Talk:Disk storage

Survey

Data are or Data is?

• Data is. As nominator, per sources for this subject and field, per reader expectations and their ease of reading, and per #MOS:STYLERET in Threaded Discussion. --A D Monroe III^(talk) 16:37, 27 October 2017 (UTC)
• Data are. This appears to be the convention in the article (at present). In arbitrary grammatical matters like this, I would say that clarity is most important, then after that convention. Is the article presently unclear in some way because "data are"? Not that I can see, so why change it? Attic Salt (talk) 17:30, 27 October 2017 (UTC)
• Data are is acceptable and should not be changed in this article per #MOS:STYLERET in Threaded Discussion; it is a matter of style.Tom94022 (talk) 18:07, 27 October 2017 (UTC)
• Data is WP is a general purpose resource and the more common general usage is a) quite adequate b) endorsed by Oxford English Dictionaries. There are many instances within the article that minor rephrasing would 'side-step' the problem. Pincrete (talk) 17:01, 28 October 2017 (UTC)
• Avoid the problem, by using words such as "information" when the collective meaning is intended. I don't actually see any cases in this article where such a change is needed, but in the example "SSDs hold fewer data" I'd change to ""SSDs hold less information" perhaps. When talking about transferring bytes, blocks, sectors, etc., the plural data is good. Dicklyon (talk) 16:46, 29 October 2017 (UTC)
• Data is. "Data", as used in the article is a mass noun. There's no justifiable reason for treating it as countable here. Data is to bytes and bits as money is to dollars and cents (or pick your favorite currency). There are certainly circumstances where it makes some sense to treat it as plural ("The questionnaire asks for several data about the subject, such as name, age, birth place..."), but this isn't one of them. It would make no sense to say that a disk holds 27 data unless, say, the data on the disk is organized into 27 pieces of information, each one of which is one datum. Also, changing to "information" to avoid the issue is a cop-out and artificially avoids a good word. --Deacon Vorbis (talk) 17:40, 29 October 2017 (UTC)
• Data is - I don't think we could reasonably say "data are" here. It just sounds odd, because it isn't something that is really countable. For example, you wouldn't know what I am saying if I said 100 kilos of data are. That's because "data" isn't a valid unit. Thus, it should be treated as a mass noun. RileyBugz^会話_投稿記録 20:13, 29 October 2017 (UTC)
• Decide on a case-by-case basis according to context. "All of that data is now stored on the hard disk" and "These three specific pieces of data are now stored on the hard disk" are both correct. --Guy Macon (talk) 22:08, 29 October 2017 (UTC)

  Sure, but with "pieces of data", it's "pieces" that's plural, not "data". --Deacon Vorbis (talk) 22:12, 29 October 2017 (UTC)

  That's what "Decide on a case-by-case basis according to context" means. I gave grammatically correct examples of "data is" and "data are". This proves that no "always use 'data is' " or "always use 'data are' " rule is needed or appropriate. --Guy Macon (talk) 21:49, 6 November 2017 (UTC)

  No, you gave an example of "pieces of data are", not "data are"; that's why I said that in your example, that "pieces" is what's plural. You can't really have pieces of a countable noun anyway (not that I can think of at least). Read the rest of the discussion and see how it's being used in the article. No one is saying that anyone should ever use "pieces of data is". And for the most part, I don't think people are advocating for always using "data is"; I'm sure not. I even said that there are times when "data are" is appropriate – just not for how it's being used in this article, which is what's under discussion here. (Also, see MOS:LISTGAP; the way I had the indenting was correct already.) --Deacon Vorbis (talk) 22:51, 6 November 2017 (UTC)

  You may very well not be advocating for always using "data is" or always using "data are", but the question at the top of this RfC is, and if one of those two choices get enough !votes (including your) that will be what will be enforced as being the consensus of this RfC. -Guy Macon (talk) 15:24, 7 November 2017 (UTC)

  Let's move this to #Clarification on grammar vs. data is/are !votes under Threaded discussion below. --A D Monroe III^(talk) 00:29, 8 November 2017 (UTC)

• It depends on context. To use clearer examples, "The project's data is backed up off-site" (mass noun – data treated as something dumped into a container); "the project's data were subject to peer review" (count noun – each datum was examined). Dicklyon is also correct in that it can simply be written around, anyway. The average case in this article's text appears to call for "data is", because it's talking about data in the abstract, as virtual piles of stuff being stored and moved around on volumes, and is not about any specific datum or set thereof.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  01:57, 30 October 2017 (UTC)

The context here is all under Data storage, so would be "data is" throughout based on this reasoning. --A D Monroe III^(talk) 22:13, 2 November 2017 (UTC)

• Data is is correct. Summoned by bot. If it seems awkward to anybody and doesn’t make it more awkward by using another word then that seems like a reasonable compromise. TimTempleton ^{(talk) (cont)} 15:56, 30 October 2017 (UTC)
• Both - depending on context. Usually 'data is' seems likely, but a 'data are' usage can also be correct. Markbassett (talk) 19:38, 2 November 2017 (UTC)

Sigh. I think the only 100% agreement we had was to not mix them. --A D Monroe III^(talk) 22:08, 2 November 2017 (UTC)

• Data is While both can be grammatically correct, "data is" is the more common usage of the two. While WP:COMMONNAME doesn't apply here (it only applies to article titles), I think we should apply the general concept here: use the most commonly recognized phrasing. Also, using both phrasings in the same article can lead to confusion, especially for those who aren't as computer-literate. Even those who are computer-literate may be confused by us using both or even just "data are" due to the latter's rarer usage. — Gestrid (talk) 14:56, 6 November 2017 (UTC)

• By !voting for "data is" instead of "choose according to context", you are proposing a rule that would call for changing the sentence "These three specific pieces of data are now stored on the hard disk" to "These three specific pieces of data is now stored on the hard disk". Is that what you intended to do? --Guy Macon (talk) 21:56, 6 November 2017 (UTC)

Incorrect. "Pieces of data" is a plural noun phrase, like "slices of apple" or "piles of sand". "Context" like this would still follow standard English grammar, of course. --A D Monroe III^(talk) 15:05, 7 November 2017 (UTC)

On what basis do you make that claim? It certainly isn't in the question asked at the top of this RfC, which makes zero mention of context. If you make a rule, people will enforce the rule as written. --Guy Macon (talk) 15:24, 7 November 2017 (UTC)

Sigh. In this article, should "data" be used as a count noun (the plural of "datum") or mass noun (like "information")? I'm sorry if my is/are simplification just for !voting somehow overrides the RfC statement as written. --A D Monroe III^(talk) 23:38, 7 November 2017 (UTC)

Let's move this to #Clarification on grammar vs. data is/are !votes under Threaded discussion below. --A D Monroe III^(talk) 00:29, 8 November 2017 (UTC)

• Data is This is common and quite adequate because I see no compelling need ever to use "Data are". If one needs to refer to a plurality of data, then be descriptive about it as in "Data bytes flow here", "Data bits are read sequentially", etc. or "The three data streams enter the device." Blooteuth (talk) 14:01, 17 November 2017 (UTC)
• Data are. In evaluating the Wiki guideline, please keep in mind that data is also jars, even for readers who don't know Latin. Shmuel (Seymour J.) Metz Username:Chatul (talk) 18:09, 20 November 2017 (UTC)

  Any basis for that? I've seen no "data are" preference outside of those focused on Latin, which is outside this subject. --A D Monroe III^(talk) 21:02, 20 November 2017 (UTC)

• Data is It boils down to ILIKEIT per what I'm used to, but decades of examining computer manuals and books means "data are" looks pretentious. Data is stored on disk or data is transferred over a cable, just like water is (not are) transferred through a pipe. Johnuniq (talk) 21:25, 20 November 2017 (UTC)
• Data is as a simple !vote jcc (tea and biscuits) 12:03, 3 December 2017 (UTC)

Threaded discussion

First, it's agreed that both "data are" and "data is" are currently used in English, so neither is "incorrect" (similar to "disc" vs. "disk"). Also note that this article currently uses "data are".

It can be noted that "data is" is the more favored use in modern dictionaries and in general English, but it's more important to note that these two uses are far from equally distributed.

The etymological origin of "data" is Latin, where it was the plural of the Latin "datum". This tends to be maintained in fields where knowledge of Latin is more common, like "hard" science and research where publishing in scientific journals is the goal. Here, using the word "datum" occurs, and "data" tends to be maintained as countable and the plural of "datum".

But outside of this, English use is otherwise. In newer "engineering" fields, where the the only end-target is products for common consumers, "data is" dominates: industry sources, consumer articles, and user manuals. Here, "data" a mass noun, used just like "information", and "datum" is virtually absent.

Thus the common reader for this subject will be used to, and expecting, "data is". When reading the article and coming across "data are", this can cause the reader to stop and wonder why our WP editor would choose that; is there some implied significance to this they've missed? It's an interruption, just like coming across "colour" in an article about an American film, or "BC" in an article on Judaism, or "phone it in" within an article about Shakespeare. Unusual wording just distracts the reader, with no benefit whatsoever. Since "data are" and "data is" are unequally correct English, we should simply go with the one more appropriate for the reader. In a few articles this choice may be difficult -- not here. --A D Monroe III^(talk) 16:37, 27 October 2017 (UTC)

It's not actually similar to "disc" vs. "disk", which in this context have different referents (disc = optical and magneto-optical media) It's not okay to write "hard disc" and "compact disk" here.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  01:59, 30 October 2017 (UTC)

Ah, but it was similar -- exactly my point. Seagate Technology, especially, used to insist that its magnetic media HDD products were "hard disc drives", and sources often did the same. Back then, WP could use either "disc" or "disk" for the same subjects. But English use evolved, and use of disc diminished to only optical, and uniformly disk for magnetic, and WP followed suit. Thus similar to "data are/is" use here: terms used to be equal, but now splitting to be area-specific. --A D Monroe III^(talk) 16:02, 30 October 2017 (UTC)

Data are is acceptable and should not be changed in this article. A D Monroe III has presented no substantial reasoning that "data are" is "unequally correct English" nor that "data is" "dominates" for encyclopedic works. Arguably this is a question of style in which case we should retain the existing style. My recollection is that this is an old issue which as best I can recall has been resolved by consistent usage in any one article of either style, that is, editors choice. Note also that there are 14,631 instances of "data are" in the English Wikipedia articles so this discussion is more suitable to MOS talk than this one article's talk. I do not understand Monroe's reasoning that this change only applies to this and "closely related articles" - it should be taken to MOS talk. Tom94022 (talk) 18:02, 27 October 2017 (UTC)

I said "dictionaries", not "encyclopedic works".

So, maybe vaguely something was said about this before? Languages and WP:Consensus can change. Similar to how "disc" changed to "disk" a way back.

Yes, there's a lot of "data are" in WP, and a lot of "data is". So, again, neither is "incorrect".

As clearly stated, this RfC isn't WP-wide; why say we should try to do that at MOS? It would be wrong if all of WP was changed to only one of either "data are" or "data is" uniformly across all articles. Each article should be judged based on its own sources and readers. Articles on scientific subjects, where use of "datum" is prevalent, should probably adopt "data are". In fact, this RfC may help clear the way for doing that in those articles, where they currently have "data is" even though contrary to the subject's sources, and it's disruptive to its readers. --A D Monroe III^(talk) 22:52, 27 October 2017 (UTC)

Oxford Eng Dict, says unequivocally that it is a mass noun. Longman says mass noun in general speech, but less so in academic and formal Eng. I don't see need or benefit in this article for academic or scientific-exactness. Pincrete (talk) 16:57, 28 October 2017 (UTC)

Perhaps the Oxford English Dictionary is equivocal:

