Interesting links, I've never actually found that Adobe book freely available, maybe I'll print it out tomorrow for a reference.
Great that you have not had any trouble with the language, yes, it does not take much more than a couple of hours to make the simplest diagrams, and then it's just fun to make more complex one (I've enjoyed coding a recursive Batcher-banyan network wiring pattern in PostScript quite alot).
I'm still a bit worried about your plans to convert eps to pdf, are you planning to just attach your diagrams as individual pages at the end of the book? If you are planning to have diagrams + captions + text on a single page, you are better off leaving them as eps files, inserting them into LaTeX figures as eps, then running the resulting dvi though dvi2ps/pdf.
Actually, XFig is not scriptable at all, but its file format is easy to "reverse-engineer" (maybe easier than reading source, despite its availability;-) ). But while in XFig you'd put "box 1 100 10 10 foo bar blah, blah", it is as easy to make PostScript statement "blah, blah 1 100 10 10 box".
Generating PS is indeed my tool of choice for quick-and-dirty visualization from within my programs, and it IS easy compared to the rest of the things one might want to do. In any case, trying would not hurt anyone.
When I was talking about PS figures, I assumed you are going to embed them in your LaTeX document. The way which works for me is first to run 'ps2epsi' (available with GhostScript) to convert them to EPS, then put within LaTeX's epsfig blocks, then 'latex', 'dvips' (from LaTeX install), 'ps2pdf' (also from GhostScript install).
I have not tried outher routes recommended here, like latex2pdf, or whatever, one which would give you hypertext links in PDF from your TOC, those might be useful as well.
If you intend to generate many diagrams using "custom scripts", you might consider making your scripts either write PostScript directly or even write scripts themselves in PostScript (the language) and have your diagram and its source be one and the same. Actually learning PostScript is not that hard (after one gets past its RPN/stack nature), you code a set of macros you intend to use (or you can draw their images in, say, Xfig, export to PS and copy/paste into your PostScript program as function), then in your code you just instantiate them where you need them, connect with lines, etc.
Adobe actually provides pretty good documentation on PostScript (this explains the popularity of the format), and GhostScript, of course, does great job in interpreting the language.
A good way to get a feeling for the language, try looking through PostScript code generated by any of cad/figure drawing tools, Xfig itself, gnuplot, etc. If nothing else, you'd learn how to make quick and dirty mods to.ps file itself, like making all lines thicker or getting rid of color.;-)
Just for fun, have a look: http://www.pvv.ntnu.no/~andersr/fractal/PostScript.html (actually I tried to find a very cool someone's 4-line.signature which generated a nice fractal tree in PostScript, saw it, like, 10 years ago on Usenet, but could not find right now.)
I am not sure about "deregulation", but in its pure form the free market approach (with the half-alive, half-rotten bodies of almost or completely failing/failed companies, paid for by investors, not general public) might've prevented it. It seems (as of now) is that all that was needed to deal with the current situation was a bit of extra capacity in the power grid, and it was not there partially because not enough companies have tried to compete on this market. It's silly that one can buy (wholesale) an entire telecom satellite fleet, but a thick copper cable (;-) ) is such a rare commodity today... Gets one thinking about the possibilities of the energy companies having just a bit too much of goverment-sponsored monopolies, not too little...
Working scale models were good fot the old days when mostly mechanical devices were patented. But now the majority of NON-TRIVIAL patentable ideas require large (if not huge) capital investment to make even a proof-of-concept breadboard thing. And VCs who invest money like to see granted patent(s) before they invest...
If anything, this requirement would shift balance towards obvious stuff. Think how hard would it be for anyone to code a mock-up of a "one-click" patent!;-)
I did not RTFA;-), but I see where 4,096 comes from -- Windows does not use 2-letter 'salt' value! A simple dictionary attack is possible if you encrypt all possible alphanum passwords either (getting 2 CDs worth of data) and compare them with a given password. In more complex case, when sailt is used, you have to encrypt all posssible passwords with all 4,096 possible salt values, thus getting 4,096 times more data. I suspect that 13 seconds is time to scan through all the data (with some of it in cache, some still on disk), if you have more data you have to spend either 4,096 or (for "smarter";-) algorithm) log_2(4,096)=12 times longer.
... after all, syntactically VHDL is an Ada derivative.
It would be an interesting thing for them to expand into an open-source VHDL design tool.
