Talk:ROT13

Missing image Cscr-featured.png Featured article star

ROT13 is a featured article, which means it has been identified as one of the best articles produced by the Wikipedia community. If you see a way this page can be updated or improved without compromising previous work, feel free to contribute.

Missing image Key-crypto-sideways.png WikiProject on Cryptography

This article is part of WikiProject Cryptography, an attempt to build a comprehensive and detailed guide to cryptography in the 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.

Template:CryptographyReader

Contents

1 Is ROT13 symmetric key? 2 Rot13 and bubble sort 3 Skylarov arrest in re after or for 4 reversion reasons 5 Furrfu 5.1 FAC Nomination, successful 6 Memfrob? 7 A source? 8 Screenshot

Is ROT13 symmetric key?

Is ROT13 'symmetric key' ? I mean there is no 'key' as such, but the procdure is same ai'nt it? --MuthuKutty 07:08, 16 Nov 2004 (UTC)

Yes, but I'm afraid I don't really see much benefit in having such a discussion in the article; the whole distinction of assymetric vs symmetric key cryptography is not really relevant as far as ROT13 is concerned, because it's not used for encryption. (I moved the following sentence here from the article: So ROT13 is called as a Symmetric Key algorithm, only that the key is non-existent.) — Matt 10:22, 16 Nov 2004 (UTC)

rot13 is essentially the more general ceaser cypher with the key fixed at 13 rot in general is symetric because the decode key can be trivially derived from the encode key and vice versa

if we encode in rot(x) then decodeing is the same process as encoding in rot(26-x) hence to decode rot13 is the same as encoding in rot13 Plugwash 03:32, 21 Nov 2004 (UTC)

Should we be linking to Bubble sort because it is another 'bad algorithm'? Sure, BS is a bad algorithm but ROT13 is pretty good at acheiving its aim of obfuscating text and not 'bad' at all. EddEdmondson 22:01, 2 Jul 2004 (UTC)

Well, bubble sort achieves its aim as well: it sorts data. It's only qualified as "bad" because it's slow compared to its competitors. I don't think ROT13 should even be qualified as a "bad" algorithm - what's bad about it? If it were billed as a secure encryption algorithm, then it would be bad. But it doesn't pretend to be secure encryption, only a sort of encoding scheme, like hexadecimal notation or Base64. So I don't think we should link to bubble sort at all. Decrypt3 00:09, Jul 3, 2004 (UTC)

Rot13 and bubble sort

Bubble sort is a fine algorithm and one that I am proud to have implemented many times. For short lists, it is ideal because it is simple, straightforward, and not prone to implementation error due to its simplicity. It is also fairly effective for lists that are already mostly sorted, such as those that have one or a few new elements added. Compared to qsort, it is faster on an already-sorted list and about the same on an almost-sorted list. ROT13, in the same way, is a fine algorithm, although I cannot say that I've ever written it into any commercial software. ROT13 serves its purpose -- which is obfuscation, not encryption -- and is simple, straightforward, and not prone to implementation error due to its simplicity. This is the similarity, not that either algorithm is necessarily "bad."

UninvitedCompany 03:54, 3 Jul 2004 (UTC)

ROT13 is "bad" only when it is mistaken for actual encryption -- which nobody but the greenest of newbies would do. This is why "DES is as good as ROT13" is funny (to crypto buffs) -- it means not just "DES is bad" but "DES looks like good encryption, but only an idiot would use it." (This is a derogatory comment; that is rather the point.)

Bubble sort, in contrast, is considered the classic example of a naïve sorting algorithm. While it does have its strengths (as you describe), it performs poorly in the general case: unless you know you'll have mostly-sorted lists, you are better off with another algorithm. Bubble sort is frequently taught first in algorithms classes because it is simple to explain and write, and is a good contrast for other (faster, in the worst and average cases) algorithms such as Hoare's quick sort and tree sorts. --FOo 06:01, 3 Jul 2004 (UTC)

Skylarov arrest in re after or for

Matt, The DMCA prohibits the act of investigating as well as talking publicly about. Kind of like burglar tool possession, no matter what you intend, is illegal in certain places. In this case, after is temporally correct (that was when he was arrested), but so is for (as the reason he was arrested was what he did). ww 16:32, 2 Aug 2004 (UTC)

reversion reasons

Matt, In the first instance you reverted, the intent was to note what DMCA regards as an offence (in the context of the point the note was attempting to make), something which was not quite evident in my judgement. In the seond instance, it was to correct the grammar. As it was orginally, and is now, the grammar (verb tense) was incorrect. I'd suggest both be restored. ww 15:14, 10 Aug 2004 (UTC)

Here is the version after your change:

The DMCA introduced a broad ban on the , circumvention (defined to include study and discussion thereof) of copy protection systems, which have too often employ insecure methods of cryptography.

