Google's bouncyballs



  • Yeah, what's up with the bouncyballs today? It's not really a failure, but I did say "WTF" when they showed up this morning, so it kinda counts.



    Anyway, it's actually kinda cool. Jscript things flying around is usually pretty horrifying ("Look! Unicorns!") but Google's implementation is cool due to the interesting way, uh... the interesting way their balls move.



    The thing that makes it is that window position affects what the balls do. It's kind of surprising you don't see that used more often. Of course, I'm using Win7, so I'm playing around moving the window to see what kind of model they're using for the ball mechanics, and I minimize everything else that was open...



    At any rate, usually there's some reason for a special Google logo. And you click on it to find out. In this case, for obvious reasons, that wasn't possible. Rather than using Google to find the answer, I decided to ask you guys.



    And that, I suppose, is [i]the real WTF[/i].





    By the way, check out the code after you're done playing with the balls.



  • Near as I can tell, the balls are connected to each other by spring forces, and bound to their equilibrium locations the same way. The mouse pointer applies an inverse square repulsive force. The balls move accordingly. The size of a ball seems to be related to the amount of potential energy stored in its springs -- the further from equilibrium a ball is, the larger it appears.

    Really cool, actually. (I didn't read the actual code)



  • TRWTF is that they block Opera 10.61, unless you mask as Firefox, even though it works 100%... and allow IE8, even though the ball sizes are messed up.



  •  Wait, what?

    First, I never visit the google site, I just type my question in the Chrome address field or the FireFox quick search thingy... but I just went to www.google.com and didn't see anything special... is that because it wants to serve me a localized page?



  • @b-redeker said:

     Wait, what?

    First, I never visit the google site, I just type my question in the Chrome address field or the FireFox quick search thingy... but I just went to www.google.com and didn't see anything special... is that because it wants to serve me a localized page?

    Same here. I even went to the non-localized page and got nothing.

    Googling for 'google bouncing ball' shows they're commemorating their anniversary.

    EDIT: Then again there's this guy who says this:

    This from the wikipedia item on Friedrich August Kekulé von Stradonitz,
    born September 7, who devised chemical valence theory and who discovered
    the chemical structure of benzene:

    "The other anecdote he told
    in 1890, of a vision of dancing atoms and molecules that led to his
    theory of structure, happened (he said) while he was riding on the upper
    deck of a horse-drawn omnibus in London. If true, this probably
    occurred in the late summer of 1855."

    Google is slyly commemoratiing Kekulé's birthday
    It's probably just a coincidence, I think, but who knows.

    EDIT 2: Here's a video of the effect.

    EDIT 3: Go to <meta http-equiv="content-type" content="text/html; charset=utf-8">http://www.google.co.uk/ to see for yourself.



  • @Valarnin said:

    TRWTF is that they block Opera 10.61, unless you mask as Firefox, even though it works 100%... and allow IE8, even though the ball sizes are messed up.

    That's not a WTF. Every company only has so much budget for testing, and usually Opera gets the axe due to this. I can guarantee that if my company's product works in Opera, it's due only to accident.



  • Maybe I should elaborate a bit. The WTF is that they blocked it by default, instead of only blocking after they confirm it doesn't work...



  • @Valarnin said:

    Maybe I should elaborate a bit. The WTF is that they blocked it by default, instead of only blocking after they confirm it doesn't work...

    But... if they don't have the budget to QA it, how could they "confirm that it doesn't work?" I think you missed the point of my post...



  • @PeriSoft said:

    but Google's implementation is cool due to the interesting way, uh... the interesting way their balls move.
     

    Where's dhromed when you need him?



  • @blakeyrat said:

    That's not a WTF. Every company only has so much budget for testing, and usually Opera gets the axe due to this. I can guarantee that if my company's product works in Opera, it's due only to accident.
     

    You poasted my reply.

    I just grabbed opera 10, and it's kinda nice. Those thumb tabs are a nice touch. Zoom has been improved, I see.

    Google is just the logo for me. Am I too late? Is there an archive of these things?



