Talk:Ƒ

Untitled

I removed this:

Font families that support languages using the African reference alphabet will have the minuscule f with a hook in its straigt form in the regular font.

Could someone make this clearer? The African reference alphabet contains both f and ƒ. EdC 23:55, 22 May 2006 (UTC)

ƒ —Preceding unsigned comment added by 83.18.76.220 (talk) 10:05, 19 July 2008 (UTC)

ƒ as a symbol for f-number

I suppose the statement is literally correct given the number of hits, but the citation isn't a verifiable reliable source. More important, the use of ƒ for f-number is in marked contrast to practice in professionally published verifiable sources. Consequently, I've restored the flag. If the statement remains, it should be qualified by mention of the disparity with authoritative sources.

The issue is under discussion in much greater detail at Template talk:F/.

The number of Google hits is a meaningless statistic; see FGA: Google result counts are a meaningless metric for more details. And anyway, the cited Google search (ƒ aperture) returns only 11,100 results (currently), whereas "f" aperture returns 46,100,000 results, which would seem to indicate that the symbol "ƒ" is used less than 2.5 hundredths of one percent of the time, hardly what I would consider "commonly used". —Bkell (talk) 07:22, 9 January 2009 (UTC)

I keep forgetting that Google searches only match whole words, even those of a single letter (except perhaps for a and I). My first search using ƒ got almost 100,000 hits ... repeating it got only 11,100—yet additional evidence that the number of Google search hits isn't a reliable source for WP. An additional clue is provided by this article's category: Uncommon Latin letters.

Perhaps the statement could say “occasionally”, but I would think a reference to the far more common usage would still be needed. Because we'd be saying, in effect, that the usage is wrong, I'm not sure I see the point of even including this section. Although it's just speculation, I think the usage on the pages Google finds is by folks who don't really understand what they're doing by using it.

Again, I think it would probably be better to remove the photography section than to fill it with caveats. JeffConrad (talk) 10:06, 9 January 2009 (UTC)

Yes, I agree. A few months ago I removed a section that said ƒ is used in mathematics to indicate a function (it's actually just an italic f); I felt confident removing that section because of my mathematical background. I have no photography background, though, so I didn't feel confident enough to unilaterally remove the photography section. —Bkell (talk) 11:42, 9 January 2009 (UTC)

The problem with the function probably arises from the HTML entity ƒ. Why they chose this name instead of florin (as it was described in most extended-ASCII character sets) is beyond me. If ƒ were the right character, what would one do for g(x), h(x), and so on?

I've fixed a minor error in this section, but unless someone comes up with a good reason for the section, I'll probably remove it. JeffConrad (talk) 22:49, 9 January 2009 (UTC)

I think that it's useful because some user might copy/paste the symbol from a photo site, and arrive at the wiki page about the symbol. That wiki page should then direct them to f-number or aperture or a similar article. I don't care for "commonly", necessarily, but there's no point in taking out the reference, which stands on its own to substantiate the claim that the symbol is in fact used for the purpose described. --jacobolus (t) 22:30, 11 January 2009 (UTC)

To clarify, Wikipedia is about describing the world, not proscribing best practices. It is possible to acknowledge that many people use ƒ for f-numbers, without suggesting that such use is "correct". --jacobolus (t) 22:35, 11 January 2009 (UTC)

Perhaps the section could remain using “occasionally” or “sometimes”. But the statement, though probably correct, would seem disingenuous without mention that simply using an italic f is far more common. And even with the qualifier, I think it would be marginal. Wikipedia is indeed descriptive rather than prescriptive, but it nonetheless requires reliable sources. Absent some other support, it's difficult to see a Google search meeting the criteria of WP:V or WP:RS, and being other than WP:OR.

I'm also hesitant to rely on web typographical practice. Many of the pages that use ƒ contain enough typographical errors to immediately impeach them as sources; though most of the errors are minor enough to have minimal impact on readability, the typography on a page should be of a very high standard to qualify as a reference. Anyone can do almost anything on a web page, and I don't think that someone does it on a web page merits inclusion in a WP article. Imagine this as an acceptable criterion for supporting factual statements; almost any claim could be made, and WP would rapidly degenerate into utter garbage. JeffConrad (talk) 04:07, 12 January 2009 (UTC)

I made some edits that I don't think will be too controversial, but I am about to restore the google reference with a {{unreliable source}} tag. I've changed the wording to indicate that the use of the florin symbol for f-number is rare, and that an italic f is the more common notation. The google source, while possibly not satisfying WP:RS and WP:OR, is sufficient to establish that this usage does in fact occur. It's better than nothing in the absence of a true reliable source. Note also the essay These are not original research, which argues that some types of straightforward, verifiable analysis of primary sources should be acceptable. The underlying fact here is not really in dispute: despite the wishes of some editors, it is clear that this symbol is occasionally used for f-number. If the usage is by unprofessional publishers, that's fine and perhaps that could be mentioned, but the usage itself cannot really be disputed in good faith. --Srleffler (talk) 06:00, 15 January 2009 (UTC)

I don't dispute the usage, and don't have a big problem with mentioning rare usage (I proposed it), but it's questionable nonetheless in light of the extremely infrequent usage (1 in 4200 if we trust the search results; recall that I got 5 times as many hits for vlave). It's also not clear from the article that a Google search isn't original research. I do take issue with including a serif f in the sans-serif text because there is an implication that it is something special without any support for that idea despite almost six months of requesting it. Mixing typefaces in running text just isn't done without good reason; if there is such a reason in this context, please show it.

