Basically, MITs reason is much more prosaic than some sort of "intellectual freedom" (*snort*). If MIT was restricted from using foreign students on a research project, that project would very likely die for lack of interest. Let's face it: US students are for the most part smarter than wasting 4-7 years in a PhD program, given the financial risk involved. Thus we have the observation that the graduate student bodies at many (most?) large research schools are foreign.
From the universities point of view, this is just good business sense: large pool of willing labor that works cheap. Also, don't be fooled by high tuition for foreign. That's just a numbers game to charge what is essentially more overhead for these same grants.
MIT sees this as a dangerous precedent. If they accepted these restrictions, it would be more difficult to refuse them later, and could really dent the supply of indentured labor available to fuel university researh programs.
The whole thing make sense in the context of an "education industry".
how easy it is these days to install software from source, compared to the old days where everyone wrote their own, usually broken Makefiles, to build and install into wacko locations../configure; make; make install is a *dream* compared to 10 years ago. And really, compared to 10 years ago, watching configure do its thing is beautiful... you think rpm dependency is hell... try having header/library dependency.
Also, rmp/deb formats are often not as convenient as source, especially for developing or maintaining software.
Finally, there is no reason not to believe that building software will get easier and easier. Easy enough that a moderately competent user should be able to build from source.
I wouldn't laugh too hard at Berkeley for accepting any class at 100 level up for graduate credit, unless you have been here at Berkeley and aced a few few of those. There is not a lot of grade inflation in math, physics or engineering at Cal. A- and up grades are limited. On the other hand, the curves can be so low that very few actually fail.
The undergrad classes I took in math (analysis 104, linear algebra 110 and num analysis 128) to satisfy minor requirements for PhD in engineering were easily as difficult as any of the graduate classes.
The difference at Berkeley is most likely due to being perceived as the best school in the UC system, which is in the most populous state (CA) in the US, which has many top-ranked departments, which is also public with incredibly low fees (order of $3k/semester). Competition is cutthroat.
Many of the grad programs at Berkeley are professional programs in disguise. That is, they are MS programs with project rather than thesis requirements. This results in an annual crop of MS grads with known skills; employers like this, especially employers with Cal degrees. Admittance to the PhD program is usually predicated on doing well in the MS program. The weakness of this is that people interested in PhD level work often could care less about the crap they teach the MS people, so one has to compete against people who only want to "get the answer" instead of understanding the material. Another weakness is that it tends very much to induce a sort of intellectual rigidity in thinking among the faculty and grad students with these kinds of programs: if one didn't learn it at Cal, it isn't worth knowing. Such hubris will of course those particular departments in the long term.
Finally, where Berkeley, Stanford, etc. really score is the reputation really does influence the *perception* of education... and too often the perception is the reality...
The easier it is for a different engineer to review your work the better. No matter where you work, someone has 1. approve the design work, 2. be legally responsible (as in getting massively sued). This is why Professional Engineers (PEs) are licensed and have a spiffy little stamp they sign tocertify the design.
So, yeah, write out everything. If you make a few blunders here and there they should be easy to spot.
Dealing with minor physical and mental illnesses used to be a hallmark of maturity and professionalism. Nowadays it seems like those of us who just get on with it are punished for our insensitivity, and/or our ability to deal with such matters independently means, _by definition_, that we don't actually, *really* suffer. Because if we did, we would obviously be in the care of "mental health" professionals.
Heh... enjoy it! Probably the last Bud you ever have, with your broken ankle, as your body core temperature drops below 90 F... At least hypothermia is painless...;)
RC in caves is a ludicrous concept, save for line of sight, but then, why bother. Cave radios work with really looooowwwwww frequencies and require rather large coil antennas to transmit through all that rock. Cave radios are pretty finicky too. This is why cave rescue organizations (good ones) have the ability to lay a mile or two of phone wire in really horrible conditions.
RC butterflies or RC anything-else just ain't gonna happen in a cave.
I asked the dad dude about this once. He harks from the old school: cobol, snobol, etc etc. His answer was that PCs move small amounts of data really fast, a needle thin stream moving at light speed. Mainframes move data like Niagara Falls.
Chiro is interesting for a couple of reasons. The most interesting is that a good chiro will actually _listen_ to you. Every doctor I have seen either for myself or with my wife in the last 10-20 years always seems to be paying attention to something else... maybe their afternoon golf date or whatever. The second interesting thing about chiro is that it only works if the patient is proactive. Can't take a pill, no no no. Gotta actually do the damn exercises. It requires more work than going to the doctor. A third interesting thing about a good chiro is that if you say your back (or whatever) hurts, they might believe you. In contrast to an MD that may say "its all in your head". (As told to my sister-in-law for years until they discovered, too late, that she had a degenerative bone disease. The pain was "all in her head". Chiro of course would not have helped her, but the orthodox medical establishment didn't really help much either.)