  • @dhromed said:

    @blakeyrat said:

    That's not a WTF. Every company only has so much budget for testing, and usually Opera gets the axe due to this. I can guarantee that if my company's product works in Opera, it's due only to accident.
     

    You poasted my reply.

    I just grabbed opera 10, and it's kinda nice. Those thumb tabs are a nice touch. Zoom has been improved, I see.

    Google is just the logo for me. Am I too late? Is there an archive of these things?

     

    Well, this is technically an archive. I'm not sure it'll do you much good, though.

    That said, Google's code without any newlines, spacing, or comments is still easier to read than a lot of other code I've seen that has those things.



  • Haha. Google, and budget constraints? You're funny... In any case, it takes more effort to implement browser sniffing than not.



  • @majorkirtar said:

    Haha. Google, and budget constraints?

    Yes. And probably time constraints, as well. You can't solve every problem by throwing money and people at it, you know... read some Mythical Man-Month if you think otherwise.

    @majorkirtar said:

    You're funny..

    I know. I watched a lot of MST3K as a kid.

    @majorkirtar said:

    In any case, it takes more effort to implement browser sniffing than not.

    Yes, but if you have an unknown quantity of completely untested code, you're better off not allowing it to run in environments it's not tested in. What if, for example, it crashed Opera? Or caused data-loss somehow? Then Opera users would be up shit-creek without a paddle, and Google would be getting the blame. (Are the odds of this code causing that low? Yes. Are they zero? No.)

    More to the point, you write software, right? You... QA your software, right? Why do I have to explain these concepts to anybody who frequents these boards? What the hell, man.



  • @blakeyrat said:

    @Valarnin said:
    Maybe I should elaborate a bit. The WTF is that they blocked it by default, instead of only blocking after they confirm it doesn't work...

    But... if they don't have the budget to QA it, how could they "confirm that it doesn't work?" I think you missed the point of my post...

    And I think you missed Valarnin's point. If you haven't tested it and can't know for sure if it works or not, give the user a warning and an option to try anyway instead of blocking them outright. Only block if you know it doesn't work.

    <rant>What is it with web applications needing to be fine-tuned for every different browser and version anyway? Going by all the hype, web is supposed to be the application platform of the future - so why does it exhibit some of the more serious problems of native applications like non-portability and trade some others in for new ones like extreme UI latency?</rant>



  • @tdb said:

    <rant>What is it with web applications needing to be fine-tuned for every different browser and version anyway? Going by all the hype, web is supposed to be the application platform of the future - so why does it exhibit some of the more serious problems of native applications like non-portability and trade some others in for new ones like extreme UI latency?</rant>

    You know, I had an awesome parody rant about microsoft and linux, with a lot of 'MICRO$HAFT' and 'linux on the desktop' but TDWTF went bugshit and I lost the whole thing.

    I blame... uh... community server? Microsoft? Chrome? I dunno. I blame something.



  • @smxlong said:

    Near as I can tell, the balls are connected to each other by spring forces, and bound to their equilibrium locations the same way. The mouse pointer applies an inverse square repulsive force. The balls move accordingly. The size of a ball seems to be related to the amount of potential energy stored in its springs -- the further from equilibrium a ball is, the larger it appears.

    Really cool, actually. (I didn't read the actual code)

    I once built something similar into a (otherwise pretty boring) graph editor I was tasked with coding. I made all the edges "springy", it was rather funny to grab a node, drag it around and watch the entire graph bounce around like a child's toy. I left it in as a hidden easter egg :)



  • New doodle today on http://www.google.co.uk/

    The logo starts greyed out, but each letter gets its color for each letter you type in the search box.



  • @Zecc said:

    New doodle today on http://www.google.co.uk/

    The logo starts greyed out, but each letter gets its color for each letter you type in the search box.

     

    Now why isn't that visible for .nl?



  • @dhromed said:

    @Zecc said:

    New doodle today on http://www.google.co.uk/

    The logo starts greyed out, but each letter gets its color for each letter you type in the search box.

