Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×
The Internet

How Many Keys Have You Pressed? 302

teardrop.ca writes "A new project created by Jason Hooper involves the counting and displaying of statistical information regarding the number of keys you have pressed since sign-up to this project. A change from the distributed problem solving projects that have been around for awhile. " Finally a truly frivelous use of distributed computing! It's a bit thin, looks like it could be easily gamed, but damn it'd be funny if the whole world did this (never mind the security and privacy issues). I'm curious how many more times some keys are pressed then others (perhaps this would explain why my spacebars always seem to break on my laptops :/)
This discussion has been archived. No new comments can be posted.

How Many Keys Have You Pressed?

Comments Filter:
  • by Zeinfeld ( 263942 ) on Sunday January 27, 2002 @02:56PM (#2910166) Homepage
    How many keys then
  • by ekrout ( 139379 ) on Sunday January 27, 2002 @02:57PM (#2910168) Journal
    The most pressed Slashdot keys...

    F-I-R-S-T P-O-S-T

    Ironic considering this might be the first post.
    • AOL's would be M-E T-O-O or A/S/L?
      L-O-L would be a close second...
      • M-E-T-O-O ?

        Comeon. You know they just use "me2", or "me 2", or even if they try to spell it out, would just do "me to". I'd have to guess that the slashes in "ASL" are dropped more often than not too out of laziness. I have to agree with you on "lol" though.
        • ...in related news today, Clippy the Talking Paperclip shot himself to death in his home, leaving behind a computer logged into IRC. His suicide note read "The spelling!! OH, GOD, the spelling!!!"

          On topic: It might be worthwhile to run a program like this as a listener in IRC, just to check if the key frequency corresponds to letter frequency in "normal" English. And yes, I know that this program doesn't measure letter frequency, which is a damned shame.
    • No the most pressed slashdot keys are
      CAN YOU IMAGINE A BEOWULF CLUSTER OF THESE?

      Somebody feels they always have to make a beowulf comment. Even if the article has nothing to do with computers.

  • Pardon? (Score:4, Insightful)

    by RC514 ( 546181 ) on Sunday January 27, 2002 @02:58PM (#2910172) Homepage
    I'm supposed to download a program that sits in the tray and records all the keys I press? I mean, to count them, they all have to pass their code, right? And it has network functionality... HELLO? Security?
    • Re:Pardon? (Score:3, Informative)

      by ekrout ( 139379 )
      No, it's not available for download yet. Their site says:

      Pulse is currently not available for public download. It is in its beta phase and is currently undergoing tests by a group of friends on the internet. This will be the place to download Pulse when it goes public.

      (NOTE: An OS/2 port is being considered.)
    • Are you afraid of Magic Lantern too? Next you are going to say that police should need search warrants to check your prostrate for drugs.
    • Jesus christ, bro.

      It's 60,000 bytes big.

      We're talking minimum 27k filesize if it's a VB application (which it probably is)

      Is the remaining 33,000 bytes big enough to contain full key-stroke logging, back-orifice-style, screenshot-capturing, cdrom-drive-opening code?

      Bah... this looks like just some guy having fun. Don't be such a bore.
      • Re:Pardon? (Score:3, Interesting)

        by ecc0 ( 548386 )
        Is the remaining 33,000 bytes big enough to contain full key-stroke logging, back-orifice-style, screenshot-capturing, cdrom-drive-opening code?

        In short; yes.
        • Re:Pardon? (Score:2, Informative)

          by MrSeb ( 471333 )
          I'm sure it is possible. Not easily, but I'm sure it is.

          Maybe using DLLs, or whatever. But there's no previously reported complaints via any search engine (tried a bunch of them).

