User talk:Wapcaplet/Archive 2

Column width
Hi, thanks for your info on width, which I fear is well over my head! All I was trying to do with my tennis article was to have three columns, one for the year, one for the winner, and one for the runners-up, in which the width of each one seemed reasonably in proportion, not too narrow and not too wide. If you go to that article you'll see what I currently have -- I think that it looks Okay, except I'd like to be able to set the width exactly and I'd like to be able to get rid of that very narrow, almost invisible column at the far right....

I'd like to get this right, so that I can copy the same format into a couple of other articles that I'll do one of these days.

Thanks for any help you can give me on this....

And by the way, are you spelling your name wrong, Pierce, "I before E, except after P"?

Best, Hayford Peirce

Enigma
Hi, here's a reply from a while back (from User talk:Matt Crypto):


 * ... I think it'd be fun to model an Enigma machine, provided I can find some good diagrams to work from, but I suspect that the workings of that might be better illustrated with this kind of diagram; I am not sure what a 3D representation would add to that. Any suggestions? -- Wapcaplet 23:22, 5 Jun 2004 (UTC)
 * I think your right about modelling the entire Enigma machine; most of the cryptography happens in the wiring, which is hidden away inside. One thing that could well be useful, I don't know if you'd be interested, would be to model the "rotor", a component of the Enigma. It'd be great to have a diagram comparable to this one: Enigma rotor from several angles or Photo of a "decomposed" rotor. The Enigma machine article is currently a Featured Article candidate (it could do with quite a bit of work first, though), and will likely be included in the WikiReader Cryptograpy. An illustration of the rotor would be useful a couple of reasons: 1) The actual shape / construction of a rotor isn't obvious from a written description. 2) Illustrations are often clearer than photos (and I don't (yet) know of any PD rotor photos). If you're interested, I'd be more than happy to dig up any photos and illustrations that might be useful (there's tons on the internet). &mdash; Matt 21:16, 27 Jul 2004 (UTC)
 * Thanks! That would be very groovy ;-) Here's various images on the internet which may be of help (there's tons of photos out there, though):


 * http://www.netzmafia.de/software/enigma/walzen.jpg
 * http://www-wi.cs.uni-magdeburg.de/~pauer/images/enigma_rotor1.jpg
 * http://www-wi.cs.uni-magdeburg.de/~pauer/images/enigma_rotor2.jpg
 * http://www.chss.montclair.edu/~pererat/u_114c.jpg
 * http://www.chss.montclair.edu/~pererat/u_085f.jpg
 * http://www.enigma-replica.com/rotors.html
 * &mdash; Matt 22:59, 27 Jul 2004 (UTC)


 * I've replied at User talk:Matt Crypto &mdash; Matt 02:52, 28 Jul 2004 (UTC) (and again &times; 4)


 * Really very nice rendering of the Enigma. Also love your car parts models. The Yafray rendering look much better - have you tried it on the enigma image? Also could you upload a version without labels, as I don't think they're currently of the same high quality as the rest of the image. ed g2s  &bull;  talk  17:07, 30 Jul 2004 (UTC)
 * Well because of the paletting - the anti-aliasing on the lines looks a bit poor - and they're also incosistent in how close they get to the part they're labelling. You could probably get away with all the numbers (which could be a tad smaller) on one side - with just straight lines. But still, great work. ed g2s &bull;  talk  22:33, 30 Jul 2004 (UTC)
 * "The aliasing problem is most likely a side-effect of reducing the image to 256 colors" - this is what I mean by paletting - turning a true colour image into a paletted/indexed one (the 256 colours being the palette). If the lines were vertical you wouldn't have this problem. I'll also created a 256 colour thumbnail (the auto-thumb is true colour and 40k, almost as big as the original). (FYI your background is off-white) ed g2s  &bull;  talk  23:37, 30 Jul 2004 (UTC)

congrats!
w, Aside from sharing the proper initial with you, I'm in awe. I'm a writing sort, and have terminal congenital drawing incompetence, so your diagrams at Engima machine are astounding, astonishing, awe-inspiring, and so on through the b's and to the end. Thanks for very very good work. I will vote for using one or more as cover images on the eventual WikiReader.