     

    Now why isn't that visible for .nl?

    Google seems to hate Benelux and Scandinavian countries. http://www.google.pt doesn't work either.

    But who cares? Take your pick:

     

    BTW, there's no google.va. I wonder what the pope uses.



  • @Zecc said:

    BTW, there's no google.va. I wonder what the pope uses.
     

    He sure ain't using the va, that's for certain.



  • @tdb said:

    And I think you missed Valarnin's point. If you haven't tested it and can't know for sure if it works or not, give the user a warning and an option to try anyway instead of blocking them outright. Only block if you know it doesn't work.

    Yes, well, that's what our company decides to do. That's not what Google decides to do. Google's philosophy is much, much safer than ours. (Although our saving grace is we have so few customers, we can almost guarantee none are using Opera.) I already explained in a previous post why it's safer that way, I'm not going to repeat that argument.

    @tdb said:

    <rant>What is it with web applications needing to be fine-tuned for every different browser and version anyway? Going by all the hype, web is supposed to be the application platform of the future - so why does it exhibit some of the more serious problems of native applications like non-portability and trade some others in for new ones like extreme UI latency?</rant>

    The W3C is basically the worst standards organization in existence. It sums to that.

    Despite that, making multi-platform software is still about an order of magnitude easier on the web than it is using anything else. Unless you don't care if your application's GUI sucks donkey ass, then use Java.



  • @PeriSoft said:

    @tdb said:

    <rant>What is it with web applications needing to be fine-tuned for every different browser and version anyway? Going by all the hype, web is supposed to be the application platform of the future - so why does it exhibit some of the more serious problems of native applications like non-portability and trade some others in for new ones like extreme UI latency?</rant>

    You know, I had an awesome parody rant about microsoft and linux, with a lot of 'MICRO$HAFT' and 'linux on the desktop' but TDWTF went bugshit and I lost the whole thing.

    I blame... uh... community server? Microsoft? Chrome? I dunno. I blame something.

    You actually type your messages directly into the text field? Trusting soul... Here, use this.



  • @blakeyrat said:

    Yes, but if you have an unknown quantity of completely untested code, you're better off not allowing it to run in environments it's not tested in. What if, for example, it crashed Opera? Or caused data-loss somehow? Then Opera users would be up shit-creek without a paddle, and Google would be getting the blame. (Are the odds of this code causing that low? Yes. Are they zero? No.)




    I wasn't really thinking about a hard crash, so you've moved my viewpoint to 50/50. On the one hand, if it crashes the browser it's obviously a bug in their implementation and should be fixed, and they may not realize it right away. On the other hand, it could be fine and if people just use and handle exceptions properly...



    Anyways, it looks like they decided to block Opera for this logo as well. :(



    -- EDIT --



    As a side point, the only thing I've personally ever seen crash Opera is Flash.

    <!-- Argh, anyone know a better html to pdf converter for PHP? I'm stuck with DOMPDF and it doesn't handle styles well at all... -->


  • @Valarnin said:

    I wasn't really thinking about a hard crash, so you've moved my viewpoint to 50/50. On the one hand, if it crashes the browser it's obviously a bug in their implementation and should be fixed, and they may not realize it right away.

    Or it's a bug in Opera, but people blame Google because it says "google.com" in the URL bar.

    I call this "Microsoft syndrome." Microsoft is responsible for perhaps 1/3rd of bugs they're blamed for. If a customer doesn't know why a particular piece of software crashes, they usually assume Windows was responsible and put the blame on Microsoft.



  • @dhromed said:

    Google is just the logo for me. Am I too late? Is there an archive of these things?

    There is an archive of those logos, but as it is not updated real-time, you are too early for it. If you dig inside, you will find pacman as well as Newton's apple (the only two "interactive" doodles I can remember before the current bouncy balls).