I've once again changed the f-number WL to aperture; as I have stated, the f-number is the denominator of the expression, not the numerator or the entire expression itself, as Sidney Ray states in the Manual of Photography, 9th ed., p. 62. I realize the article is about the opening itself, but the expression f/# is the aperture. I'm open to an alternative description that's technically correct. JeffConrad (talk) 08:07, 15 January 2009 (UTC)

Srleffler, you are correct that there do exist a handful of instances in which someone on the Internet has used ƒ to represent f-number. I don't think anyone is disputing that. However, the question is whether this is in any way significant enough to be included in a Wikipedia article. There appear to be no reliable sources that recommend or even mention this usage; the only source we have is a Google search that appears at first sight to produce about 10,000 results (but actually far fewer; as I have said several times, the number of Google search results cannot be considered reliable in any sense of the word). The mere fact that a certain usage appears on only a few hundred random Web pages does not seem significant in any way. For comparison, a Google search for Ł pound also returns some results—should we also mention in Ł that it is sometimes used to represent the pound sterling? Of course not, because realistically it isn't. —Bkell (talk) 13:15, 15 January 2009 (UTC)

"ƒ" is a rarely-used symbol to begin with. Should we have an article on it at all? If so, surely uses of that symbol are worth mentioning, even if the number is not enormous. Usage of ƒ in "ƒ/#" is rare compared to "f/#", but may not be negligible compared to total uses of ƒ (ignoring character set listings). If there were only a couple dozen pages using it, I would agree with you, but there are a lot more than that. Yes, the usage of Ł as a substitute for £ should be mentioned at Ł; at the least there should be "see also" links between £ and Ł.

I notice that the Optical Society of America, a major publisher of optics journals, uses ƒ for "ƒ/#" in article titles and abstracts, on their website.[1] I checked a half-dozen or so of the full-text articles and in every case an italic f was used in the full-text pdf. It appears that they are deliberately substituting "ƒ" for an italic f when they convert the pdf text to html, presumably to preserve the hooked appearance of the italic f, even though the web pages use a sans-serif font.--Srleffler (talk) 18:22, 15 January 2009 (UTC)

Another example: I have seen, in real life on menus and things, the "not a subset" symbol ⊄ used to represent "cents." Should we mention this usage in Subset#SYMBOLS? No, of course not, because it's a mistake. It was used by someone who had no typographic knowledge because he or she found the symbol, didn't know what it really meant, and thought it looked similar to a cents sign. Now, I can't perform a Google search to find Web pages that use ⊄ to mean "cents," because of Google's technical idiosyncrasies (the symbol ⊄ is ignored by Google), but even supposing I could, this very rare and erroneous usage doesn't deserve a mention in a Wikipedia article. We do not aim to list everything anyone has ever done. —Bkell (talk) 13:26, 15 January 2009 (UTC)

If the erroneous usage is common enough, we should document it. If it's not so common, it may not be worth mentioning. Again, at the least there should be a see also link between the articles due to the visual similarity of the glyphs.--Srleffler (talk) 18:22, 15 January 2009 (UTC)

In a sense, I agree with you both, though I'm more with Brian. If we seek to document everything, we may open a can of worms that cannot be resealed. For example, “vlave” is a common misspelling of valve, and if we rely on a Google search, it's far more common than “ƒ/”; should we therefore mention it? And what of ƒ(x)? Damn, where are the hooked g and h when I need them ...

As for this article's justification, the long-standing use of ƒ as a currency symbol would suffice. The mere inclusion of the symbol in Unicode might also suffice. JeffConrad (talk) 00:03, 16 January 2009 (UTC)

"If the erroneous usage is common enough, we should document it." No, we shouldn't. Wikipedia is not an indiscriminate collection of information. There is a threshold for inclusion in Wikipedia, and that threshold is not measured in Google hits, but by having received significant coverage in reliable sources that are independent of the subject. Has the usage of ƒ to represent f-number specifically received any coverage at all from reliable sources? So far no one has provided any evidence that it has. So it doesn't meet the threshold for inclusion in Wikipedia. We are not in a position make the subjective decision that 400 to 600 Google results means that it's notable or "common enough"; that would be original research (and again I repeat that Google results are absolutely the furthest thing from a reliable source). If it really is common, and if it really is notable, then someone should have written about it outside Wikipedia, thus providing us with a third-party source. But no one has, so it's not notable, and it doesn't belong here. —Bkell (talk) 00:51, 17 January 2009 (UTC)