Lastly, a really good chiro will have a short list of fine surgeons to refer you to if you have an injury beyond their capability. It helps to have one that has actually had back surgery...:).
Hope this helps. Don't believe the hype by any means, but for the kind of (very real) back pain I have occasionally, a $60 visit couple of times a year to the chiro is cheap relief.
in the treacherous seas of search is going to be this companies biggest hurdle. Assuming they don't founder in the sinking economy (unless they get bailed out), they had better run a tight ship. Because even if they win, they will likely lose, when the patent holders come to collect license fees.
Internet search space is seriously covered up with multiple overlapping patents on a few relatively simple notions about how to find stuff. It's a lawyers game now, not a technology game.
I did the research: Start with Salton's or Ellen Voorhees work and go forward.
Bottom line: boolean matching works "unreasonably well", clustering is more expensive for only marginally better results. Page ranks (big-ass SVDs) are done periodically, cost is cheap amortized over boolean queries. With clustering, you gotta do more work for every query.
For a small document collection (single pc), and no measurable queries (i.e., single, infrequent user), this might be nice. Then again, if it is a small enough collection to be viable, its probably overkill.
Some of these other techniques look interesting though. Might have to fire up google and investigate.
For groxis though, it doesn't really matter anyway. This entire space is covered by multiple overlapping patents. So even if they win, they lose...
I just spent around 20 hours reviewing a paper for an international engineering journal. I was not paid for this activity. My review was anonymous to the authors of the paper (accept with revisions, if you must know). And this paper was exactly in my field of expertise.
At my current billable rate, that is around $1500-2000 worth of labor, minimum. I do it for free because the authors submitting to such journals are not paid.
If it were a consulting report, you can be damn sure I would charge $75-100/hour to check someones engineering analysis.
Where do you propose to get all of this free labor for reviewing? It's damned hard work.
BSD license is a fine license for many. Others, such as myself, prefer GPL/LGPL. In fact, given a choice of proprietary or BSD, I would pick purely proprietary. I am sure many others would do the same.
One semi-hothead spouts off, 13 bazillion other buffoons jump in.
Lot more talking than thinking going on around here this morning.
Isn't peer pressure an awesome force?;> Situations like this really point out the strength of the open sources (sores heh heh heh) community. Would Bill Gates listen to anyone telling him he was full of buffalo sh*t? Don't think so. So, BP gets a mouthful of shoe leather. Once he gets tired of the taste of pavement, he will think harder before speaking...;)
The cygwin stuff is not quite there yet. It's close. I would use it a *lot* more of if I could get xterm running. The windows command terminal really really blows, even with bash running. So I see GNU/NT held back by the lack of an xserver + xlibs mostly. Once that was running, I think the rest of it would fall into place pretty fast. Why is this important? Couple of reasons. 1. Some of us are stuck in NT for at least the next few years. 2. It is much easier to train potential Linux users if there were a way to ease the transition from pure gui to not-pure-gui. This way they can see that DOS is a hopeless POS, but a modern shell is extremely useful. And that the only point of similarity between the two is the CLI. I am sure I will get roundly flamed for this, especially over point 1. But, whatever. I got stuff to do, and I have already gotten 1 pure linux box (debian!) and a dual boot box (debian!) in the office. It is just going to take a few years to complete the transition. (Actually, having a couple of NT boxes around *is* useful.).
This is really typical though. Free speech is not much different than free markets. Both depend on just exactly who gets to define the words _free_, and in this case, _speech_. It appears that the judge and plaintiffs get to define words here.
Then again, pre-web, just being on the internet was dues to the Fringe Element of Society Club.
Ah well, just an old codger here, pining for the days of kibology and p-adic insanity.
Actually, since they get to lay a big chunck of this cable under emininent domain, I believe the tax-payers have every right to set some ground rules. What right does ATT or any other utility have to clutter up my skyline with their ugly poles? Seems fair to me. ATT gets to use "public property", that is, easements taken under eminent domain, and maintained at taxpayer expense, for a non-monopoly share of infrastructure running open standards. Why is this so difficult to understand?
If ATT wants to use a private property argument, they should purchase all their own easements.