  • @Anonymouse said:

    @smxlong said:

    Near as I can tell, the balls are connected to each other by spring forces, and bound to their equilibrium locations the same way. The mouse pointer applies an inverse square repulsive force. The balls move accordingly. The size of a ball seems to be related to the amount of potential energy stored in its springs -- the further from equilibrium a ball is, the larger it appears.

    Really cool, actually. (I didn't read the actual code)

    I once built something similar into a (otherwise pretty boring) graph editor I was tasked with coding. I made all the edges "springy", it was rather funny to grab a node, drag it around and watch the entire graph bounce around like a child's toy. I left it in as a hidden easter egg :)

     

    You can get pretty complex behavior out of simple mechanics. I wrote something similar to this, too, when I was screwing around with SDL. With my uber-powerful Athlon 600, I was able to make an array of about 2000 particles, each of which tried to stay away from all of the other particles. I set up the equations so it was highly nonlinear; they just COULDN'T be any closer than, say, 20 pixels, but didn't care at all if they were further away than 25. I made something that loaded up a bitmap and used it as a map for all the particles to hate; the result was essentially a 2D liquid. It had all the right characteristics - you could slosh it, make waves, pour from the top and see it swirl around and run down slopes, set up currents within a 'tub'... it was a lot of fun. These days the same code could probably do 20k particles, 50k if you took advantage of multicore. If I had more time I'd paly with it some more.

    It's a bit surprising that people don't do this stuff more, really. It's easy as hell.



  • So apparently that cat is out of the hat now and the result is ... You don't have to click the "search" button anymore*! Surely impressive from a technical perspective with them apparently able to handle a multitude of the server load they had before without noteable latency, but from the UI standpoint, I'm a little disappointed. The great revolution, the groundbreaking paradigm change in the ways humans interact with information on the web is - a two second time save when doing a search? I already save that time by using the browser search box instead of visiting the site. So that's it?

    So what are your opinions about that groundbreaking revelation?

     

    (*I know this is old news to your guys in the states where they immediately activated the feature on the main page. However, outside the US you likely have only noticed it yet if you followed the news or if you happened to be logged in with your google account.)



  • @PSWorx said:

    So apparently that cat is out of the hat now and the result is ... You don't have to click the "search" button anymore*! Surely impressive from a technical perspective with them apparently able to handle a multitude of the server load they had before without noteable latency, but from the UI standpoint, I'm a little disappointed. The great revolution, the groundbreaking paradigm change in the ways humans interact with information on the web is - a two second time save when doing a search? I already save that time by using the browser search box instead of visiting the site. So that's it?

    So what are your opinions about that groundbreaking revelation?

     

    (*I know this is old news to your guys in the states where they immediately activated the feature on the main page. However, outside the US you likely have only noticed it yet if you followed the news or if you happened to be logged in with your google account.)

     

    You know what? I was on board with your opinion at first; now I'm thinking it's actually pretty cool. For one thing, the fast autocomplete is more useful than the dropdown box, unless you're just trying to take a screenshot of depressing common queries beginning with "why does my". I could see situations where you're typing, "OK, was this dxd8.dll or dxdll8.dll..." and you get the information a lot quicker when you get instant feedback. Just playing with it for a few seconds suggests that it could be useful not because it makes the way you search for stuff now faster, but because after you get used to it it'll change the way you search for stuff.

    Which is essentially what Schmidt (I think) meant when he said, "We want google to know what you're searching for before you do", and everyone went, ZOMG BIG BROTHER GOOGLES ARE WATCHING YOU SLEEP!!!



  • @dhromed said:

    @Zecc said:
    BTW, there's no google.va. I wonder what the pope uses.
    He sure ain't using the va, that's for certain.

    That's just because choir boys don't have one.



  • @blakeyrat said:

    I call this "Microsoft syndrome." Microsoft is responsible for perhaps 1/3rd of bugs they're blamed for. If a customer doesn't know why a particular piece of software crashes, they usually assume Windows was responsible and put the blame on Microsoft.

    not only customers, programmers too...

    i mean ms really makes shitty operating systems, when even small segmentation faults cause the applications to crash ...



