Jump to content

Talk:Fitts's law

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Mac / Windows wars[edit]

I hardly think that Mac/Win wars are the stuff of NPOV, perhaps Talk is a better place.

I might point out the following about the allegation that "the Mac menu bar is always at the top of the screen whereas the Windows menu bar is not":

  • With the advent of larger and larger screens, a menu at the top of the screen becomes further and further away from the application itself (of course, many Mac users are stuck with their tiny screens and so don't suffer too much from this) --PS4FA
(Are you implying that Mac users have smaller screens than Win users? -- Noah Slater) —Preceding unsigned comment added by (talk) 04:54, 29 August 2007
  • Should you really want to do so, it is entirely possible to write a Windows app with the menu bar at the top - that's not actually a fundamental feature of Windows itself --PS4FA
(Yes, but practically no app does at the moment so the comment is still valid. -- Noah Slater) —Preceding unsigned comment added by (talk) 04:54, 29 August 2007
  • To reduce the need for using the menu bar, Windows applications generally provide a right-click context menu that immediately appears exactly where the mouse pointer is and thus is very time efficient. The Mac also has a context menu capability, but it requires either holding down a key whilst clicking, which requires both hands, or a click and hold, which wastes time. --PS4FA
(This is totally beside the point as a) the context menu doesn't do everything the top menu bar does and b) this isn't an argument saying Macs are better overall, it's saying the Mac menu bar is better than the Win menu bar. -- Noah Slater) —Preceding unsigned comment added by (talk) 04:54, 29 August 2007
No, the Mac method (ctrl-click or cmd-click) is superior because most of the user's time should be with one hand on the mouse and the other on the keyboard, when navigating, working, or playing. Moreover, the menu shortcuts in Windows are clumsy and redundant, whereas in MacOS they're almost immediate. lysdexia 09:19, 25 Oct 2004 (UTC)
Are you saying it requires the use of your entire mouse hand to click one button? Are you wearing mittens? I figure, I have several individually articulated digits on that hand; they might as well be doing something useful. -- Wapcaplet 19:05, 25 Oct 2004 (UTC)

Markup changes[edit]

Unrelated to the above, is anyone else bothered by the addition of periods following the math markup? To me they are a little distracting. -- Wapcaplet 19:06, 25 Oct 2004 (UTC)

Yes, I don't really like the use of punctuation within math equations just for the sake of making sentences grammatical. I also think it's non-standard; I can't remember seeing any math textbooks that use punctuation like the periods and commans that have been added in the article. MichaelMcGuffin 16:14, 9 Nov 2004 (UTC)

Base of the logarithm[edit]

Changed log_2 to log in that equation. The base of the logarithm doesn't matter in the equation because there's the constant "b".

Oh, never mind, you want this here so you can talk information theory units later.

Success and implications[edit]

The list in the section "Success and implications of Fitts' law" that begins with "If, as generally claimed, the law does hold true for pointing with the mouse, some consequences for user-interface design include:" seems like original research (WP:NOR). For example, the claim that pie menus are faster because of some Fitt's law principles might be mitigated by other factors like reading lists in a circle is slower than reading lists up and down. (I actually like pie menus so I'm happy to think they are.) Are there any sources for any of the information in this section? --Ishi Gustaedr 16:42, 9 May 2007 (UTC)[reply]

Grammar edit[edit]

editor checking in here from Oxford U. Press: we will not publish Fitts's, we know enough to use Fitts' [the correct term]. when I create an account I'll come in clearer, but really, I was hoping for better to have typos argued over on the first paragraph. But, please don't fix it. As Tulving said in a review once, with the paper that wrong I can cite it as an error! — Preceding unsigned comment added by (talkcontribs)

I imagine someone will find this contentious, but this is a scientific article and we should use the correct term. Lots of people drop the S incorrectly. Chris Cunningham 15:21, 8 August 2007 (UTC)[reply]

Why is it incorrect to drop the trailing S? —Preceding unsigned comment added by (talk) 04:54, 29 August 2007
The rule is that if a plural ends with an S, you drop the trailing S. If it isn't a plural, the rule doesn't apply. End of story. Trust me on this one, my own first name ends in S. Chris Cunningham 09:14, 29 August 2007 (UTC)[reply]
You are wrong. I am sorry to say that trust is not enough for me. See this page: http://grammar.uoregon.edu/case/possnouns.html - I will quote "If a singular proper noun ends in s, add an apostrophe. [Eg.] 'Chris' exam scores were higher than any other students.'" —Preceding unsigned comment added by (talk) 18:00, 29 August 2007
Whether it is "enough for you" is beside the point, for an IP address does not have a possessive form. I could come up with my own random website to counter, but that's left as an exercise to the reader. Actually, see apostrophe. Chris Cunningham 17:21, 29 August 2007 (UTC)[reply]
Just because I don't have a user account does not make my contributions any less valid. I will quote the very page you cite: "If the word ends in an s but is singular, practice varies as to whether to add 's or only an apostrophe... In many cases, both spoken and written forms will differ between people." This clearly states that there is no clear right vs. wrong in this debate. Many other paragraphs in that section re-affirm this view-point. The only logical resolution to this is too see what is standard practice and adopt that as the correct usage. Using Google Scholar (http://scholar.google.com/scholar?q=Paul+Fitts+%22Fitts+Law%22+1954) to check academic references to Fitts' Law clear shows a predominance of the form "Fitts' Law" and hence this should be formally used here as the common standard practice. I am reverting to this usage as per my rational. —Preceding unsigned comment added by (talk) 16:59, 30 August 2007
Another note to mention that before you go reverting my changes it would behove you to actually check the formal title of the essays/papers being cited in the article as your grammatical ideas (there is no 100% correct usage by your own citation) do not correlate with any of them. —Preceding unsigned comment added by (talk) 17:03, 30 August 2007
I have no idea how to change the title and redirect, sorry about that. —Preceding unsigned comment added by (talk) 17:04, 30 August 2007
  • Relative numbers of Google results are no way to decide what is and is not "correct" on the Internet. As the law was not actually named by Fitts, there is no reason to assume that the most commonly cited form on the Internet is definitive. This is as logical as assuming that the correct plural form of "DVD" is "DVD's" because most of the English-speaking population of the world uses it. Reverting this because apostrophe illiteracy will end the human race if unchecked. Chris Cunningham 08:54, 30 August 2007 (UTC)[reply]
  • You seem to have missed his point about DVD's; Chris said it would NOT be logical to assume that "DVD's" is a correct plural just because most people use it that way. You are agreeing with him, framed as a disagreement. Then you jump, with unknown logic, to the statement that "Fitts'" would be the correct possessive for Fitts (I believe that's what you're stating), even though the section you cite also allows "Fitts's", saying "If a singular noun ends with an /s/ or a /z/ sound (spelled with -s, -se, -z, -ce, for example), practice varies as to whether to add 's or the apostrophe alone." Dicklyon (talk) 15:16, 16 March 2008 (UTC)[reply]
  • But just because YOU say so that's good enough? What evidence do you need other than it being the most commonly used form? Why have you "corrected" the references to external articles even though you have actually cited the titles wrong as NONE of them actually use your suposed "correct" form? I also propose that you change the sentence "(often cited as Fitts' Law)" to "(always cited as Fitts' Law)" unless you can provide a single accademic reference that used your "correct" version. You can't just make up history to suite your gramatical preferences, even if they are correct. Can you provide anything to back up your claims that this is the correct term - given that the correct term is what most other people use. According to WIkipedia's guidelines you must be able to verify that this is infact the correct term to use - I think I have done that amply. Where is yours? —Preceding unsigned comment added by (talk) 17:15, 30 August 2007
My bad on the references: if they use "Fitts'" then they should be listed as such. I have no problem with them being changed back individually. And I'm not "making up history", I'm using the normative term in English. Regarding "providing a single academic reference", the first ten Google results return academic papers using the correct form, so I'm not sure why you believe it's unheard of. Again, that most people get this wrong isn't a reason to use the wrong form. And dropping the S is the wrong form. Chris Cunningham 09:17, 31 August 2007 (UTC)[reply]
How can you still assert that dropping the "S" is the correct form when your only citation directly contradicts you? —Preceding unsigned comment added by (talk) 17:35, 31 August 2007

It is disputed whether the trailing "s" should be used for the phrase "Fitts' Law"

  • Either way is acceptable in contemporary usage. Leave it the way it is now, both included, with a redirect--it is just fine. Problem solved. Both camps should be satisfied. So shake hands like gentlemen and go have a beer together. Professor marginalia 15:34, 6 September 2007 (UTC)[reply]

A cursory review of the relevant sources reveals that the proper name of Fitts' Law is Fitts' Law. Grammar rules and preferences aside, Wikipedia should not be re-naming the titles of articles that are the sources for the current article in order to support certain editors' points of view on what should or should not be the correct name of an article. This is Stalinesque. -GnuTurbo 01:01, 11 September 2007 (UTC)[reply]

As I said above (it would be nice if one's "cursory reviews" included reading the talk page), any sources which use Fitts' law which have been incorrectly "corrected" in the references section should be adjusted. Your proclamation of what "the proper name" is is unfounded. The B-52's defy grammar (and indeed decency) in their defiance of the rules of the language, but it was their decision to do so. If there's a source which indicates that Fitts himself used Fitts' law then I'm happy to move the article back, noting that this is a proper name (and thus not subject to the rules of the language). if not, there's no proper name and we use the English language's normative term for a law of Fitts, which is Fitts's law. I'll ignore the Stalinist comment, though I do see the analogy between correcting apostrophe misuse and ruthlessly oppressing and slaughtering one's political enemies. Chris Cunningham 08:19, 11 September 2007 (UTC)[reply]
The Stalin reference is to rewriting history. That is what is going on on the article page. You are rewriting history when you say that because you believe, or even have many indisputable sources saying, that referring to the name of the law in a particular way violates grammar you get to re-decide how the law is referred to. No, you do not. Wikipedia does not. The people who actually use this term in the first place, they decide. And they have already decided. The name is Fitts' Law. That may be a violation of grammar. But violating grammar does not get you the result of rewriting history. It may be a historical mistake, but it is what it is. It remains Stalinesque the way the sources for this article have been rewritten (falsified) to reflect the wrong title and thus propagate a certain mistaken point of view. If you think this is like slaughtering political enemies, you are a fool. -GnuTurbo 22:54, 11 September 2007 (UTC)[reply]
For the third time, I've already said I'd rather the references used the terms they actually use. However, the article itself should use the normative form. The normative form is certainly the one used in any class I ever took on the subject, so again, the fact that a lot of people get this wrong doesn't mean we should. Rather than ranting about this you could have changed them back by now. Sheesh. Chris Cunningham 10:41, 12 September 2007 (UTC)[reply]

I've corrected the citations to reflect their titles, this seems to be the consensus even among the "Fitts's" folk. Note that the Murata paper is cited by the ACM with "Fitt's" in the title (sic), but IJHCI, the publisher, have it as "Fitts'" - I've added the DOI link to their copy of the abstract. I see that only one of the external links uses "Fitts's" in its title but all of them say "Fitts's" here - I think that should be corrected too, but I'll leave that for someone else. Bazzargh 17:26, 13 September 2007 (UTC)[reply]

Google Book Search shows 101 "Fitts's" to 278 "Fitts'" (this is in published books, not in internet pages). Certainly not overwhelming. There's a strong tendency among English speakers to drop that final s, even when logically it must be kept. Only a few names have had that dropping of final s officially sanctioned by the writing guides, like "Jesus'". But they generally also give you an escape if that final s would make the pronunciation awkward. See this site, and many others like it. As for "Fitts's", I don't think it's so awkward, but it's also not how I've ever heard it said. I always assumed based on what I heard that it was "Fitt's" law; it's too bad that Mr. Fitts gets his name mangled this way. I'd be in favor of doing it right, even if the majority of authors do it wrong. There are plenty of examples of reliable sources that follow the grammar rules, and a majority that don't is no surprise, and no reason for us to adopt the wrong convention. Of course, you can't have your choice when citing others' articles. Dicklyon 02:57, 14 September 2007 (UTC)[reply]

Per GnuTurbo, Dlabtot; general popularity, as well as leading authoritative sources appear to favour "Fitts' Law", wikipedia article should be named and use similar. However, all references and direct quotes should reflect their original usage.--ZayZayEM 03:48, 14 September 2007 (UTC)[reply]
I'm not familiar with GnuTurbo and Dlabtot; what are these? As I point out above, general popularity is not surprisingly favoring the easy but sloppy way. What do you mean by "leading authoritative sources"? Do they trump these? Dicklyon 04:20, 14 September 2007 (UTC)[reply]

More Google Book Search results: if you look at books published 1990 or earlier, only the ones with at least limited previews avaiable (this weeds out the journals whose start dates are earlier), you find 3 "Fitts'", 2 "Fitts" and 7 "Fitts's". This evidence suggests that it was originally done "right" and that the standard way has degraded over time. It argues (to me at least) for sticking with the correct (original) "Fitts's". Dicklyon 04:26, 14 September 2007 (UTC)[reply]

See this PDF [1], page 4. Fitts's is apparently correct, whereas I would have put it the other way around! User A1 10:11, 14 September 2007 (UTC)[reply]
I didn't think that we Wikipedians customarily did OR. I thought we went for authoritative sources. I don't really see why grammar and style should be an exception. I have no source published in this millennium in my library. I would defer to anyone producing, say, 2 sources since 1999 that were contrary. Eats, Shoots, and Leaves is from this millennium and favors Fitts's, according to User A1. Particularly telling would be a more recent edition of the Chicago Manual of Style, the style manual used by the University of Chicago Press. I found no sources in my library that favor Fitts' as a possessive. Following are sources that favor Fitts's:
  • Willian Strunk and E. B. White (1959) The Elements of Style 1st ed. p. 1
  • H. W. Fowler (1965) A Dictionary of Modern English Usage 2nd ed. p. 466
  • Wilson Follett and Jacques Barzon (1966) Modern American Usage p. 434 (also see pp. 254-5)
  • The Chicago Manual of Style (1993) 14th ed. p. 20
  • William Strunk and E. B. White, Jr. (1999) The Elements of Style 4th ed. p. 10
DCDuring 01:42, 22 September 2007 (UTC)[reply]

My 2¢: Besides the evidence from Google, Google Book search, and the fact that ‘Fitts’ Law’ is the term used in The Human Interface, Google Scholar shows a 2740:134 preference. That makes ‘Fitts’ Law’ over 20 times more common in the scholarly literature! Let us assume that, grammatically, one should refer to objects of Fitts as ‘Fitts’s’. When we refer to said law, are we referring to the law of Fitts (Fitts’s law), or the law whose name is ‘Fitts’ Law’ (note the quotation marks and capitalization)? I think the evidence is overwhelmingly in favour of the latter. That is, ‘Fitts’ Law’ is the name of Fitts’s law. As such, it is merely a token, a rigid designator; it need not and, indeed, does not have any internal grammatical structure. Any grammatical (in)correctness was lost as soon as the phrase became a name. The only question now is what the correct name is, not its illusory grammar. And the evidence has already answered that question. (Cf.Moore's Law’, which is also capitalized, since that is the name of Murphy’s law.) So, unless someone can come up with a good reason why Wikipedia should deviate from using the standard name of Fitts’s law (i.e., ‘Fitts’ Law’), I will soon move the page back to Fitts' Law. —David Regev 18:33, 16 October 2007 (UTC)[reply]

So you're just going to ignore the evidence from reliable sources that it was called "Fitts's law" before about 1990? It seems wrong for an encyclopedia to choose to take the side of the lazy authors in letting a name degenerate into an agrammatical form this way. Dicklyon 19:20, 16 October 2007 (UTC)[reply]
First, using terms like ‘lazy authors’ is a strong indication that you did not read my argument and do not grasp that names depends neither on grammar nor on the laziness of authors. Second, have you done the search? Limiting the search to 1990 and earlier, Google Book Search prefers ‘Fitts’’ 38:17. And, no, ‘Limited Preview’ does not mean ‘not really old’. It means that it’s not preview-less and not ‘Full Preview’ (there are no books that have a full preview available that refer to the law). Google Scholar also prefers it a whopping 188:22. In fact, the oldest journals referencing the law use its proper name as well. ‘Fitts’s law’ (note the lower-case) does not occur in the literature before 1972. So, as I stated earlier, the evidence is overwhelming that the name of Fitts’s law is ‘Fitts’ Law’. —David Regev 07:03, 17 October 2007 (UTC)[reply]
Show me what searches you did. Here are mine: Fitts-law and Fitts's-law. Notice that in the former you have to look and see which ones have apostrophes and which ones don't. I see 8 Fitts's, 3 Fitts' and 2 Fitts. GBS is admittedly not the best tool for this. If you allow "all books" like this then most of what you get are journals that started before 1990, but they are articles from later, as you can plainly see there. The evidence that you call overwhelming is really quite ambiguous in my opinion. It suggests that the original name has "degraded" over time to the more easily pronouncable name that has lost its grammatical meaning. I'd prefer to stick to the original than adopt the more "neologistic" one. Or find a source that states that the reduced form is now the "standard" name for this law, and we can say so and cite it. Dicklyon 07:20, 17 October 2007 (UTC)[reply]
As to the capitalization of Law, it seems to be an editor's style choice on whether to capitalize eponymous laws or not. We have Planck's law, Coulomb's law, and Godwin's law, but Moore's Law. All are mixed in the literature; there's no apparent evidence that some are properly considered proper nouns and others are not; I think wikipedia prefers not capitalizing when not necessary, and the List of eponymous laws supports that implicitly. Strangely, the earliest use of Moore's that I can find is Moore's Laws. Dicklyon 07:44, 17 October 2007 (UTC)[reply]
You used ‘Limited Preview’. That is limited to the books for which Google got explicit permission to show a limited preview. You must use ‘All Books’. Thus, "fitts' law" and "fitts's law". Similarly, for Google Scholar, "fitts' law" and "fitts's law". Looking through the very earliest examples (keep subtracting ten years), I see plenty more instances of ‘Fitts’’ (with the apostrophe). As to capitalization, I think it’s important enough in this case in order to indicate the status as a proper noun, i.e., a name, rather than a grammatically (in)correct description. —David Regev 08:02, 17 October 2007 (UTC)[reply]
Let me just reiterate that when you do all books, with a date limit, most of the hits are actually never articles in journals that started publication earlier. So you have to be careful what you count. It becomes very difficult to tell the actual ages, esp. when there's no preview. Many books from the 80s and 90s do have previews. I don't mean to say this is a great sample, but it does give some idea that in older books the situation is more formal. Dicklyon 17:03, 17 October 2007 (UTC)[reply]
Proper nouns do not result simply from ossification. Unless there's a definitive source to indicate that it is a proper noun, with a definitive spelling, this is definitely a live issue. And as a live issue, I'm not much in the mood for allowing the illiteracy of a collection of academics readily found through Google decide it. (I rather think that any argument on Wikipedia contain the word "Google" should be instantly disqualified as the most outrageous application of mob-rule, to be honest.) Chris Cunningham 10:39, 17 October 2007 (UTC)[reply]
The references to ‘Google’ were always accompanied with ‘Books’ or ‘Scholar’. This collection of scholars of whom you speak is not some people on the web, but the vast majority of published scholars working in the very field(s) in which this law is used. Are you really going to tell all these people that the sequence of sounds they use to refer to this object is “wrong” because its historical origin is based on bad grammar? The fact that this law is almost always designated by the same token that sounds like bad grammar demonstrates that this is its name, not a description. Otherwise, any author referring to the law would have to explain who Fitts is in the first place, which the vast majority of scholars do not do. They are referring to the entity they collectively call ‘Fitts’ Law’, not the one and only law of the known man named ‘Fitts’. The fact that a small number of grammatically-conscience people have given this law a second name should not detract from the fact that the far more common token is its proper name. Thus, the result of the ossification is which proper noun Wikipedia should use. Moreover, many of the arguments for the move have already been refuted (such as the historical descent of ‘Fitts’s law’). Finally, considering that the article started out as ‘Fitts’ Law’, remained that way for most of its lifetime, and that this is such a “live issue”, shouldn’t there have be some sort of consensus before you moved it? Shouldn’t this be discussed openly with other admins as well? Is this really Wikipedia’s crusade? —David Regev 16:37, 17 October 2007 (UTC)[reply]
Admins tend to avoid content disputes, since that's not their role, but you can use WP:3O to solicit an unbiased third opinion from a volunteer. That might not work, either, as they tend to stay away when there are already 3 or more in a discussion. Dicklyon 17:06, 17 October 2007 (UTC)[reply]
One does not need consensus to edit articles. This is a common fallacy. Wikipedia is not a democracy and its positions are not formed by mob rule. They are formed by discussion. Your results have shown that the illiterate spelling has a far from overwhelming lead, even over a random Google search, and you've not provided any argument more concrete than "lots of people misspell it, so Wikipedia should too". Chris Cunningham 11:05, 18 October 2007 (UTC)[reply]
You do need some sort of consensus when there is such a loud dispute over article titles, especially when one needs administrator privileges to move the articles (cf. Yoghurt). And my results do show an overwhelming lead: for the first several decades, ‘Fitts’s’ simply does not appear in scholarly journals, and, now, ‘Fitts’’ leads 2750:135. How is this not overwhelming? You make the very mistake I was pointing out when you say ‘lots of people misspell it’. What are they misspelling: a description of a law, or the name of the law? Fitts’s law is called ‘Fitts’ Law’—by almost everybody except a few people who mistakenly apply grammatical rules to names. Contrary to Humpty Dumpty, names do not have to mean anything; they are merely sequences of sounds chosen by history and convention. Ignoring naming conventions will simply create confusion. If you want to tell everybody in the fields of psychology, ergonomics, and interface design that they should use a different name, please do so. But Wikipedia should not be the means of this crusade. —David Regev 22:49, 18 October 2007 (UTC)[reply]
It isn't a crusade, we're not "ignoring convention", and the lead is only "overwhelming" over a particular contrivance of results. I remain opposed to forced illiteracy due to mob rule; I obtained my degree through an esteemed university which did not bow to mob rule; and Wikipedia's consensus does not indicated a deference to mob rule. This is not "loud dispute", it's kvetching from an underwhelming section of Wikipedia, and I maintain that anyone in a position to tell me, a man who graduated high school with a good result in English, how to use the language, remains in favour of the current text. Chris Cunningham 23:05, 18 October 2007 (UTC)[reply]
Regev, you seem to be the one on a crusade. You keep ignoring my explanation of why you're being misled by your search results. And there are enough people who feel strongly about leaving the article where it is that you'd probably not get far with a move proposal. Further argument here is not likely to convince anyone, since it's not a numbers game. Dicklyon 03:52, 19 October 2007 (UTC)[reply]
Let's say for the sake of argument that both usages are correct. Fitts' may or may not imply plurality, while Fitts's exclusively implies singularity. Fitts is singular, therefore Fitts's is more accurate and specific. My other comment is that opinions should be WEIGHED and not COUNTED. IMO this applies to the opinions of the editors, but also random Internet writers.(Thebigbradwolf (talk) 15:38, 19 March 2008 (UTC))[reply]
Exactly so. And more weight should be given to earlier usages, since lots of people are sloppy. The "Fitts's" form was more prevalent earlier, as I pointed out, and the shorter and less appropriate "Fitts'" more prevalent later, presumably due to carelessness and semi-literacy of so many authors. Dicklyon (talk) 16:22, 19 March 2008 (UTC)[reply]

This discussion is insane. "Fitts' law" is foremost a scientific term. It is spelled that way "Fitts' law" in every scientific article I have ever read on the subject. Type in "Fitts' law" in Google Scholar and see for yourself. The most prominent researchers in human-computer interaction on Fitts' law all spell it that way: Scott MacKenzie, Shumin Zhai, Bill Buxton, Ravin Balakrishnan, etc. The "Fitts' law" spelling is also used in text books in the field. Don't let a crazy nitty-picky grammar argument influence Wikipedia to dictate a change of the spelling of a scientific term. And yes I do have a PhD in the subject. Kristensson (talk) 00:17, 28 May 2008 (UTC)[reply]

You've made the discussion insane by re-opening it, and ignoring it. Quite a few people also use it in the grammatically correct form: [2] which was more common early in the history of the law. So what's wrong with using that? Dicklyon (talk) 04:57, 28 May 2008 (UTC)[reply]
I guess the scientists who use the term "Fitts' law" in their daily research have no say in the matter because the enlightened tribe of Wikipedia editors have the power to dictate the spelling of a scientific term. Go check how it is spelled in the CHI, UIST proceedings (leading publications venues in the field), in the leading journals (IJHCS, Human-Computer Interaction, etc.) and you will see they all spell it "Fitts' law". Just check how it is spelled in the reference section in this very article. Since these are references to the article, I would say it is fair to give them significant weight on how the law should be spelled. Glancing through the search results in your link it appears people who spell it "Fitts's" have not really used the law in their core research, rather have just mentioned it in passing. You can't just count references blindly to decide this matter, you have to actually talk to people who use it in their work. Kristensson (talk) 12:38, 28 May 2008 (UTC)[reply]
Exactly. If we wanted to just go with a count, we'd go with Fitts'. But history and grammar support the way it is now, and the count of current usage is not a compelling enough reason to change it, given that the correct way is also still in use. Dicklyon (talk) 15:19, 28 May 2008 (UTC)[reply]
That is not what I wrote. Anyways, take care. Kristensson (talk) 08:02, 24 June 2008 (UTC)[reply]
Kristensson, thank you for your input. It’s nice to have an actual expert weigh in on the issue. Regarding your analysis of the material, please see my comments above, which support you. The editors here like to repeat the claim that the earlier usage is ‘Fitts’s’, but, as I argued above in detail, this is not the case at all. Considering that you have actually seen much of this material first hand, any further thoughts would be much appreciated. Of course, that won’t convince anyone. I have argued that this is actually a historical/linguistic/philosophy-of-language issue, but, of course, said editors seem to see nothing but a grammatical issue. —David Regev (talk) 21:04, 22 June 2008 (UTC)[reply]
David Regev, I appreciate your feedback. Sorry to say though, I have given up on the issue since I don't have time convincing Wikipedia editors obsessed with their own "truth". Last reference from me is the special issue celebrating 50 years of Fitts' law research in the International Journal of Human-Computer Studies, Elsevier Publishers (one of the oldest HCI journals, used to be called International Journal of Man-Machine Studies). Note how everyone spells the law in that issue, and also check how the law is spelled in the many references in their articles (some references are very old): Elsevier Journal Link ("Fitts’ law 50 years later: applications and contributions from human–computer interaction"). But I guess scientists are lazy, ignorant and have no say on the matter :) Take care. Kristensson (talk) 08:02, 24 June 2008 (UTC)[reply]
Actually, the main problem is that we are calling out a grammar usage issue in parenthesis at the beginning of the article: "Sometimes referred to" is not accurate it is not a different way of referencing. The logical outcome is that we will update every Wikipedia entry that discusses a proper name that ends in an "s" with "(sometimes referred to as ____s')"?? Insane. dougpy

Bill Gates uses the equivalent of "Fitts' law" http://www.thegatesnotes.com/Personal/Steve-Jobs

If we're going to stick with the use of "Fitts' Law" in the body of the article, we should include pronunciation. English speakers do not typically pronounce a trailing apostrophe ("My friends' favorite.") where as "Fitts' Law" is properly pronounced "Fitts-es Law". We should inform readers who may be introduced to it here that the grammatically incorrect phrasing used by academics in their journals does not extend to how they actually pronounce it when they speak. (skia (talk) 16:31, 5 February 2013 (UTC))[reply]

ID non-negative[edit]

"The Shannon form has the advantage that the ID is always non-negative"

The derivation given in the following section, without the +1, would be strictly non-negative too if the equations correctly followed the assumptions. The assumptions imply iteration until the distance to the target is less than W/2. So if D<W/2, then N=0, it takes no steps to meet the conditions. The initial equation could more rigorously be expressed as:

where .

I find it hard to believe that the reference (Card et al) would have made such an omission. Unless the article means to say that the Shannon formula is advantageous assuming one is too lazy to add cases to one's formulas?

In fact, the Shannon form, or maybe I should say the Mackenzie form since apparently that's who invented it, implies that moving your cursor to a button takes some non-zero time dependent on the distance to the centre of the button, even when your cursor is already inside the button from the start. I'm dubious. -- Tim Starling (talk) 03:21, 16 February 2008 (UTC)[reply]

Tim, my understanding is that there is some controversy over the MacKenzie form. If you are interested read this paper, which is devoted to a discussion of the concerns you mentioned: IBM Tech Report. (Published in the International Journal of Human-Computer Studies 61(6): 791-809. "Special issue: Fitts' law 50 years later: Applications and contributions from human-computer interaction". Kristensson (talk) 09:38, 29 June 2008 (UTC)[reply]


"Fitts's law" is definitely a hard one. The question whether it should be "Fitts's" or "Fitts'" aside, shouldn't "law" be capitalized? —Preceding unsigned comment added by (talk) 08:13, 9 April 2008 (UTC)[reply]

Why would it be? The wikipedia convention is to only capitalize first word and proper nouns in titles; see List of scientific laws named after people. Dicklyon (talk) 05:41, 28 May 2008 (UTC)[reply]

What is all that bullshit? :D[edit]

A whole law for something obvious? And no, a screen edge doesn't have an infinite width. (talk) 21:10, 4 March 2010 (UTC)[reply]

Um? It has “infinite” width because you can move a mouse pointer indefinitely in one direction, and stay at the edge. And Fitts’s law is not especially obvious: just look around at all the software which violates the implied design principles. –jacobolus (t) 04:21, 5 March 2010 (UTC)[reply]

Add information about newer form used in HCI design?[edit]

Would it make sense to add mention of the refinement of Fitts' Law for 2D position-matching tasks - in particular I'm thinking of the 'effective width' definition that I think was developed by MacKenzie, see "Fitts' law as a research and design tool in human-computer interfaces" 1992.

I came to this page looking for exactly that and the Accot-Zhai development isn't it because that looks at trajectory tracking.

People have run with Fitts' Law in a number of directions; I'm not sure what's appropriate to include here and I didn't see anything come up after a quick glance at other HCI-related pages.

Novice editor, thanks for your understanding! (talk) 22:23, 7 July 2011 (UTC)[reply]

Thank you for your suggestion. When you believe an article needs improvement, please feel free to make those changes. Wikipedia is a wiki, so anyone can edit almost any article by simply following the edit this page link at the top.
The Wikipedia community encourages you to be bold in updating pages. Don't worry too much about making honest mistakes—they're likely to be found and corrected quickly. If you're not sure how editing works, check out how to edit a page, or use the sandbox to try out your editing skills. New contributors are always welcome. You don't even need to log in (although there are many reasons why you might want to).
If you know some refinement to Fitt's Law, please add references for them into the article -you're the person most qualified to know how they fit into the article!- or place them here at the talk page so we can include them for you. Diego Moya (talk) 11:16, 8 July 2011 (UTC)[reply]

Substantial proposed updates[edit]

Not much happening on this page for a couple years, but exciting stuff is happening in the research community, partly driven by the advent of interesting pointing devices for large displays (e.g. nintendo wiimote). Here are some pieces of information that will drive edits I'd like to make, and I'm first soliciting feedback or disagreement before making them.