MS has gored my very own personal ox in more ways than one. Since I do not work for MS, nor do any of my friends and family, and my Van Kampen fund is only 3% into MS, my particular vote, as a taxpaying US citizen, is to hammer MS hard.
As such, Barbara Boxer will be sure to hear my side of the story if necessary. She represents my interests, which, as I have just mentioned, MS takes little regard for.
Governments grant joint stock companies the privilege to do business as they see fit, and protect individual businessmen by and largely from personal liability. The government gives, it takes away.
So, friend, our two votes cancel out... but we aren't the only two voting, so the future remains interesting!
I don't currently use it because of libc incompatibility which I have not had time ti rectify. This does point out a *major* shortcoming in commercial software: the free software world moves much faster. It just seems ridiculous to me that a recompile from libc5 to libc6 requires me to fork out another $300 for an upgrade.
So, how does it work? Not too bad, even on a p100 with 48M ram. Other than an inordinately long startup time, the program it self runs fine once it gets under. I did notice color map problems in later versions of xfree86. The biggest issue that I have is the postscript export routine makes the assumption that one is running windows. So every line ends with ^M, and there is a binary end-of-file trailer that must be removed. This points out another serious weakness of commercial software. I (rightly) regard the postscript problem as a bug, and could fix it pretty fast if I had the source code. I am sure though that this falls way down on Bentley's priority list, so I am not holding my breath. What would be superb would be to have fig2dev support for dgn->ps, but now I am really dreaming.
Another issue is possible lack of support for graphics tablets. Mouse driven pointers SUCK compared to a 4-button puck on a graphics tablet set to move under absolute coordinates. ustation certainly supports tablets, but I have never seen any linux boxes with a tablet hooked up, or much documentation for doing so.
I tend to agree. For about the last six months, the rhetoric emanating from Redmond has taken on an increasingly delusional appearance. That Gates is a genius is not a point of contention. Neither is the empirical fact that many such people end up being bona-fide kooks. Doesn't really matter in the end. Gates is mortal, and the further he reaches the farther he will fall.
Slashdot Mirror
Basically, MITs reason is much more prosaic than some sort of "intellectual freedom" (*snort*). If MIT was restricted from using foreign students on a research project, that project would very likely die for lack of interest. Let's face it: US students are for the most part smarter than wasting 4-7 years in a PhD program, given the financial risk involved. Thus we have the observation that the graduate student bodies at many (most?) large research schools are foreign.
From the universities point of view, this is just good business sense: large pool of willing labor that works cheap. Also, don't be fooled by high tuition for foreign. That's just a numbers game to charge what is essentially more overhead for these same grants.
MIT sees this as a dangerous precedent. If they accepted these restrictions, it would be more difficult to refuse them later, and could really dent the supply of indentured labor available to fuel university researh programs.
The whole thing make sense in the context of an "education industry".
how easy it is these days to install software from source, compared to the old days where everyone wrote their own, usually broken Makefiles, to build and install into wacko locations.
Also, rmp/deb formats are often not as convenient as source, especially for developing or maintaining software.
Finally, there is no reason not to believe that building software will get easier and easier. Easy enough that a moderately competent user should be able to build from source.
I can assure that it helps to have a wife that would appreciate such effort.
I wouldn't laugh too hard at Berkeley for accepting any class at 100 level up for graduate credit, unless you have been here at Berkeley and aced a few few of those. There is not a lot of grade inflation in math, physics or engineering at Cal. A- and up grades are limited. On the other hand, the curves can be so low that very few actually fail.
The undergrad classes I took in math (analysis 104, linear algebra 110 and num analysis 128) to satisfy minor requirements for PhD in engineering were easily as difficult as any of the graduate classes.
The difference at Berkeley is most likely due to being perceived as the best school in the UC system, which is in the most populous state (CA) in the US, which has many top-ranked departments, which is also public with incredibly low fees (order of $3k/semester). Competition is cutthroat.
Many of the grad programs at Berkeley are professional programs in disguise. That is, they are MS programs with project rather than thesis requirements. This results in an annual crop of MS grads with known skills; employers like this, especially employers with Cal degrees. Admittance to the PhD program is usually predicated on doing well in the MS program. The weakness of this is that people interested in PhD level work often could care less about the crap they teach the MS people, so one has to compete against people who only want to "get the answer" instead of understanding the material. Another weakness is that it tends very much to induce a sort of intellectual rigidity in thinking among the faculty and grad students with these kinds of programs: if one didn't learn it at Cal, it isn't worth knowing. Such hubris will of course those particular departments in the long term.
Finally, where Berkeley, Stanford, etc. really score is the reputation really does influence the *perception* of education... and too often the perception is the reality...