  • @PeriSoft said:

    For one thing, the fast autocomplete is more useful than the dropdown box
     

    Apples and oranges.

    Autocomplete serves a different purpose from a suggestion dropdown. Autocomplete saves you typing where you knew you had to type it anyway, a suggestion box suggests stuff that you didn't know you might want or need to type.

    I don't know how Google implemented it, so it might still be cool, but autocomplete is not intrinsically superior to a dropdown, because it's a different thing.



  • @dhromed said:

    @PeriSoft said:

    For one thing, the fast autocomplete is more useful than the dropdown box
     

    Apples and oranges.

    Autocomplete serves a different purpose from a suggestion dropdown. Autocomplete saves you typing where you knew you had to type it anyway, a suggestion box suggests stuff that you didn't know you might want or need to type.

    I don't know how Google implemented it, so it might still be cool, but autocomplete is not intrinsically superior to a dropdown, because it's a different thing.

     

    I actually used it again just now, when searching for which rally driver was responsible for the phrase "Maximum Attack" - and again, it was really useful. The strange thing is that I can't define exactly why it was useful... it just kinda made it easier. Use it for a bit before you dismiss it.



  •  I can't. It's not here yet.



  • @dhromed said:

     I can't. It's not here yet.

    http://www.google.com/instant/ followed by "Try It Now" works for me, even when not signed in.



  • @Zecc said:

    @dhromed said:

     I can't. It's not here yet.

    http://www.google.com/instant/ followed by "Try It Now" works for me, even when not signed in.

    Man, that's still not working for me. I wonder what I did to get on Google's shitlist.



  • @blakeyrat said:

    @Zecc said:

    @dhromed said:

     I can't. It's not here yet.

    http://www.google.com/instant/ followed by "Try It Now" works for me, even when not signed in.

    Man, that's still not working for me. I wonder what I did to get on Google's shitlist.

    If it hadn't been for this thread, I wouldn't have noticed there was any difference. Are you sure it's not working? As for the shitlist, Google are like women: if you don't know what you did wrong, that's the problem.



  • @davedavenotdavemaybedave said:

    @blakeyrat said:
    @Zecc said:

    @dhromed said:

     I can't. It's not here yet.

    http://www.google.com/instant/ followed by "Try It Now" works for me, even when not signed in.

    Man, that's still not working for me. I wonder what I did to get on Google's shitlist.

    If it hadn't been for this thread, I wouldn't have noticed there was any difference. Are you sure it's not working? As for the shitlist, Google are like women: if you don't know what you did wrong, that's the problem.

    Yah, I looked over my friend's shoulder at work when it was working for him. There's really no way to miss it-- it slides the search field up when you type the first letter, then fills the rest of the page very rapidly. If I was missing that, I'd be pretty much blind.



  • @Zecc said:

    @dhromed said:

     I can't. It's not here yet.

    http://www.google.com/instant/ followed by "Try It Now" works for me, even when not signed in.
     

    Dude, the Try It Now button is just a link to google.com. No dice.



  • I missed the bouncy ball thing, and it's not in the archive :(

    @blakeyrat said:

    @Valarnin said:
    I wasn't really thinking about a hard crash, so you've moved my viewpoint to 50/50. On the one hand, if it crashes the browser it's obviously a bug in their implementation and should be fixed, and they may not realize it right away.

    Or it's a bug in Opera, but people blame Google because it says "google.com" in the URL bar.

    I call this "Microsoft syndrome." Microsoft is responsible for perhaps 1/3rd of bugs they're blamed for. If a customer doesn't know why a particular piece of software crashes, they usually assume Windows was responsible and put the blame on Microsoft.

     

    That's being a bit unfair to Microsoft there. As we all know, there's a lot of very shitty software out there, and even the good software crashes every now and then.