Remember, now for a $100-$200 one can buy an FPGA evaluation board from, say, Xilinx, they would give you a (closed-source, I guess) compiler to compile from VHDL to an FPGA bin file, you load that into the board through a parport and here it is, fully custom electronic gadget!
However all the big engineering apps, CRM apps,
and databases like Oracle only exist in x86 Linux or AIX.
I think for many applications if will be relatively easy for IBM to convince s/w vendors to port to this box. I've read somewhere a rumor that IBM some time ago just said to Cadence (the main EDA player, and IBM buys A LOT of licenses from them) not to release the next version of their suite for Suns unless Linux/x86 version is released simultaneusly -- here it goes, ic5.00-linux!
Seeing IBM to strong-arm or sweet-talk MS into releasing office on PPC would be a bit too far-fetched, but, who knows, maybe MS would want to expand beyond Intel/AMD...;-)
Paul B.
Maybe because in different situations different threshold would have to be applied. E.g., if it is a terrorist monitoring camera on a random street corner, it might not be feasible to unleash FBI agents after every guy who matched at 80%, but if that random street corner happens to be in Washington, DC across the street from the White House, 80% confidence might be a reason to trigger further actions.
And if it is a camera in the cash machine and you claim that you are Joe and want to get your $500, you better match Joe's face at, say, 99% (it can also ask you to turn a bit and face the lens if your score is lower than some threshold.
Another example, if an airport screener can realistically check 10 people out of a hundred, she chooses ones with the highest scores. Yes, it might mean that John Doe in your example will be checked, and Osama will be not, IF there are other 9 people in line with scores >=95.
Algorithms used might be the same, but exact policy is implemented by taking scores into account.
There is more than binary yes/no in this world...
Paul B.
P.S. Not that I know anything about the actual numbers or policies, but I can see the value of having the scores available to people who program the machine, but not necessarily to the screeners (if any) who operate them.
Check THIS out: http://www.mext.go.jp/english/org/science/im age/35 _01.jpg (it is a three story building, and I guess it also has its own powerplant).
For more pretty pix, of course: http://images.google.com/images?q=Earth+S imulator& ie=UTF-8&oe=UTF-8&hl=en
...or because Bill Clinton smoked weed once. Nobody is defined in any single instant.
It's not him smoking weed once, it's him maintaining the whole "War on Drugs" policy AND making non-violent people serve time in jail for what he himself did... Same with the whole "political correctness" thing, I'd never mind him fscking all the subordinate girls silly, but what right did he have to support denying the same "rights";-) to people less fortunate in life?
Proud to be a Libertarian in heart (though I can not vote for them, not a citizen of the US of A.)
It is not like he is actually trying to pay someone's living, he just wants to promote his game and make it a successful project! (If it were Leisure Suit Larry, I'd volunteer myself!;-) )
And $100/week (or even $10/week) sounds more like a token of appreciation that any kind of financial reward (at least here, in the US. I bet there are many programmers, say, in Russia (my home country) who would not mind playing with the code AND making $100/week!;-) ).
Look at it this way, if you'd get a check for $10 signed by Linus for your infinitely wise contribution to you know what, would you cash it or would you frame it and hang on the wall?;-) The latter might get you some extra geek-points (provided you care about those) and, maybe, much more in return when someone with bigger $$$ notices it on your wall...;-) Just a thought...
LHe is 4.2 K, I am not sure about LH, I know some people use solid H to cool things (I guess, Chandra space telescope was originally cooled by a big block of solid Hidrogen, then they installed a cryocooler on that).
And OF COURSE there is a whole bunch of metallic superconductors at 4.2 K, lead was most popular in 80s (IBM's superconducting supercomputer project), niobium is matherial of choice now. You can deposit/etch it using almost the same litho tools and processes as CMOS people use.
Ceramic high-Tc superconductors are harder to make, and they are less useful "for anything but straight wires", but they work at LN temperature (77K), thus the interest in developing those.
Yeah, reminds me of a story which happened in our cryoelectronics lab, back in Moscow State University (Russia). OK, some guys bought frozen dumplings ("pelmeni", for those who know...;-) ), and were not sure if they will not thaw by the time they get home, so they just poured a liter or two of LN over. Well, the next thing we've heard about those was half a week (!) later when the dumplings finally thawed enought to be separated and cooked... (pre-microwave oven days;-) )
All the code involved is GPL'd, and the data are public: everybody can grok PageRank now!