          But I'm sure the /. croud can find some, post haste ;)
        • Re:Pardon? (Score:2, Interesting)

          by DutchSter ( 150891 )
          Sure is. Consider this: http://www.TinyAps.org The notepad replacement I'm using is a grand total of 6k, written in assembler, yet it has word wrap and full search and replace. Jeez, at that rate you could write an entire kernel in the 60k download here.
    • It actually says nothing about tracking specifically which keys are being pressed, just the total number of keys are pressed. Nobody forces you to install it, it's not buried in some other "freeware" program. People will doubtlessly dissect it and log its activity to ensure it's not passing anything more than necessary. (If you wanted to log people's passwords without them being aware, would you advertise your program as a keystroke logger?) There are conspiracies, but I doubt this is one of them.
      • Is there a better way to fool people than to tell them you are doing something innocent and getting them to believe you? How many people would just assume that this program was safe even if it wasn't? Probably a lot.
  • Am I the only one... (Score:3, Interesting)

    by GearheadX ( 414240 ) on Sunday January 27, 2002 @02:59PM (#2910174)
    Am I the only one here who can see the rather hazardous applications of a program like this? I don't want anybody keeping track of how many keys I happen to be pressing when I'm typing.

    Among other things...
    • It's not like they'll be getting full logs of what you type- that would just be flat out irresponsible. If it's just keycounts, I don't see what the big issue is...

      Though I'd like to see the source to make SURE of what they're doing...
  • by MrHat ( 102062 )
    Sure that isn't statistics recording which keys you have pressed?
  • Although it *will* be good fun... it's not currently available to download.

    Oh well :)

    Pulse is currently not available for public download. It is in its beta phase and is currently undergoing tests by a group of friends on the internet. This will be the place to download Pulse when it goes public.
  • what the hell is the purpose of this? To release a statistic annually saying how many keys do users press on their keyboard?
  • by funkhauser ( 537592 ) <zmmay2 AT uky DOT edu> on Sunday January 27, 2002 @03:02PM (#2910186) Homepage Journal

    I knew a long time ago that distributed computing would truly make a difference in humanity's quest for knowledge. But I would never have imagined that we would be able to count and analyze keystroke data from users around the world. It is truly a great day for science! :p

    Blatant sarcasm aside, this is moderately interesting. Any chances we'll see a linux client?

    • > Blatant sarcasm aside, this is moderately interesting. Any chances we'll see a linux client?

      An interesting variation would be to h4x0r a PIC in-line with a PS/2 or AT-style keyboard connector. Log the pressing of keys in hardware and display a running count on an LCD display.

      In addition to knowing how many keys you'd pressed, you'd have a high probability of knowing if someone was fux0ring with your keyboard while you were away.

  • Reminds me of the many mice odometers [modometer.com] available.

    My distances were always disappointingly low compared to my colleagues - I use keyboard-shortcuts much more, I find them quicker.
  • I can see it now. Instead of KLOCs we will have MOKP (millions of keys pressed). My boss would love it!
  • So, basically, they want you to run a keystroke logger. The logger will then report back the number of times you push each key. Are we supposed to just trust them to ignore the order? I have to really doubt the sanity (or common sense, anyway) of the person who would voluntarily give away his security like that.
  • There have been programs measuring things like this for a number of years... this reminds me of a program I had once that would calculate how many miles your mouse travels [techtarget.com].

    Fun stuff. Trivial, but fun.

  • by thesupermikey ( 220055 ) on Sunday January 27, 2002 @03:03PM (#2910192) Homepage Journal
    I have spent most of my life keeping track of the keys i have pressed! a= 3 I= 3 h= 2 v= 2 e= 9 s= 6 p= 4 oy, this was a lot hard then i thought, im going to take a nap
    • I wrote:
      >have spent most of my life keeping track of the >keys i have pressed!
      >a= 3
      >I= 3
      >h= 2
      >v= 2
      >e= 9
      >s= 6
      >p= 4
      >oy, this was a lot hard then i thought, im going >to take a nap"

      I am sooo sorry.
      I'll never do anything that dumb ever

      mikey
    • I tried to do something like this once, but I foolishly used a computer to keep track of the keys pressed. Pretty soon the infinite loop I was in pressing keys and then pressing keys for the keys I just pressed and then pressing keys for the keys I just pressed got old.
  • How meny inkorrect keys r actualley pressed?

    That would be interesting to see. Possible to if you could record and check to see the most popular key pressed right before the backspace.
  • by weave ( 48069 ) on Sunday January 27, 2002 @03:05PM (#2910196) Journal
    If the question was, "Which keys get hit by me the most?" that would be easy...

    I bet it's 1, 2, 3, and 4 this past year since all I seem to do is play D2, drink potions, and die anyway....

    No offense teardrop.ca, but ah duh, did you expect any intelligent responses on this story?!

  • cool! very soon everybody could see how many times i type:

    - my credit card number
    - my real name
    - my address
    - my fav. web sites

    your personal stats page could be a nice profile.

    but wait... they would NEVER do that... wouldn't they? it's even open source! oh wait, it isn't i think ;)

    muuuh out
    [afx]
  • What is stopping any other kind of distributed program from recording key-strokes, and sending it back. I wonder if they are reading my keypresses right now.
  • RSI (Score:3, Interesting)

    by Score0, Overrated ( 550447 ) on Sunday January 27, 2002 @03:06PM (#2910205) Homepage
    The distributed version sounds dangerous for the security reasons others have pointed out ...

    But, could a local version help with RSI? I remember obnoxious programs which popped up every hour reminding you to take a break ... how about something which popped up every 1000 key strokes (or whatever) reminding you to give you wrists a rest.
  • Several years ago, a friend of mine analyzed his keypresses, and came up with a keyboard layout optimized for his particular usage.

    If he were to extend that, and have the daemon running all the time, it could optimize his keyboard layout in real time, to keep him most efficient.

    Eventually, he would need only one key.
  • I don't need a program to tell me which keys I type the most, I just have to look at the numbers and see which ones are worn the most.... (or in some cases which ones have had a marker used to draw the letters again)
  • the key that always broke on my laptop was the fucking tilde.

    I guess going to too many user porn sites back in the day.

    http://porn.com/~pornking/
  • Read the site! (Score:5, Informative)

    by jargonCCNA ( 531779 ) on Sunday January 27, 2002 @03:08PM (#2910225) Homepage Journal
    Wow, I'm amused about how many people replied to this article without actually having read the site. Jay's a good friend of mine, I know he wouldn't log the actual keys. Besides, when you go to the Privacy Policy [bitdevil.com] on the page (Yes, you can visit the mentioned sites! What wonder!), it mentions what Pulse will and won't do:

    It is the intent of Pulse to transmit the following information to dolphin.bitdevil.com on a basis whose periodicity is decided by the user through Pulse's configuration menus:
    - user's account name as provided by user
    - user's password for Project Dolphin as provided by user
    - one integer that represents the total number of keys typed since last contact with dolphin.bitdevil.com for the same purpose
    - the current time (represented by the number of seconds elapsed since midnight, 1 Jan 1970 UTC), according to the system clock on the user's computer

    What it is guaranteed Pulse does not record, collect, or transmit to this server or any other destination:
    - which keys the user types, with exception to the analysis of the very last key hit, in order to decide if it is a key that "counts" as a key being hit
    - the contents of the user's hard drive or any other attached or internal or external storage device that may hold any type of data
    - anything on the user's screen

    So, for the benefit of the lazy people who can't be bothered to actually read the important information on a product's website, there you go - the important bits of the privacy policy. Oh, can I get your addresses? I'd like to send you a hard-copy of the link on a big fscking piece of clue-by-four. Jesus.
    • Imagine, for a moment, that you are writing a program to secretly record keystrokes.

      Which privacy policy would you publish ?

      a) Trust us ... we are not recording keystrokes.

      b) We are secretly recording keystrokes and sending the to our server where they will be analysed by the FBI.

      Without proof, their privacy policy isn't worth the paper it's written on.
      • Re:Read the site! (Score:3, Insightful)

        by jargonCCNA ( 531779 )
        Well then, why not join up, and run a port sniffer at the same time? That way you'll actually have proof.

        Just because it's been written to run on Windows doesn't mean it's vanilla Micro$oft spyware.
        • Re:Read the site! (Score:3, Insightful)

          by Jeremi ( 14640 )
          Well then, why not join up, and run a port sniffer at the same time? That way you'll actually have proof


          Nope, that wouldn't prove anything except that the software wasn't sending secret info out in any obvious way. It could well be watching just for your password/credit card number, caching it when it sees it, and sending it out to a remote machine 1 bit at a time, as part of the "acceptable" data packets, over the next 24 hours.


          If you want proof, you would need to get the source, inspect it line by line, and compile it yourself.

          • No need for source code, disassemble it. Look for the system call to hook the keyboard driver, find the code that processes those keystrokes and look for anything shifty. If it's on the level it should be fairly straight forward. It's not like this is some essential thing where you have to prove that it's malicious before you decide not to use it.

            Once you're satisfied that it's ok, post it someplace and everyone who trusts you is good to go.

    • Jay's a good friend of mine, I know he wouldn't log the actual keys.

      Heh
      My ex-wife was a good friend of mine.
      She wouldn't tip brake fluid over my car :- /

      Besides, when you go to the Privacy Policy on the page it mentions what Pulse will and won't do

      And this privacy policy comes under European law also?
      Is Jay open to bribes from unscrupulous bastards who will pay for the data he collects?
      Can a melicious version of this code be put out there so a clueless windoze user downloads the wrong one? (one without a ''privacy policy'')

      Even if your mate has the best intentions, encouraging people to install spyware like this is very bad karma. You are encouraging people to take stupid risks.

  • by jarodss ( 243400 ) <mikedupuis79@@@hotmail...com> on Sunday January 27, 2002 @03:09PM (#2910230) Homepage
    Generate a script that writes "the quick brown fox jumped over the lazy dog", it uses every letter in the english alphabet, over and over just to throw off their stats.

    On a side note the keys at the support center here would be EU does not RTFM.
    • Hmm, it appears that your sentence is missing the letter 'S'.
    • by The Fred ( 462778 )
      No, the way to throw off their stats is to write a script that reads their stats and then type the less used keys more. Adding one to every key stat probably won't do that much.
    • Generate a script that writes "the quick brown fox jumped over the lazy dog", it uses every letter in the english alphabet


      So you american-english havenow scrapped the letter "S" from the alphabet by now? Way to go!

      Make that "jumps" and it should be set...
  • Imagine at the end of this project a huge database containing usernames, passwords, etc. posted on the web. If it were used like that, one user running the client on a corporate net with access to sensitive material could inadvertently take down the entire network.

    On the flip side, it would be cool to see how many keys are pressed by x number of users in x number of days. They should release the source so people can make sure it's not sending the actual -key- that is being pressed rather than the number of keys that are being pressed.
  • Make the code open source. That way, we'll see exactly what's being sent back along with what isn't. I'd like to see it done with key frequency, reporting every twelve hours or so, as opposed to sending complete logs of what you type, which I highly doubt they're stupid enough to try...
  • by Anonymous Coward
    You can almost count the number of keys pressed by parsing /proc/interrupts...
  • _That would explain why the spacebars always break on my laptops._

    For me it's always the left shift, cause of all the < and > neccesary for HTML tags.
  • by wayn3 ( 147985 )
    A new project created by Jason Hooper involves the counting and displaying of statistical information regarding the number of keys you have pressed since sign-up to this project.

    Is this a hoax? Am I on Candid Camera?

    Now why would someone want to perpetuate this when we just heard about keyboard wiretaps? Certainly not a SlashDot reader!

  • "Our research has confirmed that three keys are used most frequently by computer users: ctrl, alt, and delete."
  • Assuming that it is counting which keys as well as how many - the little web page seems to indicate how many, and his friends seem to be having a race - I bet you could figure a way to tell good spellers from bad, using that information.

    My guess would be that people who spell correctly would use uncommon letters with a certain relative frequency, but people who spell things wrong, frivelous [sic] for example, would use less uncommon letters and more common letters.

    You'd need to do a lot of training - I'm sure that many poor spellers would "look like" people with good spelling and unusual word choice.

    I don't want to be a spelling Nazi, I make occasional mistakes myself, but the editors of Slashdot need to start spellchecking their comments/articles. It isn't like any of your editorial comments are hand written, guys! Real journalists come on here and judge your entire readership, hundreds of thousands of people, based on the degree of professionalism you display.

    Stop slouching! It's two O'clock in the afternoon, PUT PANTS ON!

    Here's a java spellchecking applet [xde.net]. It certainly seems to work; I for one like to spell things correctly, I understand that other people don't and in a post I certainly have no business complaining, but I would like a little button, "spell check" next to Preview.

    There are at least half a dozen spellchecking projects on sourceforge. I haven't found any Java applets, unfortunately.
  • This reminds of that little app that has been floating around for years, the mouse odometer.