On another subject altogther, I used to live near Co Spgs and remember it fondly. You may be interested to know that RAH lived for several years at 1776 (Garden of the Gods Road?). As for wive's beliefs about addiction, well... We all understand that, don't we? ww 16:13, 2 Aug 2004 (UTC)
 * w, RAH=Robert Anson Heinlein (winner of 4 Hugos and the 1st Grandmaster Award, inventor of the water bed and the name (and concept(?) of) waldo and perhaps of the powered augmentation the Army is researching for its soldiers (see Starship Troopers), author of etc etc etc).
 * I'm not sure about the road name, but the 1776 is correct. If it's all industrial there now, that's a considerable loss as I dimly remember the area. ww 18:46, 2 Aug 2004 (UTC)
 * Having now looked up the address, it's 1776 Mesa, not Garden of the Gods... sorry about that. ww 13:29, 9 Aug 2004 (UTC)

Thanks!
Thanks for all your work on the Enigma diagrams, it's much appreciated (they look darn good!). I've nominated the exploded rotor pic at Featured picture candidates, and as ww mentions, they'd make excellent cover images for the WikiReader in Cryptography too. &mdash; Matt 01:47, 3 Aug 2004 (UTC)

Reply
Thanks for your comments about DES! I quite like the output of Dia, but I find the user interface to not-so-good (my version has few keyboard shortcuts, for example); maybe I should look to see if there's a newer version.