This article is consistent, there is no explicit policy and the author of this RfC doesn't want to take it to the MOS talk where a policy could be developed. It would seem the Oxford blog suggests leaving the article alone. Tom94022 (talk) 19:02, 28 October 2017 (UTC)

The same OED blog continues "The word data in English usage has evolved: a mass noun use, recorded in the OED from the 18th century, has become increasingly common over the past 70 years, particularly in computing and general contexts." I think as both uses are recognised as valid, the case needs to be made for deviating from common use in a "computing or general context". Pincrete (talk) 19:45, 2 November 2017 (UTC)

This article is inconsistent per that very blog summary. This subject under the subject called "Data storage" (mass noun), not "Datum storage" (count noun, where singular is used for count noun collections in English). There is no debate on "data is" in WP. There's not even a debate in the industry, because it's overwhelmingly "data is". I do not favor any sort of one-rule-fits-all in MOS for this. Instead, for the readers' sake, I want articles with subjects that use data as a count noun to use it as a count noun, and articles with subjects that use data as a mass noun to use it as a mass noun. --A D Monroe III^(talk) 15:46, 29 October 2017 (UTC)

BTW, For completeness, I've asked on MOS talk if MOS should decide this and by what means. We can see if anyone there is interested. --A D Monroe III^(talk) 16:00, 29 October 2017 (UTC)

The blog at OxfordDictionaries.com isn't a particularly reliable source, but even if it were, what Oxford U. Pr. advises as a style "rule" has nothing to do with what WP does, unless MoS adopts the same rule. And MoS would never adopt that one, because it's "context-stupid" and defies real-world usage. It's no different from ignoring the fact that "fish" and "fishes" are both valid plurals in different contexts ("the aquarium has 7 fish" refers to 7 animals; "... has 7 fishes" refers to 7 species and maybe way more than 7 animals), and insisting on always using one or other ("the aquarium has 7 fish, meaning species, and 247 fish, meaning individual animals", or worse yet, "I ate two fishes for dinner; both were tilapia").  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  02:05, 30 October 2017 (UTC)

Ah now I understand The Feeding of the 5000! Pincrete (talk) 19:39, 2 November 2017 (UTC)

There's something that may be of interest here: a user essay called the "Yogurt Principle". It covers the long sad story of the 8 year ugly battle to move Yoghurt to Yogurt, which ended only when the move finally succeeded. The "principle" it proposes is to suggest asking the question, "if a proposed change were done, would anyone later bother to try and change it back"? It could be asked here: if this article was changed to "data is", would there be any more drive-by edits to change it back to "data are"? I really doubt it. Changing can end the current drive-by edits (see article edit history), and all further discussion. The only arguments for "data are" here are based on "let's just keep it the way it is". If changed, those same arguments would then join the consensus for "data is", making it overwhelming and permanent. --A D Monroe III^(talk) 20:32, 29 October 2017 (UTC)

That's B2C's way of saying that he's OK with tyranny of the majority. I'm not so sure it applies here, as there are lots of editors who will generally try to fix treatment of data as singular. Dicklyon (talk) 21:45, 29 October 2017 (UTC)

Not quite. It's not saying "51%" is good enough", it's suggesting that for !votes that are only "just leave it be", with absolutely no preference for X or Y, that if the article was changed from X to Y, those !votes would then become votes in favor of Y. Their !vote lacking any preference may be more "let's spend as little time on this as possible", which might best be served by making the change to the most stable way and be done with it for good. That would have to be very carefully judged by the closer, of course.

On "lots of editors" for "data are", the evidence seems strongly against that. In the Data storage WP subjects, articles that use "data is" get little or no drive-by edits changing to "data are", while ones that use "data are" (like this one) get repeated drive-bys for "data is"; apparently it seems obviously wrong to them. Now, I don't have counts of any of these, searching change logs for specific changes is tedious, but this is what my experience tells me.