GPL'd? Hmm, I thought that Google did patent the PageRank algorithm (correct me if I am wrong), so re-implementing THEIR algorithm even more efficiently would be incompatible with GPL. OTOH, if it is not THEIR algorithm, it can not be called 'PageRank'
Oh, the evils of software patents...
Paul B.
Hmm, you will have a GRADIENT of the index, but hardly a BOUNDARY (almost by definition, BOUNDARY is where index is not continuous and gradient can be considered to be +-Inf; OTOH, the "continuous boundary", if such a beast exists, is nothing but a gradient).
Not that I was arguing directly with the point you were making (I do not know if they had to compensate for this effect or not), but one can definitely imagine a physical effect when you have no sharp boundary between two matherials, but some interesting optical phenomena.
I think there were even lenses made out of FLAT "gradiated" glass.
Given proper heat/pressure/surface combination you CAN get different refractive indices (though no boundary between them) -- think of mirages http://images.google.com/images?q=mirage+ desert&hl =en&lr=&ie=UTF-8&oe=UTF-8
I remember my (then-girl-)friend saying, on the very OTHER side of Atlantic: "So, you claim you do not like Fantasy? Well, read this book..." --- and the next thing I knew was me trying to bend the reality of a neighbourhood street-corner cart to carry the beer that I liked...;-)
Slashdot Mirror
Parse error. That's an interesting use of the word average. If you are an (exactly) average driver, exactly half of the people drive worse than you.
;-)
Nope, you have defined a median drived, not an average driver!
I guess I'm being a smarterer-ass!
Paul B.
Interesting links, I've never actually found that Adobe book freely available, maybe I'll print it out tomorrow for a reference.
Great that you have not had any trouble with the language, yes, it does not take much more than a couple of hours to make the simplest diagrams, and then it's just fun to make more complex one (I've enjoyed coding a recursive Batcher-banyan network wiring pattern in PostScript quite alot).
I'm still a bit worried about your plans to convert eps to pdf, are you planning to just attach your diagrams as individual pages at the end of the book? If you are planning to have diagrams + captions + text on a single page, you are better off leaving them as eps files, inserting them into LaTeX figures as eps, then running the resulting dvi though dvi2ps/pdf.
Hope this helps.
Paul B.
Feels good to have your opinion appreciated! ;-)
;-) ). But while in XFig you'd put "box 1 100 10 10 foo bar blah, blah", it is as easy to make PostScript statement "blah, blah 1 100 10 10 box".
Some points:
Actually, XFig is not scriptable at all, but its file format is easy to "reverse-engineer" (maybe easier than reading source, despite its availability
Generating PS is indeed my tool of choice for quick-and-dirty visualization from within my programs, and it IS easy compared to the rest of the things one might want to do. In any case, trying would not hurt anyone.
When I was talking about PS figures, I assumed you are going to embed them in your LaTeX document. The way which works for me is first to run 'ps2epsi' (available with GhostScript) to convert them to EPS, then put within LaTeX's epsfig blocks, then 'latex', 'dvips' (from LaTeX install), 'ps2pdf' (also from GhostScript install).
I have not tried outher routes recommended here, like latex2pdf, or whatever, one which would give you hypertext links in PDF from your TOC, those might be useful as well.
Paul B.
If you intend to generate many diagrams using "custom scripts", you might consider making your scripts either write PostScript directly or even write scripts themselves in PostScript (the language) and have your diagram and its source be one and the same. Actually learning PostScript is not that hard (after one gets past its RPN/stack nature), you code a set of macros you intend to use (or you can draw their images in, say, Xfig, export to PS and copy/paste into your PostScript program as function), then in your code you just instantiate them where you need them, connect with lines, etc.
.ps file itself, like making all lines thicker or getting rid of color. ;-)
t .html .signature which generated a nice fractal tree in PostScript, saw it, like, 10 years ago on Usenet, but could not find right now.)
Adobe actually provides pretty good documentation on PostScript (this explains the popularity of the format), and GhostScript, of course, does great job in interpreting the language.
A good way to get a feeling for the language, try looking through PostScript code generated by any of cad/figure drawing tools, Xfig itself, gnuplot, etc. If nothing else, you'd learn how to make quick and dirty mods to
Just for fun, have a look: http://www.pvv.ntnu.no/~andersr/fractal/PostScrip
(actually I tried to find a very cool someone's 4-line
Hope this helps.
