I had a look back at the older story, and was puzzled by a few things I saw:
1. The prosecutors allege he made a profit of CAN$7 million. Why is the fine only CAN$1 million?
2. The CBC story says that it will not affect his position in the company, which strikes me as odd if he could spend up to two years in jail...
3. An AC made a remark to the effect that Cowpland still expected to make his sales figures at the time of the stock sale, which suggests that he did not sell his stock to avoid an anticipated capital loss.
If the third remark is true, then the only serious allegation is that he lied to the investigators. I couldn't figure out what misrepresentation he was expected to have made from the allegations. Any enlightenment?
I had a look back at the older story, and was puzzled by a few things I saw:
1. The prosecutors allege he made a profit of CAN$7 million. Why is the fine only CAN$1 million?
2. The CBC story says that it will not affect his position in the company, which strikes me as odd if he could spend up to two years in jail...
3. An AC made a remark to the effect that Cowpland still expected to make his sales figures at the time of the stock sale, which suggests that he did not sell his stock to avoid an anticipated capital loss.
If the third remark is true, then the only serious allegation is that he lied to the investigators. I couldn't figure out what misrepresentation he was expected to have made from the allegations. Any enlightenment?
There are a number of differences between the slashdot model and conventional academic perr review:
1. Usually a journal article gets passed on to just two or three reviewers, whilst the slashdot model exposes the material to a much wider breadth of criticism.
2. The journal's reviewers will work on the article for a period of perhaps several months, and alone, whilst the slashdot feedback is conversational and immediate.
3. Reviewers feedback tends to be collected together on a take-it-or-leave-it basis, whilst in the slahsdot model the criticism is in its turn subject to criticism.
I think we will see big changes in the way academic feedback occurs in the next decades. It is already happening in computer science and mathematics: ideas at an early stage are disseminated in academic mailing lists, getting a quite different kind of feedback before being submitted to classical journal review. Also the era of the preprint has already revolutionised many subjects.
Nothing in the Wired article discussing the merits of the prosecution case (great journalism, Wired). I recall a couple of comments from the previous slashdot story that discussed the deatils of the case. Summary:
1. His alleged crime was to sell stock before announcing worse than expected profits, a case of insider dealing.
2. No-one was sure whether or not he announced his sale, and what period of notice was given.
3. The stock has risen 25% since the disputed sale.
Does anyone have any new information to add to the above? The above doesn't look exactly damning...
Scottie's ``How quaint'' reaction to the mouse in Star Trek IV is another of the rare hacker-friendly moments in movies. Perhaps the only movie moment where the hallowed GUI is not portrayed in reverential tones?
I wonder how `scientific' Maslow's hierarchy really is. Is the idea that our ultimate and most sublime motivation `self-actualisation' really subject to scientific criticism?
(For those not familiar with Abraham Maslow's hierarchy, a few points to add to Morgaine's discussion: he breaks up our needs into five categories, the lowest being self-preservation, then safety, then social needs, then self esteem, and finally `self-actualisation'. He based a theory of education upon it, based upon the idea that we try to get people to listen to their inner feelings. It has been highly influential in modern Management science.)
I was disappointed by Paxman, though. He has the ability to be much more direct, as witnessed by some of his political interviewing.
Well, but Paxman isn't as fearless as everyone on slashdot seems to think: when Matthew Parris made his famous comment about Peter Mandelson being gay, Paxman was aghast, and immediately after the show rang Mandelson to apologise for Parris' outburst.
Sounds as if Brandt didn't receive any positive feedback from his original article. Well it is easier to complain than extoll, but if that is right, it is a shame: I rather liked the original piece, and he seems to be right on target in his analysis of how Sun has come to be the way it is today.
Little factual error: he seems to think before the free *NIXes came along that Sun's offerings had achieved a total dominance amongst UNIXes. Not true: IRIX, SCO and AIX are still quite healthy in their respective user bases.
Personally I think that there is a place for regulation. I would prefer to have separate content providers from service providers wherever the technology makes this feasible.
As a rule when service providers use their leverage to push their content, they manage to displace good, diverse content with poor quality, lowest common denominator crap. It is the case with cable TV, local newspapers, satellite TV (in Europe & Asia) and I have little doubt that it would be the same with cable-ISPs.
Of course I don't. I think my ironic tone doesn't translate into ASCII all that well...