    I would display on screen the number of miles your mouse has rolled. Maybe he could piggyback this functionality... since the whole thing is pretty pointless anyway. (Except for the RSI/take a break after so many keystrokes function.)

    Actually it would be an interesting analysis to compare keyboard use to mouse use, per user, especially if it could be compared across platforms.
  • And then become depressed to find that the key most commonly used was backspace :P
  • by _typo ( 122952 )
    So the FBI's key logger gets announcements on Slashdot now? How much are they paying for this?
  • by willybur ( 217434 ) on Sunday January 27, 2002 @04:09PM (#2910474) Homepage
    This has already been done, in a way. A guy took a bunch of newspapers, and tallied up the numbers of times each letter occurred. He did this over a long period of time, and came up with a ranking chart. This list can be used to crack monoalphabetic ciphers, using frequency analysis (the most common code letter would translate to be most common real letter, and so forth). This is how you crack the Cryptoquotes in the newspaper.

    In terms of frequency, here are the percentages (out of _The Code Book_, by Simon Singh, page 19):
    • a: 8.2%
    • b: 1.5%
    • c: 2.8%
    • d: 4.3%
    • e: 12.7%
    • f: 2.2%
    • g: 2.0%
    • h: 6.1%
    • i: 7.0%
    • j: 0.2%
    • k: 0.8%
    • l: 4.0%
    • m: 2.4%
    • n: 6.7%
    • o: 7.5%
    • p: 1.9%
    • q: 0.1%
    • r: 6.0%
    • s: 6.3%
    • t: 9.1%
    • u: 2.8%
    • v: 1.0%
    • w: 2.4%
    • x: 0.2%
    • y: 2.0%
    • z: 0.1%
    • This has already been done, in a way. A guy took a bunch of newspapers, and tallied up the numbers of times each letter occurred.

      These results would be valid for any users who sit down at their keyboard and type out newspaper articles day-after-day.

      The results for normal use would be skewed depending on the user. For examples, I bet I hit "l" followed by "s" a lot more than standard Engligh texts would suggest. It also only addresses the finished document, not gross keypresses. Using myself as an example again, I bet the backspace key has a shockingly high frequency.

      But, it doesn't address the real point of this program at all: to count raw keypresses. I have no idea how many keypresses I make in a day. Whatever it is, I'm sure my estimate would be low.

    • The letter distribution is based on frequency of use. You only get one z and q but a lot of e's and a's.
      • The letter distribution is based on frequency of use
        If you are referring to the distribution of letters over the keyboard area, you are wrong. The original QWERTY design still dates back to the early ages of mechanical typewriters. With the mechanical system not being fully optimized, it was decided to move the often used letter conbinations as far apart as it was possible to physically slow the typist down and allow time for the "type" to move back from the paper before the next letter is launched. If you have ever typed on an old mechanical writer you know what i mean - typing too fast or accidentally pressing two keys at once would result in both types bolting forward and getting stuck.

        Just look at Dvorak (an alternative keyboard layout). After getting used to it, you are easily 10% faster compared to QWERTY.
    • There's much more to this than first meets the eye. Here are some things to consider when making the counts:

      • Press a key and hold it down - autorepeat. One keypress generates numerous characters of input.
      • Capitalization - Does "A" count the same as "a"? The "A" requires pressing TWO keys: the SHIFT key as well as the key for the letter a. (Assuming, of course, that the CAPS LOCK key is inactive! Otherwise, the "a" now requires the SHIFT key!)
      • Other "Meta" keys - What about the CTRL key? ALT key? Windows Key?
      • Non-US keyboards - It should not make a difference if a user's keyboard is designed for a different language; does it? (Think of German, Greek, Arabic, Korean, Japanese)
      • That's right, can't count just bytes, either! DBCS (Double-Byte Character Sets)
      • Then also consider non-character-generating, but frequently used keys like the arrow keys and the page-up, home, end etc. keys.

      I'm sure there's more, but I would really be interested to see how well the program captures every single keypress!

