Talk:Cryptosystem

From Wikipedia, the free encyclopedia

WikiProject on Cryptography This article is part of WikiProject Cryptography, an attempt to build a comprehensive and detailed guide to cryptography on Wikipedia. If you would like to participate, you can choose to edit the article attached to this page, or visit the project page, where you can join the project and see a list of open tasks.

Contents

[edit] Naming: "Cryptosystem" vs "Crypto system"

"crypto-system" or "cryptosystem" instead of "crypto system"? Sorry to mention alternative spellings on a crypt* article... ;-) Matt 20:53, 10 Mar 2004 (UTC)

Mostly what I see, and what I use in my own stuff is crypto system. That's why I chose it for this. If a redirection page is needed, well, I guess we should add one. As for being sorry about alternative spellings, check out talk:cryptography for the cypher vs cipher controversy. ww 18:02, 11 Mar 2004 (UTC)
Coming back to this again, I think "cryptosystem" is the correct spelling in formal writing. Some evidence: a Google test reveals 10 times more hits for "cryptosystem" than "crypto system"; and HAC and Schneier's AC both use "cryptosystem", as does Ross Anderson's "Why cryptosystems fail" — Matt 10:08, 16 Apr 2004 (UTC)
Matt, As I suggested in re cipher v cypher above, this is a bit of teapot tempest. With the balance being toward cryptosystem in others usage, I don't suppose that I have much room to insist on anything else.
Since there are multiple references to crypto system here and there, a redirect page will be needed. Satisfactory? ww 13:36, 16 Apr 2004 (UTC)
Yes, sorry to go on about spelling (pedantic nit-picking is something I try and avoid, sorry, try to avoid...) My excuse is that I think there is a real danger of things looking slightly amateurish, just because of the spelling. e.g., recently I saw a title of a paper with the spelling "cryptoanalysis" rather than the more common "cryptanalysis". Probably both are OK historically, I'm not sure, but I deduced immediately that the authors weren't primarily cryptographers (which was true). w.r.t. cipher vs cypher, I'm a lot happier with both spellings than I was previously, but first a quick quote from the WP Manual of Style:
>>> * Articles which focus on a topic specific to a particular English-speaking country should generally aim to conform to the spelling of that country. A reference to "the American labour movement" (with a U) or to "Anglicization" (with a Z) may be jarring. However, a reference to "the American labour movement" would be okay on New Labour. <<<
Thinking along these lines, along with noting that very recently the academic world has standardised on "cipher", I would suggest that "Rijndael is a block cypher" is jarring for a similar reason, even though "the Enigma cypher" is OK. — Matt 14:02, 16 Apr 2004 (UTC)

Matt, I give up on the ':' after a certain point. It seems my limit is 4. Anyway, we use the same gloriously absurd language in which my father's middle name was actually pronounced kuhnicht once upon Chaucer's time or perhaps a little before. And still spelled that way! And remember you can spell an aquatic animal with fins, ghoti -- per GBS! I agree that some spelling choices convey hints or allow one to make WAG about the writer, but the content is after all more important than the spelling. At least in this absurdist orthography. Groucho, Chico, and Harpo could have done better.

The WP style guide also says that it's agnostic on AE vs BE spelling but that the spelling in an article shouldn't be changed around just for that. Though many folks do. Peruse the foolishness about cypher v cipher at Talk:cryptography for an example. Imran even remembered a good example of cypher which I'd forgotten about before he reminded me.

I say, if Shakespeare could spell his own name five different ways, who am I, a next to functional illiterate if you watch my typing, to say anything different. It's mostly harmless, especially when redirect can take out most of the search confusion.

Got to go, unfortunately. ww 20:51, 16 Apr 2004 (UTC)

[edit] Illustrative example

This article discusses the nature of cryptosystems using as an illustrative example the design and implementation an influential modern cryptosystem—PGP. PGP is computer software, as essentially all practical cryptographic systems now are. It is more complex than earlier hand or mechanical rotor systems, but was chosen because it is far more typical of modern cryptographic practice. Version 2.6x is used as an example rather than say the OpenPGP standard, as it is conceptually similar, but has fewer algorithm choices.

[edit] Design overview

encryption, integrity, non-repudiation, key exchange

[edit] First-level issues

MD5, RSA, IDEA, hybrid use of encryption for efficiency purposes, key distribution, digital signature of certificates / messages, key vetting, open source, use of publicly known and examined primitives

[edit] Second-level issues

RSA/IDEA key choices, random inputs

[edit] Third-level issues

format and content of certificates, choice of certificate vetting mechanism, lack of truly secure channels for key distribution and vetting,

I removed the above from the article. Whoever started it has probably lost interest in trying to finish it, since its been more than a year, and in its current state it does not seem to have any useful information. Arvindn 17:16, 22 July 2005 (UTC)