Here is the version previous to your change (and after the revert):

The DMCA introduced a broad ban on the circumvention of copy protection systems, which often employ insecure methods of cryptography.

To my eye (and I'm not an English language expert), the grammar was better before your change ("copy protection systems often employ insecure methods"); the phrasing could be improved, I guess. Regarding what the DMCA regards as an offence, I think this is something we should be careful about describing, since we aren't (I suspect) lawyers. The DMCA does not ban the study of circumvention methods per se; it may well ban the public dissemination of products which circumvent copyright protection, and so forth. However, the only reason we mention the DMCA here is to explain why people use signatures like "Encoded with ROT26 — circumvention will be prosecuted"; I think the current sentence is quite sufficient for this purpose. Any more is unnecessary, and (as we aren't lawyers) we are quite likely to get it wrong. — Matt 00:28, 11 Aug 2004 (UTC)

Matt, Oops. Should have been 'employed' after my change. Wtih that modification, I still prefer, on English grounds, my version.

As for the DMCA, I'm not a lawyer either. But,... Richard Smith, of PharLap fame, has publicly said that he expects object code debuggers to be banned under the DMCA, if any prosecutor notices. It is in fact, absurdly drafted, banning even investigation into (w or w/o public disemination of the information, regardless of reduction to practice or embodiment into a product whether sold or not). Felten at Princeton was threatened for proposing to publish the results of an analysis (invited as part of a test project) of an (RIAA?) security scheme. Ferguson has declined to discuss the flaws he found in Intel's scheme, even though Intel has said they have no objection. It's worse than you would have WP suggest here; we should not err in the other direction, either. This was the background for my parenthetical addition here. In comparison to RIP, DMCA is even dumber. RIP implements boneheaded public policy and is based, at least in part, on a mistaken impression as to the operation of some crypto systems. DMCA is the moral equivalent of the Indiana legislature's (almost) enactment that pi=4. And it implements equally brain dead public policy. ww 14:44, 11 Aug 2004 (UTC)

Furrfu

FAC Nomination, successful

Interesting, well written. Image? -- Fredrik | talk 16:49, 17 Oct 2004 (UTC)

• 'Support - Interesting. Well written. JOHN COLLISON [ Ludraman (http://en.wikipedia.org/w/wiki.phtml?title=User_talk:Ludraman&action=edit&section=new)] 18:09, 17 Oct 2004 (UTC)
• Support. Excellent. , just needs a picture. Zerbey 18:22, 17 Oct 2004 (UTC)
  • Has an excellent diagram now. JOHN COLLISON [ Ludraman (http://en.wikipedia.org/w/wiki.phtml?title=User_talk:Ludraman&action=edit&section=new)] 12:27, 19 Oct 2004 (UTC)
    • That'll work :) Zerbey 15:01, 19 Oct 2004 (UTC)
  • Why? It doesn't really seem like a picture sort of article. Pcb21| Pete 08:35, 18 Oct 2004 (UTC)
• Object. 1) There are no references. 2) The article is a bit short. I can't think of a lot to add, but a better description of the cryptanalysis is needed. There is nothing said about frequency analysis which is needed to break the encoding. 3) As for the image: I don't think one is required here, but adding one would be good. Jeronimo 11:33, 18 Oct 2004 (UTC)
  • As I said above, now has a diagram. JOHN COLLISON [ Ludraman (http://en.wikipedia.org/w/wiki.phtml?title=User_talk:Ludraman&action=edit&section=new)] 12:27, 19 Oct 2004 (UTC)
  • 1) There are inline references, and some of the external links also serve; I've now added a reference to the Jargon file definition. Do you think there's an obvious need for more? 2) I've added a paragraph to show how ROT13 could be easily broken if used as encryption using frequency analysis or pattern words. Does this suffice? — Matt 13:19, 19 Oct 2004 (UTC)
    • Good work, Matt! Fredrik | talk 14:34, 19 Oct 2004 (UTC)

Good work indeed - support. Jeronimo 19:35, 20 Oct 2004 (UTC)

