8
$\begingroup$

We need to make these tag naming decisions (mostly together with synonyms).

  1. Do we want or ? (The other one would be made a synonym).

    ✔ It looks like is better. (Done.)

  2. Should we use the spelling cipher or cypher? (Now there are , , .)

    ✔ It looks like cipher is the only widely used spelling in cryptographic context. (Done.)

  3. How to tag questions about differential cryptanalysis? is one character too long (maximum length is 25, and this is not likely to change just for us). For now we have , which is not really optimal here.

    ➾ Is there some useful abbreviation, which still is recognizable?
    ✘ Seems like is okay.
    ✔ I changed the tag on the only question, and got autodeleted now. (Do we need a synonym here?)

  4. It looks like should be a synonym for .

    While technically TLS is a different protocol than SSL, they are quite similar (TLS 1.0 is more similar to SSL 3.0 than SSL 3.0 to SSL 2.0), and often implementations know (versions of) both protocols.

    SSL is the preferable tag name since the acronym is more known than TLS. (I'm just now writing a tag wiki excerpt + tag wiki which mentions both names.)

    ✔ (synonym made.)

  5. and are about the same thing (public-key cryptography). They have 6 and 4 questions now (with one question having both tags).

    ➾ Which one do we prefer (or better some even other name)?
    ✔ After Thomas' answer (and noone who later said something against it), I just merged both of these as .

(I mark here what seems community consensus as ✘, and what is done with ✔.)

I now mark this as completed, as the initial bag of tag names seems to be done. Open a new question for new ones, this will provide better visibility.

$\endgroup$
7
  • $\begingroup$ The 25-character limit applies to all SE sites, and previous requests to increase it have been declined. (Linked request on Gaming was status-declined; similar questions on meta.SE are closed as duplicates.) I think we should proceed as though 25 will always be the limit. $\endgroup$ – Dave DuPlantis Jul 15 '11 at 14:50
  • $\begingroup$ @Dori: Could we have tls be a synonym to ssl? (Explanation added to the question.) $\endgroup$ – Paŭlo Ebermann Jul 20 '11 at 17:46
  • $\begingroup$ @PaŭloEbermann Done. :) $\endgroup$ – HedgeMage Jul 25 '11 at 17:55
  • $\begingroup$ Community wiki? $\endgroup$ – user182 Jul 27 '11 at 15:23
  • $\begingroup$ @Muzz5: I think only moderators (or people with some higher reputation than me) can make a post community wiki. But you should be able to add edit proposals right now, then a moderator can approve them. Or simply post as comments, I will add them. (If you think about not gaining reputation: There is no reputation gain on the site metas, it is directly copied from the main site.) $\endgroup$ – Paŭlo Ebermann Jul 27 '11 at 15:28
  • $\begingroup$ Ah, sorry. I intended them to mean "this seems community consensus", e.g. discussion is done. What would be a better icon? $\endgroup$ – Paŭlo Ebermann Aug 6 '11 at 12:19
  • $\begingroup$ @Dori: I now use ✘ instead of ✔ where the discussion is finished, but the synonym not yet made. For now, the only important point is gpg. $\endgroup$ – Paŭlo Ebermann Aug 7 '11 at 21:19
7
$\begingroup$

would be the optimal tag for Differential Cryptanalysis. But this tag name is (one character) too long.

Currently is in use, but that is a bit poor. (Also, not really specific.) In the chat Thomas Pornin just proposed the naming .

in crypto.SE on can assume that "crypt" is involved pretty much everywhere

Google finds mainly some economic meaning (comparing business choices), but I think with a good tag wiki (and the tag autocompletion) it could be usable.

$\endgroup$
2
  • 2
    $\begingroup$ I think that [differential-analysis] is great. $\endgroup$ – davidlowryduda Jul 21 '11 at 21:10
  • $\begingroup$ I just changed the one occurence of differential to differential-analysis. I think we will not really need a synonym here (because of completion), but we will see. $\endgroup$ – Paŭlo Ebermann Jul 23 '11 at 15:38
5
$\begingroup$

I know I'm late to the party on this one, but we may want to consider making tls the default tag name and ssl the synonym to it, since tls is the way forward at this point, and awareness should hopefully increase over time while use of the SSL acronym fades away.