You foolishly said, "are there other crypto-related articles in need of illustrative assistance?" ;-) One thing that a lot of the cryptography articles lack, in terms of illustration, is simple diagrams showing people communicating / various protocols / public key cryptography, etc. http://www.nuitari.de/crypto.html has several good examples of the type of diagram I'm thinking of. What might be quite useful is to have some "primitives" (aka Clip art), like keys, padlocks, Alice and Bob etc, so that these kind of diagrams can be put together rapidly. Alternatively, Rijndael is in need of a few diagrams; John Savard has a few 3D diagrams (e.g. ) that illustrate the encryption (the state is 2D); maybe one could be created along those lines? Alternatively, if you've got enormous amounts of free time, you could even build a model of a M-209 (Hagelin) cryptograph, a purely mechanical encryption machine, which looks fiendishly complex to me. &mdash; Matt 15:07, 5 Aug 2004 (UTC)
 * Yes, that was rather hasty of me :-) I like the idea of having some standard clip art to use; it'd give a nice level of consistency to crypto articles. The reference site you gave me has good ones; I especially like the meat grinder for hash functions. John Savard's site looks like a wonderful resource - we should get him into Wikipedia! The M-209 looks like it could be a great challenge, but it'd likely require a lot more reference material than is readily available on Google. Ideally, I'd have a replica sitting next to me to model from; I'll peek around the web and see if I can turn up anything useful, though. If you know of any books I might find at the local library that would aid me, let me know. For now, I'll see if I can put together some basic clip art, and see whether I can wrap my head around AES before trying to illustrate it. -- Wapcaplet 18:18, 5 Aug 2004 (UTC)
 * I've made some clip art showing a key, locked and unlocked palock, a long key (perhaps to illustrate the difference between a 64-bit key and a 128-bit key), a lock-box (for trading public keys or whatnot), Alice and Bob. More to come! I'll post the SVG versions at some point. -- Wapcaplet 21:43, 5 Aug 2004 (UTC)
 * I've made a few more (not posted yet), with an open lock-box, an ink pen (for digital signatures), and public and private key (shown by an open/closed hand holding a key). I'm striving for language-independent illustrations and having difficulty coming up with something good for plaintext and cryptotext; with an English phrase (and an accompanying scramble of letters), I could pretty clearly convey the idea of a plaintext message, but how to do that without using a language? Two ideas I've come up with are (1) Use digits, e.g. "12345" for plaintext and "35142" for cryptotext, but that doesn't quite accomplish what I'm after, since it implies that the plaintext is just rearranged; (2) use a recognizable picture, for instance a depiction of a five-pointed star, and an accompanying "scrambled" version of the picture in which the star has been chopped up and rearranged (maybe a puzzle-style cut-up of the picture); but that doesn't really explain how possession of a "key" helps to solve the puzzle. Any suggestions? If no better options present themselves, I may just go with English text. Can't come up with much of anything better for a hash function, other than the meat grinder (maybe paper shredder, but that seems more like a way to destroy something, rather than transforming it); problems with the meat grinder include the fact that it appears to destroy the original, and does not really produce a unique output. The idea with hashing is to produce a short, distinct summary of sorts. I'd welcome your input. -- Wapcaplet 04:02, 6 Aug 2004 (UTC)
 * It's a very nice start &mdash; I especially like the "long key"! I can't really think of anything good for plaintext; perhaps we could copy this discussion onto the Cryptography WikiProject page, and get one or two other cryptographically-minded people to think about it? Likewise, I can't think up an alternative to a meat-grinder / shredder, but again, I'm not very imaginative; I guess the basic security point about hash functions (in cryptography) is that they're one way .. you can go one direction, but not the other. &mdash; Matt
 * W, I too am amazed. What are you doing in CS when you have a clearly graphically enabled brain. It's usually for the artistically impaired. Anyway, I like everything you've done (that I've seen), and appreciate the effort. I too have tried to illustrate asymmetric key crypto, but aside from looking like kid's refrigerator art, the concepts I came up with are not as elegant as the open/closed fist thing. Very good!
 * On another one of your diagrams I would suggest caution. The term digital signature is unfortunate and was ill chosen, but I suspect we are stuck with it. It has legal connotations in re authorized consent/agreement contexts and the crypto engineering does not well overlay even the ordinary use of 'signature' much less the legal use. The article has some coverage of this, but in the interests of brevity (and not being elided) it's brief. I would think a notary stamp might have some potential, except that it's such a good fit for the fundamental meaning of digital certificate. Maybe marking the back of a hand with UV ink at a concert or something. You see that I'm running out of gas -- a very small tank for me in these matters.
 * Good work! User:ww 14:03, 9 Aug 2004 (UTC)
 * Another idea or two just fell into my head. Since long keys are sometimes much looooonger than shrt keys, perhaps a not as wide stem in the long key to stress the loooongnessss? One graphic idea for lockbox I've seen is a stereotypic pirate's treasure chest -- it stresses the value (and secrecy (& privacy(?))) of the enclosed thing(s). Kind of cartoonish, though. There is much known about Alice and Bob; a graphic representation of them might reflect some part of that; the suggested icons are awfully, well, GUIconic. And in regard to graphic representation of plaintext vs cryptext (spelling ideas on that? I've been dithering for 4 years...). There is a typographic convention called Greek which uses random Greek alphabet letters to suggest actual text. Since most machines/browsers can render some of the Greek characters, how about that as a possibility with, of course, the canonical "attack at dawn" as the plaintext sample? There's probably something wrong with this, but it was just a graphical idea after all.... ww 14:14, 9 Aug 2004 (UTC) ...though it does have a literary pun sort of meaning with the Shakespeare line about it being Greek to me (from Juilius Caesar is it?) ww