    I would probably guess that CTDs (crash to desktop) are divided like this:

    55% - Internal application errors (i.e. bugs in the program)

    20% - Driver bugs

    15% - Hardware (for me, this number is historically several times larger)

    5%  - Malware

    3% - External forces (generally followed by the application not handling it gracefully, like a yanked flash drive or a dropped network connection)

    2% - Bugs in the OS (or other system-level component provided by Microsoft)

    @dhromed said:

    @Zecc said:

    [quote user="dhromed"]

     I can't. It's not here yet.

    http://www.google.com/instant/ followed by "Try It Now" works for me, even when not signed in.
     

    Dude, the Try It Now button is just a link to google.com. No dice.

    [/quote]

    It works for me, but only if I click on the "Go to Google.com" link on google.ca

     



  • @dhromed said:

    @Zecc said:

    @dhromed said:

     I can't. It's not here yet.

    http://www.google.com/instant/ followed by "Try It Now" works for me, even when not signed in.
     

    Dude, the Try It Now button is just a link to google.com. No dice.

    It's working for me this morning. Whee. I'm not dead-last to get it activated.



  • @blakeyrat said:

    @davedavenotdavemaybedave said:
    @blakeyrat said:
    @Zecc said:

    @dhromed said:

     I can't. It's not here yet.

    http://www.google.com/instant/ followed by "Try It Now" works for me, even when not signed in.

    Man, that's still not working for me. I wonder what I did to get on Google's shitlist.

    If it hadn't been for this thread, I wouldn't have noticed there was any difference. Are you sure it's not working? As for the shitlist, Google are like women: if you don't know what you did wrong, that's the problem.

    Yah, I looked over my friend's shoulder at work when it was working for him. There's really no way to miss it-- it slides the search field up when you type the first letter, then fills the rest of the page very rapidly. If I was missing that, I'd be pretty much blind.

    I think it might be very connection dependent. Or maybe I'm just blind. I can't see it updating when I use it - just has the results as soon as I type. As far as I can tell, though, it's simply updating in the time it takes my eyes to move from the search box to the results.



  • @blakeyrat said:

    It's working for me this morning. Whee. I'm not dead-last to get it activated.
     

    D:



  • @pkmnfrk said:

    I would probably guess that CTDs (crash to desktop) are divided like this:

    55% - Internal application errors (i.e. bugs in the program)

    20% - Driver bugs

    15% - Hardware (for me, this number is historically several times larger)

    5%  - Malware

    3% - External forces (generally followed by the application not handling it gracefully, like a yanked flash drive or a dropped network connection)

    2% - Bugs in the OS (or other system-level component provided by Microsoft)

    I beg to differ. The main cause of crashes, in my experience, is configuration issues - about 95% of the time. (NB: I would say, for example, that a driver-caused crash where the driver was inappropriate (rather than just flawed) is a configuration error.) Application bugs are the next most common, comprising ~4.99999%. The remainder are the MS operating system failures, although that's still vastly overstating their share by many orders of magnitude. I've done a lot more desktop support in the last couple of years than I'd like, but even so, I can't remember seeing any crash in that time that was caused by the Windows OS being buggy itself. Badly set-up, sure, plenty of those. WTFs in the 'correct' settings, plenty. Actual bugs in the OS which cause real crashes? Vanishingly rare.



  • @blakeyrat said:

    @Zecc said:

    @dhromed said:

     I can't. It's not here yet.

    http://www.google.com/instant/ followed by "Try It Now" works for me, even when not signed in.

    Man, that's still not working for me. I wonder what I did to get on Google's shitlist.

     

    I dunno, but when my friend and I were looking through page and pages of results for network cameras that had been left out on the internets, it refused to serve us any more pages when we got up to page 12, because it thought we were a bot.

    Shame, really - spying on Japanese cafes, zoos, and random peoples' front entryways really never gets old.



  • @davedavenotdavemaybedave said:

    I think it might be very connection dependent. Or maybe I'm just blind. I can't see it updating when I use it - just has the results as soon as I type. As far as I can tell, though, it's simply updating in the time it takes my eyes to move from the search box to the results.

    What I meant is you can't miss the search box moving from the middle of the page to the top of the page. Unless you're the type of person who looks at the keyboard when he types, I guess.