$\endgroup$
3
$\begingroup$

In the academic world of researchers who read and write articles about cryptographic algorithms, it is "cipher" everywhere, and "cypher" nowhere (well, almost nowhere, at least). If you need "massive" arguments, consider that Schneier uses "cipher" and that "Cypher" is the name of a "Matrix" character, which says it all.

I would recommend over on the basis that the GnuPG web site calls it, well, "GnuPG", and is www.gnupg.org. "gpg" is just the abbreviated name of the command-line interface on some systems.

I personally prefer over the shorter one, because what it means is immediately clear with the long name (which happens to match the name of the Wikipedia page on that subject, by the way). Length is not very important (in my opinion) because of the auto-completion performed by the interface.

$\endgroup$
1
  • 2
    $\begingroup$ About differential-cryptanalysis, the problem with too long sits in the engine. I added a screenshot to the question. (+1 for the other points.) $\endgroup$ – Paŭlo Ebermann Jul 15 '11 at 13:50
3
$\begingroup$

For anything related to asymmetric cryptography, I suggest (thus merging and into ).

Rationale: the tag is supposed to be what the average poster would first think about the subject at hand. The most striking feature of asymmetric cryptography is the apparent paradox of a key being made public. Also, the traditional expression for talking about asymmetric encryption, digital signature and their ilk, is "public key crypto".

$\endgroup$
2
  • 1
    $\begingroup$ Obviously the "average posters" added the other two tags instead :-) $\endgroup$ – Paŭlo Ebermann Aug 10 '11 at 17:46
  • 1
    $\begingroup$ Yeah you cannot let the average posters post by themselves, they tend to mess it up. But maybe a bit of guidance (autocompletion, synonyms...) could do marvels... $\endgroup$ – Thomas Pornin Aug 10 '11 at 22:01
2
$\begingroup$

As noone else wants to answer (are you all in the US timezones?), I'll start about cipher/cypher.

Wiktionary lists cypher as an alternative form of the main form cipher:

Alternative forms

  • cypher, less common than cipher but still in use in English. [...]

Wikipedia has separate disambiguation entries on Cipher and Cypher, where the latter mainly links to Royal Cypher (and quite some names of bands, films, albums, characters, persons), while the first one lists the encryption algorithms as the first entry, and also the plain title article refers to these.

The Cipher article still mentions that both spellings are valid, but uses the i-spelling.

Google searches on Cipher and Cypher show that the y-spelling is used about double as often. But inspecting the first some results shows mostly non-cryptographical meanings (movie, game, drug, company, aircraft, ...), while the i-spelling at least for half the hits on the first page has the cryptographical meaning.

Thus my recommendation: We should use cipher for the tag-names, using the y-form as a synonym. (For somy historical ciphers which were actually named with y, like the cypher No 1 mentioned in the wiktionary article, we could use the y-form.)

$\endgroup$
1
  • $\begingroup$ In research papers, I've only ever seen cipher used. I've never seen cypher used. So cipher should definitely be the primary tag. $\endgroup$ – D.W. Aug 18 '11 at 9:18
2
$\begingroup$

Would it make sense to rename to ? Just a thought because appears a little odd to me.

$\endgroup$
2
  • $\begingroup$ There seem to be two uses of the random tag: creating random numbers (i.e. PRNGs) (6×), and using random numbers (2×). Would it be useful to use separate tags here, e.g. [randomness] and [prng]? (With the right tag wiki to distinguish them, of course). $\endgroup$ – Paŭlo Ebermann Sep 6 '11 at 14:20
  • $\begingroup$ @Paŭlo actually, perhaps, yes. Then we could blacklist the random tag, because I don't think "random" is a great tag to have, really. $\endgroup$ – user46 Sep 6 '11 at 16:24
1
$\begingroup$

For and , they are the same thing, so I think we should go with the more concise one ().

I mean, would you rather type out or ?

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .