There Is No Objective Evidence That Keyboards Are Useful

Especially on this wiki, no-one has shown any hard evidence that keyboards are useful.

I TRIED REALLY HARD TO WRITE A COUNTERPOINT USING JUST MY MOUSE BUT I FAILED. IT TURNS OUT EVEN THIS ACCESSIBILITY SOFTWARE IS CALLED 'ON-SCREEN KEYBOARD'.

People have talked about them being useful for typing, but psychologically one might prefer to just use the right or left mouse button alone, and click and select characters/words on their screen using the clipboard.. pasting them around into different places. All one would need was a character map tool and a mouse. No keyboard has shown to be of any use based on this proposal - and no-one has jumped in and shown any evidence.

You're focusing too much on psychology, whereas you should be focusing on psychiatry.

There are no statistics out there, nor has anyone shown us that keyboards are useful yet. Until further notice, therefore, keyboards are unproven empirically to have any use whatsoever.


You are being silly. This one is fairly easy to solve: have a typing contest just like a run-of-the-mill secretarial typing test. Most people would agree that's a fairly good metric. If somebody finds a way to type faster with a mouse, I applaud them.

You cannot have a typing contest on a wiki easily - who will organize it and who will prove that the results sent in are valid and unbiased? The wiki medium makes it impossible for the results of the test to be reliable.

Typing tests are just a speed test - and not a psychology or preference based test.

Nobody claimed it was the be-all-end-all of factors. It is one of many possible metrics. It's a starting point. Note that an employer may not care what personal preferences are; they just want productivity. Thus if you like Python but your employer has evidence that Java is "more productive", then you are stuck with your employer's weighing of evidence if you want to continue there. (It may be that individual differences overshadow average productivity numbers.) Ultimately, the customer assigns the weights. There are two problems with the theoretical evidence mentioned in TheoreticalRigorCantReplaceEmpiricalRigor. First, they are often not numeric. GoodMetricsUseNumbers (and preferably objectively countable).

No numbers have been posted that keyboards are useful. And even if numbers were posted, 99 percent of people wouldn't give a damn - because humans are already intelligent enough to know that keyboards are useful. Similarly, 99 percent of languages use a type system of some sort - even the dynamic and weak types - so there obviously are types, and pages like "there are no types" are useless. 99 percent of people also know that python is easier to use for real work than machine language, and 99 percent of people know that GOTO statements are not to be used in great quantity - therefore continually debating about such topics that are obvious, and demanding "objective evidence" is a waste of time. OOP (parts of or ideas that come from it) are obviously useful. Debating whether OO is useful is just as silly as debating whether or not keyboards are useful. Keyboards are obviously not hammers and cannot be used for hitting nails very well. This is obvious. Different tools. Goto statements can be used occasionally but should not be used typically. This is also obvious. Yet you continually DEMAND OBJECTIVE EVIDENCE for these most obvious, common sense topics that do not require MOUNDS OF OBJECTIVE EVIDENCE.

Once again, we demand evidence and at least a few wiki pages with people providing objective evidence that keyboards are useful. I demand that you ask people without a keyboard, since anyone with a keyboard would be biased. The results cannot be biased. I demand you spend weeks of your time coming up with this evidence.

Second, the weight of those factors toward "better" is disputed. My premise is that if the theory is good, then it should produce abundant empirical results also.

You still have not shown any numbers or empirical evidence that keyboards are useful. You don't even show any statistics or references - not one ounce of evidence has been provided (except someone else's BookStop - and it is only one puny book - not ten, not one hundred).

One argument is that it's too expensive and thus theory should be accepted at face value; but if doing it right is expensive, that's the way it is. Theoretical evidence can highlight potential issues to study further, but it is not a substitute for empirical. If this topic is to poke holes in these issues, it hasn't worked.

It sure has worked. It has clearly shown how hard it is for anyone to provide objective evidence ON THE SPOT within 10 seconds or 10 days.. like Top constantly demands. Where is the EVIDENCE? Where are the STATISTICS? Provide them RIGHT NOW or this topic has absolutely no merit. Keyboards, we declare, are USELESS. They are academic toys. Unless you provide objective evidence, statistics, and proof right now (not just blabbering about how we could do some typing tests), there is no empirical proof that keyboards are useful.

There's no need to significantly revise my stance. -- top

Extracted interesting part to AlternativesForKeyboards.


To resummarize the thread-mess:

There exists plenty of statistics documented by many in the HCI field, and summarized by JefRaskin in the book TheHumaneInterface. Some might claim that referring the reader to this book is a BookStop, and thus constitutes a failure to address the problem. This is false; in no way does this attempt to ShiftTheBurdenOfProof?; evidence in support of a claim was asked for, and evidence has been provided. Now the ball is in the reader's court; it is now up to the reader to partake of the voluminous quantities of research clearly demonstrating the advantages of keyboards over other (then-current) input methods. -- SamuelFalvo?

By one single person? That's not rigorous enough. That's the point of this page, to demonstrate that it can be endlessly argued for any topic whatsoever, that there is no objecctive evidence even when there is. It is a cop out.

OK, just one more time, with as much emphasis as is apparently required to get the point across: there are numerous papers on the topic, authored by many people. They are summarized by JefRaskin. What part of numerous and summarized don't you understand? -- SamuelFalvo?

There are numerous papers that state lots of things, mostly written by academics without practical experience.

Indeed. Some people even make vague, deprecatory statements about things they've neither bothered to look up, examine, properly judge, and (in general) experience.

The point is that at some point we have to use our intuitions and not just keep demanding objective evidence for commonly understood topics such as whether or not python is better than machine language for every day programming! Why even bother asking whether GOTO's are generally bad, when we already know so? Of course in certain cases machine language can be handy, as can a goto statement. But to beat those dead horses and to continually demand more and more objective evidence costs us time. At some point, we draw the line.

You need to define your objective criteria for superiority. If your primary criterion is speed, then the keyboard is useful. If your primary criterion is usability by people with only one hand then it sucks greatly. Ultimately TheSearchForTruthIsFutile.


Most of the above is just opinions and psychology. Keyboards are mostly just in people's minds. There is no hard evidence.

'key' and 'board' seem to be TautologyMachines! I demand a falsifiable objective definition of 'keyboard' before we move forward on this subject! And by 'falsifiable definition', I require both an algorithm and an empirical, objective, rigorous set of metrics that can exactly determine whether or not something is a keyboard. And, just to be a prick^H^H^H^Hroper advocate of reason, I'm gonna fight you every step of the way because I believe ObjectivityIsAnIllusion. And when I say ObjectivityIsAnIllusion, I actually mean 'usable' objectivity, by which I actually mean 'absolutely provable' objectivity (given no axioms or assumptions). Also note that I consider empirical measurements with potential for error to be subjective (under my HumptyDumpty definitions) because you can't 'absolutely prove' that the world isn't built on collective consciousness and is therefore possibly subjective by nature. So, now, just TRY coming up with an algorithm that uses 'empirical' metrics to determine something 'objectively' when even metrics themselves are subjective! I DARE YOU!

Hahahaha! EverythingIsRelative. And I'm impervious to your whiny complaints that 'EverythingIsRelative' is an absolute statement, and therefore contradicts itself because, whenever I find it convenient, I believe that absolutes, contradictions, and even truth itself are relative! It's all psychology anyway.

And DON'T YOU DARE call me a hypocrite or inconsistent in my beliefs or demands or accuse me of ShiftingTheBurdenOfProof for demanding proofs just because I'm unwilling to offer support for the assertion that 'key' and 'board' seem to be TautologyMachines. That would be rude of you, and nobody (especially me) deserves such rudeness. And my demands and vigorous handwaving aren't rude at all! I'm only rude when I'm in one of my 'reptilian' moods, and you don't have the right to say something I do is rude unless you can also convince me of it. You think you can prove something unreasonable to me, who won't listen? I don't think so! I've got just as much right to choose what constitutes 'reasonable' logic as you do, and I choose the one I make up on the spot! I'll even start six new topics to talk about it.

Note that I did not write any of the above text in this section. It appears to be a (bad) parody on my style. -- top

{Note that I understand the humorous but subversive attack on EverythingIsRelative. I consider it well done, funny and insightful but sadly it nonetheless ridicules the truth in EverythingIsRelative. -- GunnarZarncke}

It is mixing in different kinds of arguments and debates. I'm a practical person who will accept practical evidence if practical evidence is the topic at hand. Other arguments are about philosophical purity issues which are NOT about the practical realm. You intermix these in a way that mis-characterizes my opinion. To my knowledge I generally separate them. If I slip somewhere, then point at that specific spot rather than over-generalized accusations. If it doesn't illustrate anything, then it's nothing but social intimidation on your part which appears to be directed at me. Thus, I take offense to your little works here. In other words, you are troll. If I made a parody page such as AnalTypingWillCureCancer?, some of you would go ballistic on me. By the way, I do consider popularity as a legitimate form of empirical evidence. BUT, not the only. -- top When you say "you", you do realize more than one person are creating content on this page, Top? Not just one person is ridiculing or questioning you here.

Try a sig, Sid. Unless you provide some kind of unique signature (make up something, I don't care), I will treat you as one big dripping ball of top-hating slime. ;-P

Many people here would prefer to use signatures if it was viable. Unfortunately, it is not - and the wiki is not designed to be a web forum or BBS.

I happen to know that all my programming errors come into the system through the keyboard. Therefore, it should be clear to anyone that the keyboard is the problem. "Down with keyboards!!!" ;->


Seriously, this would fall into the category of "productivity test". First one devises techniques to measure productivity, and then tests with and without keyboards. Now it's true that different people will favor different metrics over other, perhaps because of different usages for their own shops, and that's fine. We can let them plug their own weights into each metric to see the result. We are providing a framework for them to make decisions, not giving them a final answer. -t


See also: ObjectiveAdvantagesOfWoodenPencils
CategoryHumor or what?

Definitely was intended as so.

CategorySpitefulAndBadHumor?

EditText of this page (last edited March 14, 2012) or FindPage with title or text search