      BTW, it would be really interesting to compare the distribution of keypresses required for an experienced Emacs user compared to an experienced vi user for typing in say, identical computer programs! One could then see which one was more efficient. Then, of course, one could argue that some keypresses require more "work" than others. A home-row "d" being less work than pressing the digit "1"; but of course is that a "1" from the main group of keys, or the "1" on a numeric keypad? Let the editor wars continue! ;^)

  • Someone should calculate how much energy it takes to press a key, and write a program that counts how many calories you've burned. "Mom, let me have that extra desert, I've been playing Quake for two weeks!"
  • From a usability standpoint this could be used to determine ways to decrease the number of keys pressed to complete a task. If it could keep track of which keys were pressed it would be even more useful to application developers concearned with usablity and design issues.
  • ...when I can just ask the VX2 Corporation [vx2.cc] how many keys I've pressed?
  • How many keys could a typing geek type if a typing geek could type keys?

    A typing geek could type as many keys as a typing geek would type, if a typing geek could type keys.
  • Open source? (Score:2, Insightful)

    I wonder, will this key-counting thing be open source? I wouldn't trust it if it weren't opensource... it might count more than just keystrokes. We all know how popular spyware is there days. If the author is serious about this, he would make the program open-source... just my .02
  • I don't think there'll be people 'battling' for the top position on this ladder. All it proves is who wastes the most time posting to slashdot.
  • by alue ( 253363 )
    how frequently you press the keys on your keyboard in relation to each other is to just look at your keys. The friction generated by the movements of your fingers against the keys wears away the surface of the keys so that (1) the printed symbols begin to fade away and (2) the surfaces of the keys becomes smoother and smoother.

    There's a smooth shiny oval-shaped area about 2/3's the way across my space bar (starting from the left) where I'm evidently accustomed to tapping that particular key. =)
    • I can't entirely comprehend how it happened, but I've seen keyboards that not only have keys on the home-row wiped clean of markings, but they are rippled. There is a discernable ridge of plastic where one should not be.

      The problem with looking at friction is that a lot of times you end up rubbing across keys that you aren't actually using.

      While it might be a usefull metric for an individual to evaluate their personal use, it becomes meaningless if you want to compare two different users, as I've seen people who literally stab the KB with their fingers while typing.
  • `grep key /proc/interrupts` already gives a nice count under Linux. Remember that all keys generate one interrupt when pressed, and one when released. IIRC, some generate more than one int per press (extended keycodes?), and there may be an issue with debouncing.
  • Windows ME (Score:2, Funny)

    by Dwedit ( 232252 )
    When I ran Windows ME, Alt, Ctrl, and Delete were pretty high on the chart.
  • This isn't special...the FBI already logs all of my keystrokes for me...

    i wonder if they'd be nice enough to hack together something to let me look at my stats :-)
  • Something like this could be used to truly improve UI's. By studying key presses (and mouse movements), and seeing the particular things that *slow* people down in their interactions, better UI's could be created.

    Things like the Fitaly keyboard (and an IBM equivalent, and others), were created by using large texts to estimate pen motion and such. Actual user interaction would be even more valuable (although it's hard to say if it'd be much different; certainly some things like cursor motion and other navigation would come out higher in real world analysis, than using text analysis).

    The concept of automated collection and analysis of user interaction efficiency is pretty exciting. It really could amount to more than "which keys have you pressed the most." (With me, it'd be backspace :-)

    -me
  • Ahh... it may tell you which keys you press but not which part of the key.

    "this would explain why my spacebars always seem to break on my laptops"

    If I look closely at my laptop spacebar, I notice that the texture on the right edge of the spacebar is much smoother than the left side. Conclusion: I press the spacebar with my right hand much more often than my left.

    Hah, your program couldn't figure that out =P (And pressing just one side of a wide key will probably cause it to break in the long run!)

  • One copy of Diablo: $59.95
    One Logitech Mouse Man: $95.99 (now completely ruined)
    Internet Access: $34.95 / month

    The crackly noise whenever I move my index finger: Priceless

  • Oh, you mean RC5's not frivolous? There are people dying in the world due to our continued lack of scientific knowledge regarding various diseases and our own gene structures.

    Try one of these instead, please...

    Genome@Home [stanford.edu]: gene structure
    Folding@Home [stanford.edu]: protein folding
    United Devices [ud.com]: cancer and anthrax
    Parabon Pioneer [parabon.com]: cancer
    Entropia [entropia.com]'s FightAIDS@Home [fightaidsathome.org]: AIDS

Real Programmers think better when playing Adventure or Rogue.

Working...