Paul B.
As in this guy:
m l
Former VP of International Business at SuSE Joins SCO As VP of SCOsource in Europe
http://biz.yahoo.com/prnews/030819/latu060_1.ht
Paul B.
I am not sure about "deregulation", but in its pure form the free market approach (with the half-alive, half-rotten bodies of almost or completely failing/failed companies, paid for by investors, not general public) might've prevented it. It seems (as of now) is that all that was needed to deal with the current situation was a bit of extra capacity in the power grid, and it was not there partially because not enough companies have tried to compete on this market. It's silly that one can buy (wholesale) an entire telecom satellite fleet, but a thick copper cable ( ;-) ) is such a rare commodity today... Gets one thinking about the possibilities of the energy companies having just a bit too much of goverment-sponsored monopolies, not too little...
Just my $0.02 (or 1/4 of an kWt/hour)
Paul B.
Very good point, really.. Where are my modpints when I want to use them!
Working scale models were good fot the old days when mostly mechanical devices were patented. But now the majority of NON-TRIVIAL patentable ideas require large (if not huge) capital investment to make even a proof-of-concept breadboard thing. And VCs who invest money like to see granted patent(s) before they invest...
;-)
If anything, this requirement would shift balance towards obvious stuff. Think how hard would it be for anyone to code a mock-up of a "one-click" patent!
Paul B.
I did not RTFA ;-), but I see where 4,096 comes from -- Windows does not use 2-letter 'salt' value! A simple dictionary attack is possible if you encrypt all possible alphanum passwords either (getting 2 CDs worth of data) and compare them with a given password. In more complex case, when sailt is used, you have to encrypt all posssible passwords with all 4,096 possible salt values, thus getting 4,096 times more data. I suspect that 13 seconds is time to scan through all the data (with some of it in cache, some still on disk), if you have more data you have to spend either 4,096 or (for "smarter" ;-) algorithm) log_2(4,096)=12 times longer.
Paul B.
... after all, syntactically VHDL is an Ada derivative.
It would be an interesting thing for them to
expand into an open-source VHDL design tool.
Remember, now for a $100-$200 one can buy an FPGA
evaluation board from, say, Xilinx, they would
give you a (closed-source, I guess) compiler to
compile from VHDL to an FPGA bin file, you load
that into the board through a parport and here it
is, fully custom electronic gadget!
Paul B.
However all the big engineering apps, CRM apps, and databases like Oracle only exist in x86 Linux or AIX. ;-)
I think for many applications if will be relatively easy for IBM to convince s/w vendors to port to this box. I've read somewhere a rumor that IBM some time ago just said to Cadence (the main EDA player, and IBM buys A LOT of licenses from them) not to release the next version of their suite for Suns unless Linux/x86 version is released simultaneusly -- here it goes, ic5.00-linux!
Seeing IBM to strong-arm or sweet-talk MS into releasing office on PPC would be a bit too far-fetched, but, who knows, maybe MS would want to expand beyond Intel/AMD...
Paul B.
http://www.theonion.com/onion3311/microsoftpatents .html
Check out the link that someone posted in that IBM Mainframe thread!
Paul B.
I was so much expecting:
...Individuals know how to make other individuals. ;-)
Paul B.
Maybe because in different situations different threshold would have to be applied. E.g., if it is a terrorist monitoring camera on a random street corner, it might not be feasible to unleash FBI agents after every guy who matched at 80%, but if that random street corner happens to be in Washington, DC across the street from the White House, 80% confidence might be a reason to trigger further actions.
And if it is a camera in the cash machine and you claim that you are Joe and want to get your $500, you better match Joe's face at, say, 99% (it can also ask you to turn a bit and face the lens if your score is lower than some threshold.
Another example, if an airport screener can realistically check 10 people out of a hundred, she chooses ones with the highest scores. Yes, it might mean that John Doe in your example will be checked, and Osama will be not, IF there are other 9 people in line with scores >=95.
Algorithms used might be the same, but exact policy is implemented by taking scores into account.
There is more than binary yes/no in this world...
Paul B.
P.S. Not that I know anything about the actual numbers or policies, but I can see the value of having the scores available to people who program the machine, but not necessarily to the screeners (if any) who operate them.