The easier it is for a different engineer to review your work the better. No matter where you work, someone has 1. approve the design work, 2. be legally responsible (as in getting massively sued). This is why Professional Engineers (PEs) are licensed and have a spiffy little stamp they sign tocertify the design.
So, yeah, write out everything. If you make a few blunders here and there they should be easy to spot.
Edit-Compile-Run-Debug
Edit-Compile-Run-Debug
Programmers would seem to me to be very well conditioned for EQ addiction.
Dealing with minor physical and mental illnesses used to be a hallmark of maturity and professionalism. Nowadays it seems like those of us who just get on with it are punished for our insensitivity, and/or our ability to deal with such matters independently means, _by definition_, that we don't actually, *really* suffer. Because if we did, we would obviously be in the care of "mental health" professionals.
Heh... enjoy it! Probably the last Bud you ever have, with your broken ankle, as your body core temperature drops below 90 F... At least hypothermia is painless... ;)
RC in caves is a ludicrous concept, save for line of sight, but then, why bother. Cave radios work with really looooowwwwww frequencies and require rather large coil antennas to transmit through all that rock. Cave radios are pretty finicky too. This is why cave rescue organizations (good ones) have the ability to lay a mile or two of phone wire in really horrible conditions.
RC butterflies or RC anything-else just ain't gonna happen in a cave.
I asked the dad dude about this once. He harks from the old school: cobol, snobol, etc etc. His answer was that PCs move small amounts of data really fast, a needle thin stream moving at light speed. Mainframes move data like Niagara Falls.
Chiro is interesting for a couple of reasons. The most interesting is that a good chiro will actually _listen_ to you. Every doctor I have seen either for myself or with my wife in the last 10-20 years always seems to be paying attention to something else... maybe their afternoon golf date or whatever. The second interesting thing about chiro is that it only works if the patient is proactive. Can't take a pill, no no no. Gotta actually do the damn exercises. It requires more work than going to the doctor. A third interesting thing about a good chiro is that if you say your back (or whatever) hurts, they might believe you. In contrast to an MD that may say "its all in your head". (As told to my sister-in-law for years until they discovered, too late, that she had a degenerative bone disease. The pain was "all in her head". Chiro of course would not have helped her, but the orthodox medical establishment didn't really help much either.)
:).
Lastly, a really good chiro will have a short list of fine surgeons to refer you to if you have an injury beyond their capability. It helps to have one that has actually had back surgery...
Hope this helps. Don't believe the hype by any means, but for the kind of (very real) back pain I have occasionally, a $60 visit couple of times a year to the chiro is cheap relief.
in the treacherous seas of search is going to be this companies biggest hurdle. Assuming they don't founder in the sinking economy (unless they get bailed out), they had better run a tight ship. Because even if they win, they will likely lose, when the patent holders come to collect license fees.
Internet search space is seriously covered up with multiple overlapping patents on a few relatively simple notions about how to find stuff. It's a lawyers game now, not a technology game.
I did the research: Start with Salton's or Ellen Voorhees work and go forward.
Bottom line: boolean matching works "unreasonably well", clustering is more expensive for only marginally better results. Page ranks (big-ass SVDs) are done periodically, cost is cheap amortized over boolean queries. With clustering, you gotta do more work for every query.
For a small document collection (single pc), and no measurable queries (i.e., single, infrequent user), this might be nice. Then again, if it is a small enough collection to be viable, its probably overkill.
Some of these other techniques look interesting though. Might have to fire up google and investigate.
For groxis though, it doesn't really matter anyway. This entire space is covered by multiple overlapping patents. So even if they win, they lose...
I just spent around 20 hours reviewing a paper for an international engineering journal. I was not paid for this activity. My review was anonymous to the authors of the paper (accept with revisions, if you must know). And this paper was exactly in my field of expertise.
At my current billable rate, that is around $1500-2000 worth of labor, minimum. I do it for free because the authors submitting to such journals are not paid.
If it were a consulting report, you can be damn sure I would charge $75-100/hour to check someones engineering analysis.
Where do you propose to get all of this free labor for reviewing? It's damned hard work.
was raising the bar far higher than anyone ever before imagined.
Before e, wm's were not very interesting.
BSD license is a fine license for many. Others,
such as myself, prefer GPL/LGPL. In fact,
given a choice of proprietary or BSD, I would
pick purely proprietary. I am sure many others
would do the same.
Lawyerclysm imminent!!! News at 11!!!
;> ;)
One semi-hothead spouts off, 13 bazillion
other buffoons jump in.
