Wikipedia:Articles for deletion/Cycles per byte


 * The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review).  No further edits should be made to this page.

The result was merge to Encryption software &mdash; Amakuru (talk) 10:35, 20 September 2019 (UTC)

Cycles per byte

 * – ( View AfD View log  Stats )

This is a dictionary definition of a metric used to evaluate the performance of cryptographic software. Though the metric is frequently used in the field, it seems unlikely that the article can ever grow to more than a dictionary definition: sources explaining the history, significance, or usage of "cycles per byte" seem lacking. BenKuykendall (talk) 15:24, 4 September 2019 (UTC)
 * Note: This discussion has been included in the list of Science-related deletion discussions. BenKuykendall (talk) 15:24, 4 September 2019 (UTC)
 * Note: This discussion has been included in the list of Computing-related deletion discussions. BenKuykendall (talk) 15:24, 4 September 2019 (UTC)


 * Keep Per WP:DICDEF, this is the "perennial source of confusion is that a stub encyclopedia article looks very much like a dictionary entry, and stubs are often poorly written; another is that some paper dictionaries, such as "pocket" dictionaries, lead users to the mistaken belief that dictionary entries are short, and that short article and dictionary entry are therefore equivalent." The stub can quite readily be expanded from sources such as Cryptography for Developers''. Andrew D. (talk) 19:01, 4 September 2019 (UTC)
 * It only necessarily looks like a dictionary entry to someone who knows nothing about the topic. The nominator clearly addressed this. Imagine one of your favourite "rescued" articles, perhaps the "clurn" [fictional example, ok?], a measure of cloth used in Norfolk in the 14th century. Someone creates Clurns per fortnight, so you argue this can be easily expended[??] from "Hiring clothworkers in 14th-century Norfolk", by Lucian B. Withers, 1873. I think your submission is simply invalid, being based on a logical fallacy. Imaginatorium (talk) 19:30, 4 September 2019 (UTC)
 * A fictional straw man is unhelpful. The nominator's difficulty seems to be that they can't envisage how the current stub might be expanded.  If we consider an actual example of a similar measure – instructions per second – we see that it has a table of processor speeds, expressed in that unit and given as a timeline.  The relevant feature of the source which I cited is that it gives a series of examples which are likewise expressed in the unit in question.  For example, it says that "SHA-512 requires 12 cycles per byte...".  Such examples might usefully be added to expand the article.  My !vote stands. Andrew D. (talk) 23:22, 4 September 2019 (UTC)

 Relisted to generate a more thorough discussion and clearer consensus.
 * Delete: as per nom, this is a unit that can be used in cryptography. There is lots and lots to say about cryptography, and possibly even Cryptographic processing rate, but nothing to say about an obvious measurement other than dicdef. Imaginatorium (talk) 19:30, 4 September 2019 (UTC)
 * Merge - (to Encryption software) there are dozens of books, news pieces and journal articles that have a couple of lines on the topic. This isn't enough to warrant an article as we never reach Sig Cov. However, WP:ATD applies, and would seem relevant to that article. Nosebagbear (talk) 09:54, 5 September 2019 (UTC)
 * Merge to any of a number of cryptography articles. This is a frequently used term, but practically all related material is much more relevant to specific applications than to the measure itself. (This, by the way, is nicely demonstrated by the book proffered above, which I doubt Andrew looked at closely outside of searching for occurrence of the words...) Being a recent invention, it doesn't have an interesting history either, as do many units. In it's current state, it doesn't merit an article. Merge and split later, if lots of content emerges contrary to expectations. -- Elmidae (talk · contribs) 12:55, 5 September 2019 (UTC)

Please add new comments below this notice. Thanks,  Sandstein   18:22, 11 September 2019 (UTC)
 * Merge I find Elmidae's analysis persuasive. It's documentable and worth documenting, but not with a stand-alone page. XOR&#39;easter (talk) 01:46, 17 September 2019 (UTC)
 * Merge As stands there is insufficient content beyond the definition itself. No harm to split down the track if more content on the topic is incorporated. pinchies (talk) 12:45, 17 September 2019 (UTC)
 * Comment: re the merges did you have a redirect target in mind? Nosebagbear recommended Encryption software; we could even make a new section Encryption software. Alternatively, could redirect to Algorithmic_efficiency. Preferences? Other ideas? BenKuykendall (talk) 14:39, 18 September 2019 (UTC)
 * Encryption software (probably with a new section) was what occurred to me, but I'm distinctly a non-technical expert. If algorithmic efficiency seems more appropriate I'm happy to go with that. Nosebagbear (talk) 14:42, 18 September 2019 (UTC)


 * Merge to Encryption software. There is some crypto-specific interest here, which would be lost in an article about general algorithmic complexity.  There are cryptographic attacks which depend on observing how long it takes to encrypt/decrypt something.  For example, if you can observe that decrypting two strings take the same amount of clock cycles, even if you can't see the inputs, you may be able to deduce that they're identical.  Of course, anything we say in Encryption software needs to be vetted against WP:RS, not the random ramblings of a wikieditor who knows just enough about crypto to be dangerous.  -- RoySmith (talk) 14:09, 19 September 2019 (UTC)


 * The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.