Check THIS out:m age/35 _01.jpg
S imulator& ie=UTF-8&oe=UTF-8&hl=en
http://www.mext.go.jp/english/org/science/i
(it is a three story building, and I guess it also has its own powerplant).
For more pretty pix, of course:
http://images.google.com/images?q=Earth+
Paul B.
...or because Bill Clinton smoked weed once. Nobody is defined in any single instant.
;-) to people less fortunate in life?
It's not him smoking weed once, it's him maintaining the whole "War on Drugs" policy AND making non-violent people serve time in jail for what he himself did... Same with the whole "political correctness" thing, I'd never mind him fscking all the subordinate girls silly, but what right did he have to support denying the same "rights"
Proud to be a Libertarian in heart (though I can not vote for them, not a citizen of the US of A.)
Paul B.
It is not like he is actually trying to pay someone's living, he just wants to promote his game and make it a successful project! (If it were Leisure Suit Larry, I'd volunteer myself! ;-) )
;-) ).
;-) The latter might get you some extra geek-points (provided you care about those) and, maybe, much more in return when someone with bigger $$$ notices it on your wall... ;-) Just a thought...
And $100/week (or even $10/week) sounds more like a token of appreciation that any kind of financial reward (at least here, in the US. I bet there are many programmers, say, in Russia (my home country) who would not mind playing with the code AND making $100/week!
Look at it this way, if you'd get a check for $10 signed by Linus for your infinitely wise contribution to you know what, would you cash it or would you frame it and hang on the wall?
Paul B.
LHe is 4.2 K, I am not sure about LH, I know some people use solid H to cool things (I guess, Chandra space telescope was originally cooled by a big block of solid Hidrogen, then they installed a cryocooler on that).
And OF COURSE there is a whole bunch of metallic superconductors at 4.2 K, lead was most popular in 80s (IBM's superconducting supercomputer project), niobium is matherial of choice now.
You can deposit/etch it using almost the same litho tools and processes as CMOS people use.
Ceramic high-Tc superconductors are harder to make, and they are less useful "for anything but straight wires", but they work at LN temperature (77K), thus the interest in developing those.
Paul B.
Yeah, reminds me of a story which happened in our cryoelectronics lab, back in Moscow State University (Russia). OK, some guys bought frozen dumplings ("pelmeni", for those who know... ;-) ), and were not sure if they will not thaw by the time they get home, so they just poured a liter or two of LN over. Well, the next thing we've heard about those was half a week (!) later when the dumplings finally thawed enought to be separated and cooked... (pre-microwave oven days ;-) )
Paul B.
No, heavily overclocked CPUs should be cooled with liquid Helium!!!
;-)
Said by someone who is actually trying to design and sell one of those superconducting CPUs!
Paul B.
All the code involved is GPL'd, and the data are public: everybody can grok PageRank now!
GPL'd? Hmm, I thought that Google did patent the PageRank algorithm (correct me if I am wrong), so re-implementing THEIR algorithm even more efficiently would be incompatible with GPL. OTOH, if it is not THEIR algorithm, it can not be called 'PageRank'
Oh, the evils of software patents...
Paul B.
Hmm, you will have a GRADIENT of the index, but hardly a BOUNDARY (almost by definition, BOUNDARY is where index is not continuous and gradient can be considered to be +-Inf; OTOH, the "continuous boundary", if such a beast exists, is nothing but a gradient).
Not that I was arguing directly with the point you were making (I do not know if they had to compensate for this effect or not), but one can definitely imagine a physical effect when you have no sharp boundary between two matherials, but some interesting optical phenomena.
I think there were even lenses made out of FLAT "gradiated" glass.
Paul B.
Given proper heat/pressure/surface combination you CAN get different refractive indices (though no boundary between them) -- think of mirages+ desert&hl =en&lr=&ie=UTF-8&oe=UTF-8
http://images.google.com/images?q=mirage
Paul B.
I remember my (then-girl-)friend saying, ;-)
on the very OTHER side of Atlantic: "So,
you claim you do not like Fantasy? Well,
read this book..." --- and the next thing
I knew was me trying to bend the reality
of a neighbourhood street-corner cart to
carry the beer that I liked...
Just that you asked...
Paul B.
Not that I agree with your ~/.signature, ;-)
but I do appreciate the point!
Paul B.