Carmack's coments on development environments
on
John Carmack Answers
·
· Score: 2
Thought provoking comments on the differences between the development envirnoments he has used, and it is nice to see that he thinks that Linux is getting there: the games development envirnoment is probably the furthest away from Linux's natural strengths.
I'd be interested to see what Carmack thinks of tools like VMware: does he see it as bing useful to have NT and Linux on the same machine? Would it make his life easier?
Juln writes: Lets see... he bought DOS from someone else, had the personal computer market handed to him on a platter through his mother sitting on a bank board with the then-chairman of IBM, and has since then been one of the biggest thieves of intellectual property in the industry.
Well, Gates didn't lie: these moves *did* revolutionise the industry...
One would probably translate it by `My hat, which has three corners'. The point is that this policy of directly translating `der' by `which' doesn't work, since the noun referred to by which may be the wrong one, since English lacks the gender distinctions of German.
I concluded that the metalanguage must have a phrase structure that generalises any particular language one wants to embed in it, which of course means that the metalanguage contains expressions that can't be directly represented in the language one wants to translate into.
The solution to this difficulty is to have semantics-preserving transformation rules in the matalanguage that allow one to manipulate the expression into something that does correspond to something in the target language. Hard work, but I think it could be made automatic.
With respect to your other point, I think the `cultural baggage' is a matter of semantics and not syntax. The problem these machines are expected to solve doesn't require human intelligence, and the nice thing is the single machine could in principle be able to translate into a hundred different languages, something even the very cleverest humans find difficult...
Lastly, it is probably easier to get people to agree on a common metalanguage, than to get them to learn a common language...
One might have said the same thing about IBM fifteen years ago...
It really comes down to what suceeds in the market. Microsoft charges an awful lot more for its OS than 3Com does (I saw $30 vs 50 cents here on/. not too long ago), and Microsoft can't afford to break into *every* market the way it did with IE.
(1) Not every language has every tense. German has fewer tenses than english, and another poster said that Chinese has none.
(2) Language can't be described in A BNF grammar: it isn't sophisticated enough to capture singular vs. plural, gender, case, verb declensions etc. Phrase structure grammar extend BNF grammar s with parameters to capture these, and Chomsky showed that these are sufficnet to capture all of natural language.
I would guess that the meta-language design is based upon transformational grammar, which exposes the essential similarities between sentences like `The door is closed' and `Close the door!'. This would allow it to express subtleties like different ways of representing the same sentence.
(1) Translating into the metalanguage is in general ambiguous. However language is equipped with lots of hints to disambiguate just these cases, and so for the cases where language is put to use, it should be possible to apply heuristics.
(2) German also has much less tense structure than English. The passage of time is indicated using modal particles, and the metalanguage will need to possess transformation rules for switching between tensed and modal representations.
I gather from a linguist friend that systems such as these exist already. The exciting thing about this proposal is that it covers such a wide variety of languages and carries the stamp of the UN.
Let me just add something to the above, since I haven't made myself clear in what I have said in the above.
In German it is possible to use the definite article to refer back to something used in the previous sentence, rather like `it' in English: but with the crucial distinction that what we refer back to must be of matching gender. So if a masculine, feminine and neuter word occur in the sentence it is possible to refer to any of them with the `it'. This ability to refer on the basis of gender must be captured in our syntactic model. Similarly the case system allows one to have multiple indirect objects (one accusative, one datave and one genetive, for example) directly attached to a verb, where in english one would use a preposition.
Esperanto wouldn't work: the point about Esperanto is that it is a pared down language without the anomalies that most natural languages have.
Unfortunately one *needs* these anomalies in order to translate back into the natural languages: one needs to know all about the 3 genders and 4 cases to translate automatically into German, for example.
Each dialect has its own distinct syntax: naturally the project will pick just one for each language.
Linguists are pretty good at providing models of informal language use. People are quite sloppy in observing formal linguistic rules, but even these deviations tend to follow their own rules.
I think the metalanguage would need to carry all of the specific case/gender/tense information for each of the languages that can be embedded in it.
It sounds possible to me: phrase structure grammars can express the syntactic structure of all natural languages (Noam Chomsky's result). The meta language will be very complex though...
I am interested to see how they will actually get this article written. Will it be a survey-style article organised by topic (eg. section of the availabe technologies, subsection crypotography discussing the balance of powers between white hats and black hats)?
It might not be a bad idea to put together a web resource of of the quality information available on electronic terrorism and countermeasures.
Did the US pull out of Vietnam because it was bankrupt? Economics matters, but the ability of impoverished nations to fight on should not be underestimated. The objective of war remains to incapacitate your enemy's fighting forces...
I doubt taking the NYSE offline for two days would do any harm at all to the US's *military* effectiveness.
Slashdot Mirror
1. The prosecutors allege he made a profit of CAN$7 million. Why is the fine only CAN$1 million?
2. The CBC story says that it will not affect his position in the company, which strikes me as odd if he could spend up to two years in jail...
3. An AC made a remark to the effect that Cowpland still expected to make his sales figures at the time of the stock sale, which suggests that he did not sell his stock to avoid an anticipated capital loss.
If the third remark is true, then the only serious allegation is that he lied to the investigators. I couldn't figure out what misrepresentation he was expected to have made from the allegations. Any enlightenment?
1. The prosecutors allege he made a profit of CAN$7 million. Why is the fine only CAN$1 million?
2. The CBC story says that it will not affect his position in the company, which strikes me as odd if he could spend up to two years in jail...
3. An AC made a remark to the effect that Cowpland still expected to make his sales figures at the time of the stock sale, which suggests that he did not sell his stock to avoid an anticipated capital loss.
If the third remark is true, then the only serious allegation is that he lied to the investigators. I couldn't figure out what misrepresentation he was expected to have made from the allegations. Any enlightenment?
1. Usually a journal article gets passed on to just two or three reviewers, whilst the slashdot model exposes the material to a much wider breadth of criticism.
2. The journal's reviewers will work on the article for a period of perhaps several months, and alone, whilst the slashdot feedback is conversational and immediate.
3. Reviewers feedback tends to be collected together on a take-it-or-leave-it basis, whilst in the slahsdot model the criticism is in its turn subject to criticism.
I think we will see big changes in the way academic feedback occurs in the next decades. It is already happening in computer science and mathematics: ideas at an early stage are disseminated in academic mailing lists, getting a quite different kind of feedback before being submitted to classical journal review. Also the era of the preprint has already revolutionised many subjects.
1. His alleged crime was to sell stock before announcing worse than expected profits, a case of insider dealing.
2. No-one was sure whether or not he announced his sale, and what period of notice was given.
3. The stock has risen 25% since the disputed sale.
Does anyone have any new information to add to the above? The above doesn't look exactly damning...
Scottie's ``How quaint'' reaction to the mouse in Star Trek IV is another of the rare hacker-friendly moments in movies. Perhaps the only movie moment where the hallowed GUI is not portrayed in reverential tones?
(For those not familiar with Abraham Maslow's hierarchy, a few points to add to Morgaine's discussion: he breaks up our needs into five categories, the lowest being self-preservation, then safety, then social needs, then self esteem, and finally `self-actualisation'. He based a theory of education upon it, based upon the idea that we try to get people to listen to their inner feelings. It has been highly influential in modern Management science.)
more direct, as witnessed by some of his political interviewing.
Well, but Paxman isn't as fearless as everyone on slashdot seems to think: when Matthew Parris made his famous comment about Peter Mandelson being gay, Paxman was aghast, and immediately after the show rang Mandelson to apologise for Parris' outburst.
Little factual error: he seems to think before the free *NIXes came along that Sun's offerings had achieved a total dominance amongst UNIXes. Not true: IRIX, SCO and AIX are still quite healthy in their respective user bases.
As a rule when service providers use their leverage to push their content, they manage to displace good, diverse content with poor quality, lowest common denominator crap. It is the case with cable TV, local newspapers, satellite TV (in Europe & Asia) and I have little doubt that it would be the same with cable-ISPs.
Of course I don't. I think my ironic tone doesn't translate into ASCII all that well...
I'd be interested to see what Carmack thinks of tools like VMware: does he see it as bing useful to have NT and Linux on the same machine? Would it make his life easier?
Lets see... he bought DOS from someone else, had the
personal computer market handed to him on a platter through his mother
sitting on a bank board with the then-chairman of IBM, and has since
then been one of the biggest thieves of intellectual property in the
industry.
Well, Gates didn't lie: these moves *did* revolutionise the industry...
One would probably translate it by `My hat, which has three corners'. The point is that this policy of directly translating `der' by `which' doesn't work, since the noun referred to by which may be the wrong one, since English lacks the gender distinctions of German.
I concluded that the metalanguage must have a phrase structure that generalises any particular language one wants to embed in it, which of course means that the metalanguage contains expressions that can't be directly represented in the language one wants to translate into.
The solution to this difficulty is to have semantics-preserving transformation rules in the matalanguage that allow one to manipulate the expression into something that does correspond to something in the target language. Hard work, but I think it could be made automatic.
With respect to your other point, I think the `cultural baggage' is a matter of semantics and not syntax. The problem these machines are expected to solve doesn't require human intelligence, and the nice thing is the single machine could in principle be able to translate into a hundred different languages, something even the very cleverest humans find difficult...
Lastly, it is probably easier to get people to agree on a common metalanguage, than to get them to learn a common language...
One might have said the same thing about IBM fifteen years ago...
/. not too long ago), and Microsoft can't afford to break into *every* market the way it did with IE.
It really comes down to what suceeds in the market. Microsoft charges an awful lot more for its OS than 3Com does (I saw $30 vs 50 cents here on
Two mistakes in the above:
(1) Not every language has every tense. German has fewer tenses than english, and another poster said that Chinese has none.
(2) Language can't be described in A BNF grammar: it isn't sophisticated enough to capture singular vs. plural, gender, case, verb declensions etc. Phrase structure grammar extend BNF grammar s with parameters to capture these, and Chomsky showed that these are sufficnet to capture all of natural language.
I would guess that the meta-language design is based upon transformational grammar, which exposes the essential similarities between sentences like `The door is closed' and `Close the door!'. This would allow it to express subtleties like different ways of representing the same sentence.
Two points:
(1) Translating into the metalanguage is in general ambiguous. However language is equipped with lots of hints to disambiguate just these cases, and so for the cases where language is put to use, it should be possible to apply heuristics.
(2) German also has much less tense structure than English. The passage of time is indicated using modal particles, and the metalanguage will need to possess transformation rules for switching between tensed and modal representations.
I gather from a linguist friend that systems such as these exist already. The exciting thing about this proposal is that it covers such a wide variety of languages and carries the stamp of the UN.
Let me just add something to the above, since I haven't made myself clear in what I have said in the above.
In German it is possible to use the definite article to refer back to something used in the previous sentence, rather like `it' in English: but with the crucial distinction that what we refer back to must be of matching gender. So if a masculine, feminine and neuter word occur in the sentence it is possible to refer to any of them with the `it'. This ability to refer on the basis of gender must be captured in our syntactic model. Similarly the case system allows one to have multiple indirect objects (one accusative, one datave and one genetive, for example) directly attached to a verb, where in english one would use a preposition.
Esperanto wouldn't work: the point about Esperanto is that it is a pared down language without the anomalies that most natural languages have.
Unfortunately one *needs* these anomalies in order to translate back into the natural languages: one needs to know all about the 3 genders and 4 cases to translate automatically into German, for example.
Each dialect has its own distinct syntax: naturally the project will pick just one for each language.
Linguists are pretty good at providing models of informal language use. People are quite sloppy in observing formal linguistic rules, but even these deviations tend to follow their own rules.
I think the metalanguage would need to carry all of the specific case/gender/tense information for each of the languages that can be embedded in it.
It sounds possible to me: phrase structure grammars can express the syntactic structure of all natural languages (Noam Chomsky's result). The meta language will be very complex though...
No mathematics prize either, and I believe that subject existed then...
The film adaptation of his most famous novel `The Tin Drum' was outlawed in Oklahoma until last year as child pornography.
Hope the legislators are embarassed...
I am interested to see how they will actually get this article written. Will it be a survey-style article organised by topic (eg. section of the availabe technologies, subsection crypotography discussing the balance of powers between white hats and black hats)?
It might not be a bad idea to put together a web resource of of the quality information available on electronic terrorism and countermeasures.
... is a futile measurement: I saw a `Programming Pearls' column in an old CACM that had the best illustration.
:= f(1);
The author of the article (a software consultant) found a piece of code of the following form:
a1
a2 : =f(2);
etc for 500 lines. Needless to say the (ir)responsible programmer was paid per line...
Did the US pull out of Vietnam because it was bankrupt? Economics matters, but the ability of impoverished nations to fight on should not be underestimated. The objective of war remains to incapacitate your enemy's fighting forces...
I doubt taking the NYSE offline for two days would do any harm at all to the US's *military* effectiveness.