Coverage in secondary sources is not the only possibility. Direct observation of usage of the symbol in primary sources also suffices. (We can't do analysis on primary sources, but simple observation is permitted.) Of course the sources have to be reliable, but we have an abundance of such sources. Each of the web abstracts of articles in OSA journals is a reliable source. A nice, citable reliable source with a DOI, in fact. Feel free to replace the Google search reference with a half dozen references to OSA-published articles, if you wish. I'm pretty sure I saw an IEEE publication and a couple of magazines on the list too.--Srleffler (talk) 04:44, 17 January 2009 (UTC)

I'd have a problem with reference to OSA-published articles when only the online abstracts use the symbol ƒ and f is used in the articles, especially since f is the conventional symbol in most cases (e.g., “gf-value”). This sure looks like a typographical malaprop; I obviously can't be sure without asking, but assuming that the persons responsible for the web page fully understood this seems shaky. JeffConrad (talk) 05:11, 17 January 2009 (UTC)

Okay, just so we don't have to keep guessing how many sources Google gives, I've looked at all of them. See User:Bkell/ƒ aperture. Essentially Google gives about 141 sources that use ƒ in f-number, f-stop, or the f/ notation, though not all of them do so consistently, and I don't think I'd classify any of them as reliable sources. The most reliable pro-ƒ sources seem to be Apple Aperture (since Apple generally seems to be reasonably careful with its typography—more so than most other software developers, anyway) and a single Kodak page, but even those are sketchy as sources and I wouldn't feel comfortable referencing them in a Wikipedia article to back up a claim that the usage of ƒ in f-number is even remotely common. The fact that Google produces only 141 instances of this usage from its entire database, which contains over a trillion pages [2], seems to imply that this usage is exceedingly rare.

I propose a compromise. According to the official Unicode code chart [3], the character ƒ can also represent "script f". This provides a reliable, third-party source justifying its use when a script f is needed, so we can give as examples the use of ƒ to denote a mathematical function and the florin currency (for which we can also cite [4]) and f-numbers. We should not imply that the character ƒ is the only choice, the correct choice, or even a common choice for these, but I would be fine with mentioning it as a possibility, which seems to be in accordance with the official Unicode position. What do you think? —Bkell (talk) 07:19, 17 January 2009 (UTC)

I don't have a big problem with a mention, as I've indicated several times. I still think it's marginal, though, and ask a very basic question: what justifies the use of a script f? This is not a common typographical practice.

It now seems obvious that Outdoor Photographer intentionally uses U+0192 in place of f in printed articles that are sans-serif type; the a in the base typeface is non-cursive, and I'm unaware of any true serif font that has a cursive f but not a cursive a (a few other characters also significantly change appearance in true italic fonts). OP also use “ƒ-stop” in sans-serif type, so at least they're consistent. I'm not sure this really has any bearing on the issue, though.

Additionally, I still take issue with including a serif f in addition to one that matches the rest of the text without some explanation, because doing so is likely to confuse content and appearance. The explanation probably could be as simple as “(or f when the text is in a serif typeface)”; a longer explanation would probably be better in a note. JeffConrad (talk) 07:51, 17 January 2009 (UTC)

I somehow forgot to mention: nice job reducing the data, Brian. JeffConrad (talk) 08:44, 17 January 2009 (UTC)

I like Bkell's suggestion. The Unicode chart provides a reliable source that supports the use of the symbol a script f or as a symbol for function, so we should mention that. I certainly agree that the article should not imply that the hooked f is the only choice, a correct choice, or a common choice for either of these, or for f/#. It is clearly none of those. --Srleffler (talk) 16:11, 17 January 2009 (UTC)

I also agree on mentioning use as an alternative to f(x), though it sure would be nice to see some examples. I think great care is indicated in mentioning use in place of f/. There obviously are a few people that do it; OSA are probably as reliable source as one can find, but as I've said, I find it a bit disturbing that practice on the web site (or perhaps simply for sans-serif type) differs from that in print (or perhaps serif type). In setting for appearance, which is what the web site seems to do, it's almost the same as saying lower-case ell and one are interchangeable. The OSA abstracts also seem to use ƒ whenever f is used as a quantity symbol, so if the OSA web site is cited, this general usage should be mentioned lest the citation give the mistaken impression that the somewhat unusual treatment is reserved for f/. I don't see how this could be done without a substantive footnote, but perhaps the best approach is to avoid the citation, and if objections come up, direct the objectors to this discussion, and if that doesn't suffice, then figure out an appropriate citation. If some sort of note is added, OP might also be mentioned. They aren't in the same league as OSA, but they're far more credible than most of the sites using ƒ, and they are consistent between online and print practice. JeffConrad (talk) 01:58, 18 January 2009 (UTC)

WikiProject Statistics discussion about usage of ƒ for mathematical functions

In case anyone watching this page is interested, there is a discussion at Wikipedia talk:WikiProject Statistics#ƒ or f about whether the symbol ƒ should be used for mathematical functions in Wikipedia articles. —Bkell (talk) 17:43, 18 November 2010 (UTC)