-Fitt's law is in fact not nearly the rock-solid model it's described as in many, many places, including this article. I can cite journal articles in which the model fails spectacularly (R^2 drops around 0.8 or lower), by simply extending the range and combination of distances and widths tested.

-The information theoretic claim about 'bits per second' is flat wrong, and in the original research was simply asserted by analogy and not mathematically demonstrated. R. William Soukoreff published at least one paper showing that the information theory does not hold up.

-Competing models to Fitts's law exist, and outperform it. They have existed all they back to within ~5 years of the original model being published. A recent study rather thoroughly validated a specific alternative to Fitts's law that requires 3 parameters instead of 2 (Shoemaker, 2012) The F-test was used to compare models with different numbers of parameters fairly. Opinion: everyone loves a single number in bits/second to quantify performance, and so gravitated to this candidate law in the face of evidence that it's not objectively as good at predicting movement times.

-The Accot-Zhai steering law is not a general, appropriate extension of Fitts's law to 2D tasks; it only applies to pulldown menus or similar. The basic Fitts's law model as-is applies to 2D tasks with *circular* (or spherical in higher dimensions) targets, with some important fixes, which have been studied and documented in published works by e.g. Jacob Wobbrock.

-Yves Guiard has made the criticism that prescribing a very specific accuracy of a human motor task and then only measuring time duration is problematic, as opposed to allowing the user to freely make a speed/accuracy tradeoff each time and characterizing this. This is not a mainstream POV but is an interesting idea at the forefront of research.

-The section that walks through the iterative sub-movement derivation of Fitts's law is one of many over-simplified and incorrect attempts to provide an underlying explanation for why the law holds (which, recall above, it doesn't actually). Absent a community consensus around a single correct and complete explanation for why a movement time prediction model holds, what justification is there for including just one in the article? There isn't even consensus around what the law should be.

I anticipate push-back on these edits because they reflect some very recent research that contradicts over 50 years of (flawed) research, and I want to make sure I can make edits that conform to WP standards before wasting time on edits that will be reversed. So please fire away at the above or request more detailed justification of problematic assertions. Cheers. --eecharlie 14:15, 3 November 2014 (UTC) — Preceding unsigned comment added by Eecharlie (talkcontribs)

I've done a first cut of most of these updates, and wanted to share my to-do list with any reviewers/editors whose attention is grabbed. There are some placeholder citations that I plan to fill out, which in some cases will mean that References are flipped over to inline citations, per the current editing plea on the page. I also need to review content such as 'success and implications' to determine which statements are facts backed by research and which are hypotheticals based on intuition. Experienced editors may want to comment on whether at some point the page should be renamed or split up to accommodate alternative models.

Finally, a quick note that the unsigned comment from a couple years ago was also by me. I still haven't figured out how to properly sign my name. --Eecharlie (talk

Old English poetry[edit]

I removed the hatnote

because the string fitt does not occur in that article, except in the phrase "a burial fitting a king". —Tamfang (talk) 03:40, 11 December 2015 (UTC)[reply]

And the only mention of a law is "Kaluza's law". I also can't seem to find any such named law in the poem, so unlikely that Beowulf readers would end up here. Looks fine to me. Opencooper (talk) 04:11, 11 December 2015 (UTC)[reply]

Effect of rescaling D/W[edit]

The statement the model implies that a target distance and width combination can be re-scaled arbitrarily without affecting movement time, which is impossible should be substantiated by a reference to literature. IMHO, it is not obvious that this is impossible. (Never mind that newer studies show that for large D's and W's Fitts' law does not fit very well, as someone pointed out above.)

Gabriel (talk) 13:15, 15 January 2018 (UTC)[reply]

The illustration is wrong[edit]

The variable D in Fitts's Law is the distance from the starting point to the center of the target, but the illustration shows D as the distance to the EDGE of the target. Since the illustration is wrong this should be corrected. — Preceding unsigned comment added by (talk) 12:59, 29 March 2018 (UTC)[reply]

Seems to be fixed. Kriomet (talk) 22:21, 11 February 2021 (UTC)[reply]


This article has been reorganized and significantly extended by User:Mantury as part of the coursework for the course 'Interaction Techniques and Technologies' (summer 2019) at the University of Regensburg, Germany. —Preceding undated comment added 21:06, 20 January 2020 (UTC)