 * w, That's 'alien wombats living under the Greenland ice', thank you very much. It's mine of course, I told you I was a word person.... I didn't say anything about a non lurid imagination.
 * As for visual analogies with digital signatures, the problem is quite real. The legal beagle types got their knickers all in a twist in the mid 90s and wrote all sorts of stuff (mostly bumph) about the legal significance and use and potential of digital signatures. Several legislatures (here and elsewhere) followed with actual law on the subject. Also mostly bumph. Not good as law based on misreading of the engineering is likely to be (and was in this case) not too good and that causes real trouble in real life.
 * In law, there are several aspects of a signature on some document. Only one, and in some sense the least important, is the 'no one else can produce it' thing. Which isn't true anyway, given skilled forgers. Probably more important is what is called ceremony, and is related to what used to be used to prove seissen (sp?) for a piece of land, namely who was there, who knew whom, who watched it happen, who noticed that Alice was (or was not) staggering drunk (or high on something) at the time, that Mallory did (or did not) have a pistol to her head just before the siganture, .... NONE of which does a digital signature, even one done correctly with a correct implementation of a high quality algorithm(s) used in a provably secure protocol, achieve. Statutes may enact and courts hold that a digital signature is proof positive (just as good as a regular signature), but calling a pig a horse doesn't make it so. I resist letting a sloppy analogy of language become established and lead to poor thinking. Such small things can have large consequences.
 * But of course I have no better ideas, being a non visual type. The notary seal is the only thing that comes to mind and it really does fit much better with the idea of certificate than with a signature. I know, maybe a tattoo somehow? Or something.
 * It's important enought to warrant signficant effort to get as 'right as possible' though. Your choice might turn out to be rather influential, WP being what it is. That's why I brought it up.
 * As for Alice and Bob, there's an external link at the article to an hilarious piece about them. Maybe that'll give you some ideas. Think Mata Hari maybe, or Dudley DoRight's nemesis, Snively Whiplash, from the Rocky and Bullwinkle series. In fact, they might give you some good ideas for much of this since Boris and Natasha were spies and had to do many secret and confidential things; Ward's people had lurid imaginations. Spy vs Spy might prod a few brain cells too. There's at least one book of collected panels. That guy was most peculiar; but then so was the rest of the usual gang of idiots down at Mad. ww 19:05, 9 Aug 2004 (UTC)
 * The drugs are making me imprudently bold. I've had an idea for visual representation of random number generation. Think numbered ping pong ball delivery mechanisms. Eg, rotating bingo drums on a spit, the sort that come in the 'home' bingo sets. Or, my favorite, a 'magical' fountain of balls out of which comes a series ('magically chosen', of course). My favorite since it stresses the (currently) unknowable quality. Imation's logo may have a hint for you. Or what about Mickey's magical broom (from Fantasia) -- or something not traceable to Disney, they have legal friends in low places -- whacking numbered balls out of the fountain. Something similar relying on radioactivity requires that the viewer understand the unpredicability of decay mechanisms, not something the average viewer is likely to bring to the party. Consider that some poll or other (Gallup?) has reported that more than 50% of the US population thinks there was no landing on the moon. Faked in Hollywood or Area 51 by Langley or some other secret nefariousity. Re-reading this, I think the ibogaine has gotten to me. ww 15:09, 11 Aug 2004 (UTC)
 * W, "Blessed"? Hah! You yourself have contributed to the agony. It was 'alien wombats under the Greenland ice sheet', and you persist in talking about other alien wombats (which might not be under that ice). Do you have more information? Do you know about other alien wombats? You have no idea the problems...
 * Never mind.
 * Your report of still further poll 'evidence' of dunce is simply depressing. And I was depressed enough already. I'm considering a move to Death Valley (not depressed yet enough to learn to breath water, nor to move off the continent). Soon maybe.
 * Other ideas have come to me. (It was a very LARGE voice, and very very deep, too). How about a billiard cue selecting the numbered balls (for RNGs)? Or a pen with 10011010 flowing out of it (for digital signature)? Or serial porcupines passing messages along (for secure channel)? Or a cat herder (for a crypto system admin coping with users/configuration problems/OS insecurities/...)? Or maybe for sys admins generally. Or a gangster connection (think George Raft or Cagney or Edward G Robinson) (for the bribery, extortion, threat ... attack)? Or a cephalopod 9with earphones?) (prob octopus) (for Eve the eavesdropper)? Or a little kid skipping along tossing flowers (= message copies) out of a basket (for insecure channel)? Or trenchcoat spy type (think Peter Lorre?) (for the task of distributing keys safely -- perhaps an up to date PL would use the porcupine line?)? Or a judge in robes or maybe a wig (for Trent the trustworthy -- I know, lots of judges aren't, but the best I can come up with)?
 * Enough. This is getting out of hand. ww 14:23, 12 Aug 2004 (UTC)