  • @blakeyrat said:

    @davedavenotdavemaybedave said:
    I think it might be very connection dependent. Or maybe I'm just blind. I can't see it updating when I use it - just has the results as soon as I type. As far as I can tell, though, it's simply updating in the time it takes my eyes to move from the search box to the results.

    What I meant is you can't miss the search box moving from the middle of the page to the top of the page. Unless you're the type of person who looks at the keyboard when he types, I guess.

    Wait, how do you get a search box in the middle of the page? They abandoned that years ago, didn't they? goes and looks ooh, classic mode, that's better ooh, cool, it brings up the results page as soon as I start typing. Guess it just sucks to be you, then ;)



  • @davedavenotdavemaybedave said:

    I beg to differ. The main cause of crashes, in my experience, is configuration issues - about 95% of the time. (NB: I would say, for example, that a driver-caused crash where the driver was inappropriate (rather than just flawed) is a configuration error.) Application bugs are the next most common, comprising ~4.99999%. The remainder are the MS operating system failures, although that's still vastly overstating their share by many orders of magnitude. I've done a lot more desktop support in the last couple of years than I'd like, but even so, I can't remember seeing any crash in that time that was caused by the Windows OS being buggy itself. Badly set-up, sure, plenty of those. WTFs in the 'correct' settings, plenty. Actual bugs in the OS which cause real crashes? Vanishingly rare.

    Assuming "many" is at least 3, that works to around one in ten billion (10^10) bugs being due to the OS. Somehow I doubt your sample size is large enough to draw that kind of conclusions. It would probably take something like the entire tech support history of the whole world to manage that.



  • @tdb said:

    @davedavenotdavemaybedave said:
    I beg to differ. The main cause of crashes, in my experience, is configuration issues - about 95% of the time. (NB: I would say, for example, that a driver-caused crash where the driver was inappropriate (rather than just flawed) is a configuration error.) Application bugs are the next most common, comprising ~4.99999%. The remainder are the MS operating system failures, although that's still vastly overstating their share by many orders of magnitude. I've done a lot more desktop support in the last couple of years than I'd like, but even so, I can't remember seeing any crash in that time that was caused by the Windows OS being buggy itself. Badly set-up, sure, plenty of those. WTFs in the 'correct' settings, plenty. Actual bugs in the OS which cause real crashes? Vanishingly rare.

    Assuming "many" is at least 3, that works to around one in ten billion (10^10) bugs being due to the OS. Somehow I doubt your sample size is large enough to draw that kind of conclusions. It would probably take something like the entire tech support history of the whole world to manage that.

    Or, alternately, they're only responsible for a fraction of a bug.

    Either way, the point is that there is a fuckton more non-OS software out there than there is OS-software. Since all software is buggy, it stands to reason that there are a fuckton more non-OS bugs than there are OS-bugs.



  • @pkmnfrk said:

    I missed the bouncy ball thing, and it's not in the archive :(
    http://dotty-dots.appspot.com/

    Not the exactly same, but very close.



  • @blakeyrat said:

    I call this "Microsoft syndrome." Microsoft is responsible for perhaps 1/3rd of bugs they're blamed for. If a customer doesn't know why a particular piece of software crashes, they usually assume Windows was responsible and put the blame on Microsoft.

    I believe this is the only reason why people thought Windows Me and Windows Vista were shit. Yes, Vista had some deficiencies that were addressed by the various updates (and are now included in the Service Packs). But... both of these operating systems coincided with a shift in driver model. (VXD to WDM for Windows Me, can't remember the names for Vista). The manufacturers provided absolutely terrible drivers that crashed the operating system, and users blamed Microsoft.

    I seem to remember reading a study that showed something like 30% of BSODs on Vista in the first year were caused by Nvidia drivers. 30%!!!

    FWIW, I happily used Windows Me until Windows XP showed up. My uncle was one of the many that absolutely insisted Windows 98 SE was way way way better, but his computer never seemed to be any more stable than mine.


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.