'Just guessing here, but...Mr. Rappoports GPA is probably about 0.6 and he'll be text-messaging his friends during his shift at MacDonald's this time next year...'
Paul Graham kicked off a flood of mail filters implementing Bayesian filtering with his
"A Plan for Spam" article in August 2002, though it was far from a new concept. In fact, ifile has used a Naive Bayes classification algorithm since August 1996 to automatically file mail into folders. In academic circles, Bayesian methods have been used in text classification for many years, and for spam detection prior to Graham, as evidenced by the 1998 workshop paper A Bayesian Approach to Filtering Junk E-Mail by Sahami, et al.
I, myself, remember discussion of AI versus Bayesian versus fuzzy set etc. methods being compared for text classification and search in the 1980s. Here, for example, is the announcement of a presentation in 1990 by James Coombs to Brown Computing in the Humanities Users' Group which includes Bayesian classification.
Well, maybe. For example, the DX costs would have included loss due to manufacturing flaws; that is, the SX cost was already factored in and covered. If the cost of additional handling of the SXes was commensurable with the discounted price, then there is no real production cost to SXes as it is already covered by the DX production. (I think that makes sense;^)
As well, they quite possibly SXed whole wafers or lots of DXes using SPC (statistical process control) methods (at least later in the chip's lifecycle), additionally reducing FP testing costs.
'Of course, we can also think of the 486SX which was a copro impaired version of the genuine DX...'
The impairment was not deliberate, however. The SX models were simply a means to sell DXs in which the FP functions were not working properly due to glitches during manufacture.
'Yeah, because doing something for the betterment of society without wanting to get rich off of it is just un-American...'
It's especially humorous (in a sad way) as one of Alexis de Tocqueville's (the French author, not the intitution) main, favorable points about the US was the rich fabric of volunteerism, community spirit, and neighbourliness.
Yes, but 'behavioural analysis' was not needed to find it as MS was quite upfront in acknowledging the copyright holders in the XP documentation. It is a minor point in ESR's response, but on these do credibility hang as the overall important message gets lost in arguments of the details.
Yes, I think he should have taken a deep breath and counted to ten before replying:^) For example:
Proprietary outfits lift code from elsewhere all the time. It is known from behavioral analysis of the Microsoft TCP/IP stack, for example, that they swiped their code from BSD. So there may well be be immense amounts of stolen IP in proprietary code, hidden by commercial secrecy.
Given that the BSD TCP/IP stack was released under the BSD licence and that MS provides the proper acknowledgements in their documents, this is a red herring. Thus, ESR's concluding sentence does not follow, although it may be correct for other reasons.
'I don't think so, Mr. Crosby! Cuts out all middlemen? The RIAA are still there taking their fat chunk. The artists get a tiny chunk. Of course, if you're smart enough to release tracks directly to Apple...but most RIAA-promoted artists can't do this.'
While you are correct, direct to internet is very much still in its infancy. As more artists, especially those who aren't cute, move to this, the RIAA (affiliated record companies) will become less and less important. The hard part will be identifying and finding the musicians which interest you! That is: a good indexing system is required.
'Latin America overheats and goes into a draught.'
I, too, enjoy a draught when I overheat; my favourites being Guinness or my own home brew. I am sad, however, when there is a drought and I cannot enjoy same;^)
'Yeah. It's not like I'm paying your money or anything, so why in the hell would you want to do what I say?'
Well, one reason is that your name is being associated with a very visible work. If it is unpleasant to look at or use, then it will be a detriment to your future employment. Some contracts are not worth taking because they will have a negative impact. Your job should be to provide what the client needs and get him to appreciate it (the harder part).
'The arrogance of programmers is only exceeded by their snottiness.'
This, alas, is also too often true; especially when dealing with user interfaces.
Just a joke, honest!. Actually, any of the major distributions will almost certainly do you just fine. Each have their ups and downs. My very first was slackware, about 10 years ago, but I wouldn't suggest it to a new user today unless you are very comfortable with command line configuration and post-installation setup. I still use it because I like it.
Oh! I didn't take offence to your response. The 'I think' remark is in reference to the bad taste in my mouth that I got after reading the page you indicated and the other response to my question. My thanks was honest.
What retard came up with this piece of crap? Montaigne must be turning over in his grave! (I've been 35 years out of high school; I'm glad to see that low standards have gotten even lower.)
'I had 5 paragraphs - 1 intro, 3 body - 1 half-assed conclusion.'
I've seen several references to 'five paragraphs' in the various comments. Would someone explain what this means and, more importantly, what five paragraphs has to do with writing an essay?
SPARK only covers programme verification. There are other formal methods which cover specification (e.g. Z) as well as various theorem provers which can link all these parts together to ensure no contradictions, etc.
A search on Google for formal methods will give you a lot of stuff. The first site that comes up is a good starting point.
