and then making us sit and twiddle our thumbs for two years.
Of course, "faster better cheaper" means that a single failure is not catastrophic - so the loss of the previous Mars Polar mission was a total disaster (it was one of the big missions) and left us with nothing going towards Mars for years, while the Climate Orbiter failure is unfortunate, but only a six month to one year delay.
Meanwhile, the last of the old dinosaurs speeds on towards Saturn - in spite of the stupid and misguided protests over its RTG power source. People in the planetary science community are praying that Cassini does an Energizer Bunny on them, because the next look at Saturn is at least another decade away...
Belluzzo might be joining Microsoft, says this cnet/NYT article.
"When Belluzzo's resignation was announced yesterday, Silicon Graphics spokesman John Cristofano would say only that the departing CEO was leaving to take a position at a company that doesn't compete with SGI." Well, not yet, anyway... this sounds very fishy.:)
Slashdot Mirror
But: on average, it works.
True, it is biased towards incremental progress rather than revolutions, but that is the way science works most of the time. Most of us do not recreate whole systems of thought like Feynman (another iconoclast idol) - the times that require that are few and far between. (Though for physics, the current impasse with GUTs might be one of them.)
The reason we have to submit to the tyranny of peer review is simple: no one is an expert on everything. With our increasingly narrow specialisations, I know next to nothing about topic A at wavelength B, though I'm the world's expert on topic C. So if someone says A affects C in some way, I have much less chance of judging his claim correctly than another expert in A. But I should hear about, not have that view supressed by others, right?
So now we have preprint servers. One little 386 (yes!) at LANL archives all the submitted preprints in astronomy and physics on a daily basis - some people submit them after peer review and acceptance at a major journal (to prevent embarrassing retractions), some people submit them as soon as they send in a paper (to establish priority) and some people just publish papers on the preprint servers (and we know them well, as kooks of various kinds).
In this day, when results are shared at conferences and off webservers, journals are having an increasingly hard time justifying subscriptions. (I read preprint abstracts daily, and never use dead tree journals...) So they are evolving into keepers of standards - if its published in the Fancy Journal of UnGnomon News, it must be good stuff on Gnus!
That's quite enough raving - but as for this comment in the article: 'He [..] migrated to a "much more livable" environment at Cornell' - let me just add: "yeah, right!"
What made you choose Linux over a standard so-called off-the-shelf solution like Solaris or NT? Or easy-ware like Macs? If you were already a convert to Open/Free software, did you consider the *BSDs?
And have you had occasion to regret your decision? In terms of scalability, stability, price, TCO, or your favorite marketroid jargon term?
That's all one question. Really! Just ignore the man behind the curtain :-)
That way, when I notice a little something like "Insightful" spelt "Insghtful", or a grammatical error that bothers the obsessive-compulsive in me, I can post it to the vast faceless group running slashdot (-: without the pangs of guilt that prevent me from adding minor quibbles to CmdrTaco's overflowing (I'm sure) mailbox.
Oh, and this comment also demonstrates the first +5 post gravity effect - I think shuffling the few top +5 postings is a neat idea.
Just my "me too" for the day...
But while we're speaking hypothetically, combine the coke machine with a finger daemon (like so - or try finger coke@l.gp.cs.cmu.edu) and what do you get? The Coca Cola weather service!
Even more fun - maybe you could telnet in from your Playstation to the coke machine and start a multiplayer contest... the possibilities are endless! :)
Oh, and yes, I forgot to put TM and (R), but Coke, Coca Cola and Playstation are trademarks of ... wait a minute, this is fair use of trademarks, for dummies, even. Nyah, nyah, nyah!
So something like the Sojourner Rover died in a month (its minimal design goal was a week) - if this had been one of the older Cadillac-style overengineered no expense spared missions, it would still be chugging away and getting good science done, instead of having provided us with tantalising glimpses of stuff and then making us sit and twiddle our thumbs for two years.
Of course, "faster better cheaper" means that a single failure is not catastrophic - so the loss of the previous Mars Polar mission was a total disaster (it was one of the big missions) and left us with nothing going towards Mars for years, while the Climate Orbiter failure is unfortunate, but only a six month to one year delay.
Meanwhile, the last of the old dinosaurs speeds on towards Saturn - in spite of the stupid and misguided protests over its RTG power source. People in the planetary science community are praying that Cassini does an Energizer Bunny on them, because the next look at Saturn is at least another decade away...
Belluzzo might be joining Microsoft, says this cnet/NYT article.
:)
"When Belluzzo's resignation was announced yesterday, Silicon Graphics spokesman John Cristofano would say only that the departing CEO was leaving to take a position at a company that doesn't compete with SGI." Well, not yet, anyway... this sounds very fishy.
--------