(I have moved the following from MOS Talk as more appropriate here. --A D Monroe III^(talk) 01:49, 11 November 2017 (UTC))

With counting of "data is" vs "data are" being presented as evidence for some WP preference, I should point out that's not evidence for any preference of data as a plural of datum vs. data as a mass noun. In phrases like "different types of data are used", the subject is "types", not "data", so "are" is used regardless of data being a mass noun. There are many other indicators of count vs. mass, such as "fewer data" vs. "less data". English being what it is, the varieties pertinent variations on this are virtually endless. A better indicator (though still not perfect) would be counting the occurrence of "data" vs. "datum" in articles; in general, any article that uses "data" as a plural is at least somewhat likely to use its singular form as well, while articles using it as a mass noun are very unlikely to use "datum". --A D Monroe III^(talk) 15:03, 10 November 2017 (UTC)

Search shows 190 instances of "of data are" in WP articles so Monroe's example aboove really doesn't illustrate much - regardless of the possible miscounting or mischaracterization by my somewhat crude search the fact is that there are thousands of ambiguous usages of these terms in WP articles which makes it a matter of style covered by MOS:STYLERET MOS:RETAIN which clearly states:

The author of this RfC has stated neither "data is" nor "data are" is incorrect but he prefers "data is" "per reader expectations and their ease of reading." which is his POV about style that he is entitled to but is not supported by the actual usage in WP. Personaly my POV is although I use "data are" this is a writing style issue and therefore we must apply the MOS guidance to retain existing styles. The reasoning hereinabove is POV, none of which amounts to a substantial reason to modify this article given the thousands of articles in WP using "data are".

BTW there is ample WP:RS reliable sources that this is a style question and other style guides do address it. Also see Oxford dictionay blog quoted above. So unless someone comes up with a substantial reason this article should remain unchanged. Tom94022 (talk) 20:36, 11 November 2017 (UTC)

Since this thread has been switched from counting "data are" other articles (with one last attempt at cooked counting) to MOS:RETAIN, I'll continue with that. See MOS:RETAIN #MOS:STYLERET below.

Close this discussion

In the absence of consensus this discussion should be closed without changing the article.

An opinion at MOS on this specific subject is that this discussion falls under Retaining Existing Styles which states, "... editors should not change an article from one styling to another without "substantial reason" " I agree.

There are about 30,000 WP articles that use one form or the other of data are/data is. In these articles WP editors use "data are" over "data is" by a ratio greater than 4/3 so that actual style used by actual WP editors demonstrates there there is no "substantial reason" for a change. The best unsubstantial reason raised above is that "data is" is "more common" in general works. It is not relevant in WP where editors prefer "data are;" making WP more of a technical publication where "data are" is more common. Either is acceptable.

I would note that this has been a very stable article and I am the only editor participating in this discussion who contributed to the article. It is one of about 17,000 WP articles exclusively using "data are." So I ask why put all this effort into imposing a rule in this one article and oppose taking it to MOS? There is nothing about its usage herein that is any different than its usage in the other 17,000 article! If this is not a question of retaining an existing style then shouldn't it be discussed at MOS? If it is a retention question then the discussion should be closed with the article unchanged. Tom94022 (talk) 18:56, 8 November 2017 (UTC)

The RfC exists to establish a substantial reason to change; common use and reader expectation for this subject.

This RfC isn't for all of WP; what other articles do isn't relevant.

Claims that verge on WP:OWN are not relevant.

MOS is a dead horse here. --A D Monroe III^(talk) 02:49, 9 November 2017 (UTC)

The only valid reason to close down an RfC before 30 days is WP:SNOW, and the votes are not one-sided enough to invoke SNOW. Closing down RfCs early is undesirable, because a significant number of people take two-week vacations and we don't want them to come back and discover that there was an RfC that they were not able to comment on. Let it run.

BTW, a careful reading of WP:LOCALCON may help some of the editors who are participating in this discussion. --Guy Macon (talk) 06:04, 9 November 2017 (UTC) --Guy Macon (talk) 06:04, 9 November 2017 (UTC)