Lot more talking than thinking going on around
here this morning.
Isn't peer pressure an awesome force?
Situations like this really point out the strength
of the open sources (sores heh heh heh) community.
Would Bill Gates listen to anyone telling him he
was full of buffalo sh*t? Don't think so. So,
BP gets a mouthful of shoe leather. Once he gets
tired of the taste of pavement, he will think
harder before speaking...
The cygwin stuff is not quite there yet. It's close. I would use it a *lot* more of if I could get xterm running. The windows command terminal really really blows, even with bash running. So I see GNU/NT held back by the lack of an xserver + xlibs mostly. Once that was running, I think the rest of it would fall into place pretty fast. Why is this important? Couple of reasons. 1. Some of us are stuck in NT for at least the next few years. 2. It is much easier to train potential Linux users if there were a way to ease the transition from pure gui to not-pure-gui. This way they can see that DOS is a hopeless POS, but a modern shell is extremely useful. And that the only point of similarity between the two is the CLI. I am sure I will get roundly flamed for this, especially over point 1. But, whatever. I got stuff to do, and I have already gotten 1 pure linux box (debian!) and a dual boot box (debian!) in the office. It is just going to take a few years to complete the transition. (Actually, having a couple of NT boxes around *is* useful.).
Last I heard, he had his access pulled.
Then again, I have been usenet free
(and darn proud of it!) for a couple of
years now.
Maybe there is hope yet...
Archimedes Plutonium disappeared.
This is really typical though. Free speech is not much different than free markets. Both depend on just exactly who gets to define the words _free_, and in this case, _speech_. It appears that the judge and plaintiffs get to define words here.
Then again, pre-web, just being on the internet was dues to the Fringe Element of Society Club.
Ah well, just an old codger here, pining for the days of kibology and p-adic insanity.
ciao
Just a nit...
Livermore NL is in Livermore. Berkeley NL is the one in Berkeley.
Actually, since they get to lay a big chunck of this cable under emininent domain, I believe the tax-payers have every right to set some ground rules. What right does ATT or any other utility have to clutter up my skyline with their ugly poles? Seems fair to me. ATT gets to use "public property", that is, easements taken under eminent domain, and maintained at taxpayer expense, for a non-monopoly share of infrastructure running open standards. Why is this so difficult to understand?
If ATT wants to use a private property argument, they should purchase all their own easements.
And so are you!
MS has gored my very own personal ox in more
ways than one. Since I do not work for MS, nor
do any of my friends and family, and my Van Kampen
fund is only 3% into MS, my particular vote,
as a taxpaying US citizen, is to hammer MS hard.
As such, Barbara Boxer will be sure to hear my
side of the story if necessary. She represents
my interests, which, as I have just mentioned,
MS takes little regard for.
Governments grant joint stock companies the
privilege to do business as they see fit, and
protect individual businessmen by and largely
from personal liability. The government gives,
it takes away.
So, friend, our two votes cancel out... but we
aren't the only two voting, so the future remains
interesting!
I bought ustation for linux about 3 years ago.
I don't currently use it because of libc incompatibility which I have not had time ti rectify. This does point out a *major* shortcoming in commercial software: the free software world moves much faster. It just seems ridiculous to me that a recompile from libc5 to libc6 requires me to fork out another $300 for an upgrade.
So, how does it work? Not too bad, even on a p100 with 48M ram. Other than an inordinately long startup time, the program it self runs fine once it gets under. I did notice color map problems in later versions of xfree86. The biggest issue that I have is the postscript export routine makes the assumption that one is running windows. So every line ends with ^M, and there is a binary end-of-file trailer that must be removed. This points out another serious weakness of commercial software. I (rightly) regard the postscript problem as a bug, and could fix it pretty fast if I had the source code. I am sure though that this falls way down on Bentley's priority list, so I am not holding my breath. What would be superb would be to have fig2dev support for dgn->ps, but now I am really dreaming.
Another issue is possible lack of support for graphics tablets. Mouse driven pointers SUCK compared to a 4-button puck on a graphics tablet set to move under absolute coordinates. ustation certainly supports tablets, but I have never seen any linux boxes with a tablet hooked up, or much documentation for doing so.
In summary, good program.
I tend to agree. For about the last six months, the rhetoric emanating from Redmond has taken on an increasingly delusional appearance. That Gates is a genius is not a point of contention. Neither is the empirical fact that many such people end up being bona-fide kooks. Doesn't really matter in the end. Gates is mortal, and the further he reaches the farther he will fall.