• Support. Looks like a good article — there's a lot more about the topic there than I even realized existed. A picture would be nice, but not crucial, in my opinion. Factitious 07:19, Oct 19, 2004 (UTC)
• Support'. A picture is not needed here, imnsho. Anárion 08:58, 19 Oct 2004 (UTC)
• Object. Good article, I find it very interesting but it needs references for my support. [[User:Norm|Norm]] 09:11, 19 Oct 2004 (UTC) Support. [[User:Norm|Norm]] 15:32, 19 Oct 2004 (UTC)
• Support once a couple more good references are added cleaned up. The new diagram is great to represent the topic and explain it. Maybe it should even be moved up though. The references need to be one consistent style. - Taxman 14:56, Oct 19, 2004 (UTC)
  • I don't suppose that you could you suggest the type of references that are needed — citing specific sources for certain facts, or general references? — Matt 15:04, 19 Oct 2004 (UTC)
    • Someone add a "reference" to a joke paper about how ROT13 was especially good if applied twice, and at least one user changed their decision to support after this reference (along with a jargon file one) was added. Sometimes it seems like people don't care what the content of an article is, so long as it "ticks all the boxes" like "has an image", "has references", "has a certain length first paragraph". Pcb21| Pete 15:53, 19 Oct 2004 (UTC)
      • I added these references. The reference to a joke paper was fine because it was referred to from the text in a section about the use of ROT26 as a joke — literally a reference, as opposed to further reading; it's a useful example of a certain social aspect of the topic. It's true, of course, that the line between "References", "External links", "Sources" and "Further reading" gets pretty blurred sometimes on WP. The Jargon file reference is more of a "Sources", I guess. — Matt 16:06, 19 Oct 2004 (UTC)
    • Well I didn't qualify it because it could use any references that would verify the facts in the article. Specifically a reference that detailed the frequency analysis info would be helpful. Also the references, external links and citation in general needs to be cleaned up. There are 3 in article external links that appear to be intended as footnotes that get the standard external link numbering [1], etc. Then there is a single superscripted footnote with inconsistent numbering to the external links. The inline external link citations are not listed in the references or external links sections. Were all of the external links in the external links section used as references or are they just there for additional information for the reader? One syle should be chosen for citation, for ex. either inline, consistently numbered external links (also listed in the references/external links section), or more like MLA style with (author, year) inline and the full citation at the end. I can work on some of that, but those familiar with the subject will need to provide the additional quality references. - Taxman 17:02, Oct 19, 2004 (UTC)
      • For frequency analysis, I've added a couple of books to the frequency analysis article; I think they'd be a little too off-topic in this article. — Matt 13:45, 25 Oct 2004 (UTC)
• Support. Mpolo 16:59, Oct 19, 2004 (UTC)
• No vote yet. This is a good article, I'd just like a little bit of an explanation on the Unix command. Is there a way for laymen to understand its syntax? The article has improved with the new picture and the formatting of the tables. I'd be happy to support as soon as I can get some information verified. Support[[User:MacGyverMagic|Mgm|^(talk)]] 20:13, Oct 19, 2004 (UTC)
• V nz pbaprearq gung srnghevat guvf rapelcgvba flfgrz pbhyq pbafgvghgr n oernpu bs angvbany naq vagreangvbany frphevgl. Jrer gur EBG13 frphevgl cebgbpny gb snyy vagb gur jebat unaqf, rnpu naq rirel fvatyr wbxr ba gur Hfrarg jbhyq fhqqrayl orpbzr genafcnerag gb bhe rarzvrf... naq gura jurer jbhyq jr or? Arireguryrff, V ibgr Fhccbeg. shap(gnyx) 19:48, 20 Oct 2004 (UTC)
  • Thank you, func, for you support. Now, if anyone else thinks it would be funny to write their comment in ROT13, I swear, things will get very nasty. :) JOHN COLLISON [ Ludraman (http://en.wikipedia.org/w/wiki.phtml?title=User_talk:Ludraman&action=edit&section=new)] 22:00, 20 Oct 2004 (UTC)
• Another Fhccbeg, err, 'Support here. *Kicks himself for not thinking of that joke first* Kiand 20:23, 20 Oct 2004 (UTC)
• Support, I really like the table and the reversals. --Alex Krupp 04:36, Oct 23, 2004 (UTC)
• Support. Still can't understand the memfrob bit, but everything else dealt with. jguk 22:15, 25 Oct 2004 (UTC) Object It's difficult for me as a layman unfamiliar with computing technology to understand bits of this. But it's nearly there. If I have time I get round to it, I'll amend the article myself to deal with my objection. Main points of confusion/suggested improvements are:

  1. <s>Move the explanation that ROT-13 stands for 'rotate by 13 places' to the lead section. The 'description' section would then need a slight tweak to avoid duplication;
  2. Needs a brief explanation of what 'eBook copy protection systems' are;
  3. I can't believe many (non-computing bods) are familiar with what 'ASCII' is. A brief (one sentence, say) explanation would be helpful';
  4. I have absolutely no idea what the 'memfrob()' section is about. It should either be removed or reworded so a layman can understand it;
  5. It would be helpful if the lookup table included 'Description' could also be included in the Trivia section.
  6. The word 'newbies' in the Trivia section is slang. Whilst I believe a not too formal style is appropriate for many articles, I don't think articles should use slang either.

jguk 17:39, 23 Oct 2004 (UTC)

• • I've tried to address some of these. 1) Done. 2) I've added "Ebooks (books available in digital or electronic format) sometimes include technical measures to enforce copyright." 3) I've added "Instead of using the sequence A-Z as the alphabet, ROT47 uses a larger alphabet, derived from a common character encoding known as ASCII." 4) I've reworded this slightly. I think that a layperson would probably have to follow the links to GNU C library byte, XOR, binary to understand this fully, but he or she should (hopefully!) get the gist. I think memfrob() is worth mentioning, but it's not sufficiently notable to warrant more than a terse paragraph. I've also moved the "Variants" section to the end, so this technical paragraph will be the last thing a reader gets to in the article. 5) Done. 6) Replaced "newbies" with "newcomers". — Matt 13:45, 25 Oct 2004 (UTC)

• Support Dysprosia 06:07, 24 Oct 2004 (UTC)

Memfrob?

I really dont think the memfrob operation is a variant of ROT13 and isnt related. MEMFROB is based on XOR with a magic number (which is always reciprocal for any number and is a very common way to obscure data) and ROT13 is based on pretty much modulo arithmetic. Completely different operations... at best memfrob should be just a *see also* link 19:58, 21 Nov 2004 (UTC)

memfrob() is similar to ROT-13 in purpose, a sort-of generalised ROT-13 for binary data. It wouldn't really be worth putting it in its own article. You can also consider them to be similar in form, too, from a more abstract point of view. They both take sequences of elements of a group G and add a fixed public constant <math>k\in{G}<math> to each:

<math>C_i = P_i + k<math>

In both cases, the addition of <math>k<math> has the property that doing it twice recovers the original:

<math>P_i = P_i + k + k = C_ i + k<math>

For ROT13, the group is the integers modulo 26 under addition, k=13; for memfrob(), it's bytes under XOR with k=42. — Matt 20:36, 21 Nov 2004 (UTC)

A source?

the coding method had been in use for decades before being applied to computers,

I don't suppose anyone knows of a source for this statement? — Matt 21:34, 21 Nov 2004 (UTC)

As I said in reply to David (q.v.), who beat your question to me by 11 seconds, I should have said "centuries", and I could have said "millennia", not "decades". Robin Patterson 23:56, 21 Nov 2004 (UTC)

*chuckle*, yes, but I at least have the claim to have put a question on your "User:" page before David's comment! Anyway, I've had a stab at rewording it. — Matt 01:48, 22 Nov 2004 (UTC)

Screenshot

I removed the Pan screenshot. Am I the only one who thinks this is picturitis, and the screenshot adds nothing to the article? All it basically says is that newsreaders often have support for it, which is mentioned in the article; showing a menu with the option in it is overdoing it. JRM 12:51, 2004 Nov 23 (UTC)

Hmm, no — you're probably right. (I added it after wracking my brains for ideas after someone nominated it on WP:FAC and there was a request for illustration.) — Matt 13:53, 23 Nov 2004 (UTC)

I suspected as much ("hey, if it's a FAC, it must have A Picture (tm)! Why doesn't it have A Picture (tm)? It definitely needs A Picture (tm)!" :-) Some day, there'll be a FAC that has no images and to which no images can be added to improve it, and that'll make for some heated tangential discussion... JRM 14:14, 2004 Nov 23 (UTC)

No, the proceduralists would win the "debate" with a thumping majority, and there is no way that that hypothetical FAC would be become a FA. Pcb21| Pete 14:25, 23 Nov 2004 (UTC)