Note that at some point, one has to hope that what the client wants is what he has described. A tax calculation programme will not be of use if he really wanted a customer relations management system:^)
Have you thought about getting a mathematician/numerical analyst (e.g. grad student) to take a look at your system? Brute forcing singular matrices usually doesn't scale well. Being able to identify the critical points would probably give you more bang for your euro.
This would be better fixed by using a base 12 number system; then it would allow both the most common practical divisions (2, 3, 4) and ease of scaling by shifting the duodecimal point! That is: metric and Imperial in 1!
Slashdot Mirror
And that's why I buy undyed detergents for my hunting clothes, as the dye highlights you in the eyes of deer, moose, etc.
(I know, totally OT)
Well, at least he could get free ring tunes
Er, was this before or after eating it?
Ronald MacDonald Sings Britney Spears' Greatest Hits!
I don't need a TV; I have a front loading washing machine;^) Now, if they could get that on my wrist, I'd be interested!
IANAL, but isn't that a proof of "prior usage" and makes the patent invalid?'
And even earlier, as freshmeat points out:
I, myself, remember discussion of AI versus Bayesian versus fuzzy set etc. methods being compared for text classification and search in the 1980s. Here, for example, is the announcement of a presentation in 1990 by James Coombs to Brown Computing in the Humanities Users' Group which includes Bayesian classification.
Well, maybe. For example, the DX costs would have included loss due to manufacturing flaws; that is, the SX cost was already factored in and covered. If the cost of additional handling of the SXes was commensurable with the discounted price, then there is no real production cost to SXes as it is already covered by the DX production. (I think that makes sense ;^)
As well, they quite possibly SXed whole wafers or lots of DXes using SPC (statistical process control) methods (at least later in the chip's lifecycle), additionally reducing FP testing costs.
The impairment was not deliberate, however. The SX models were simply a means to sell DXs in which the FP functions were not working properly due to glitches during manufacture.
It's especially humorous (in a sad way) as one of Alexis de Tocqueville's (the French author, not the intitution) main, favorable points about the US was the rich fabric of volunteerism, community spirit, and neighbourliness.
I believe so too, but I have only looked at the XP docs (I mostly use Linux, FreeBSD, and (when I can get work on it, VMS :^)
Yes, but 'behavioural analysis' was not needed to find it as MS was quite upfront in acknowledging the copyright holders in the XP documentation.
It is a minor point in ESR's response, but on these do credibility hang as the overall important message gets lost in arguments of the details.
Given that the BSD TCP/IP stack was released under the BSD licence and that MS provides the proper acknowledgements in their documents, this is a red herring. Thus, ESR's concluding sentence does not follow, although it may be correct for other reasons.
So you mean that the Bush administration officials are taking action against themselves as they are responsible?
While you are correct, direct to internet is very much still in its infancy. As more artists, especially those who aren't cute, move to this, the RIAA (affiliated record companies) will become less and less important. The hard part will be identifying and finding the musicians which interest you! That is: a good indexing system is required.
I, too, enjoy a draught when I overheat; my favourites being Guinness or my own home brew. ;^)
I am sad, however, when there is a drought and I cannot enjoy same
Well, one reason is that your name is being associated with a very visible work. If it is unpleasant to look at or use, then it will be a detriment to your future employment. Some contracts are not worth taking because they will have a negative impact. Your job should be to provide what the client needs and get him to appreciate it (the harder part).
'The arrogance of programmers is only exceeded by their snottiness.'
This, alas, is also too often true; especially when dealing with user interfaces.
Just a joke, honest!. Actually, any of the major distributions will almost certainly do you just fine. Each have their ups and downs. My very first was slackware, about 10 years ago, but I wouldn't suggest it to a new user today unless you are very comfortable with command line configuration and post-installation setup. I still use it because I like it.
Oh! I didn't take offence to your response. The 'I think' remark is in reference to the bad taste in my mouth that I got after reading the page you indicated and the other response to my question. My thanks was honest.
Sorry, that would violate the tenets of XP;^)
What retard came up with this piece of crap? Montaigne must be turning over in his grave!
(I've been 35 years out of high school; I'm glad to see that low standards have gotten even lower.)
I've seen several references to 'five paragraphs' in the various comments. Would someone explain what this means and, more importantly, what five paragraphs has to do with writing an essay?
A search on Google for formal methods will give you a lot of stuff. The first site that comes up is a good starting point.
Note that at some point, one has to hope that what the client wants is what he has described. A tax calculation programme will not be of use if he really wanted a customer relations management system :^)
It uses acl2, a lisp language based prover.
Have you thought about getting a mathematician/numerical analyst (e.g. grad student) to take a look at your system? Brute forcing singular matrices usually doesn't scale well. Being able to identify the critical points would probably give you more bang for your euro.
This would be better fixed by using a base 12 number system; then it would allow both the most common practical divisions (2, 3, 4) and ease of scaling by shifting the duodecimal point!
That is: metric and Imperial in 1!