• motsats (unregistered)

    a.k.a Gridlock: http://en.wikipedia.org/wiki/Gridlock

  • (cs) in reply to motsats

    re: /bin/true

    v1.0: exit 0 # "true" v1.1: exit 1 # "enum Boolean.TRUE" v1.2: exit # default v1.3: return # this works in "C" v1.4: exit # but apparently not in shell v1.5: exit FILE_NOT_FOUND v1.6: <empty script> # turns out we don't need to do much here

  • jedicus (unregistered)

    My guess:

    v1.0: <empty script> v1.1: add version number v1.2: add copyright notice v1.3-1.6: update copyright notice with the current year

  • (cs)

    Traffic like that doesn't happen so easily if people fricken remember NOT to pull into an intersection they can't get out of...

  • (cs)

    Hm... cars everywhere, angry drivers who refuse to be the one who gets traffic moving again, (because it would require going around the block)... I've played this game before.

  • jmc (unregistered)

    I love how those 2 santas seem to be enjoying the jam.

  • (cs)

    I blame the drivers. The junctions are clearly boxed (the yellow diamonds) so they should not have pulled out until there was somewhere to get out. Only a moron blindly pulls forward without thinking just because the lights told them to.

  • Cargo Cult (unregistered)

    Mac OS X?

    Soyuz:bin afoster$ ls -al /usr/bin/true
-r-xr-xr-x  1 root  wheel  38096  6 Feb 07:41 /usr/bin/true*

Soyuz:bin afoster$ file /usr/bin/true
/usr/bin/true: Mach-O universal binary with 2 architectures
/usr/bin/true (for architecture i386):	Mach-O executable i386
/usr/bin/true (for architecture ppc7400):	Mach-O executable ppc

    38kB, just to return true! And works on two separate processor architectures! (I believe it's also cryptographically signed...)

    Now there's progress for you.

  • Cesium (unregistered)

    Incidentally, yes, the program that returns its argument until killed, is 13 KB as a binary.

  • (cs)

    The malpractice ad reminds me of a billboard here for a news radio channel. It bears a picture of a bunch of cars with a banner across the top:

    "Depend On Us For Traffic"

    Yeah, thanks for all the traffic, guys.

  • BlueKnot (unregistered) in reply to MET
    MET:
    I blame the drivers. The junctions are clearly boxed (the yellow diamonds) so they should not have pulled out until there was somewhere to get out. Only a moron blindly pulls forward without thinking just because the lights told them to.
    Which describes the majority of drivers today.
  • (cs)

    The man page for /bin/true should be in the Tao of Programming: "The true utility does nothing, successfully. The untrue utility does nothing, unsuccessfully". Now that's deep.

  • (cs)
  • brl (unregistered)

    For someone loving to read legalese, try to look at clear instead of true, while that has one line doing something, the many corporations claiming rights in it are....

  • (cs)

    You know, there is also a program called false

  • (cs) in reply to MET
    MET:
    I blame the drivers. The junctions are clearly boxed (the yellow diamonds) so they should not have pulled out until there was somewhere to get out. Only a moron blindly pulls forward without thinking just because the lights told them to.

    Do we have any verification that the lights were broken, not the users?

    Asshat 1 drives into intersection while light is yellow, even though he has no way out. Asshat 2 misses his light thanks to asshat 1. Decides to just run it once there's an opening. ...griddlock!

  • Eduardo Habkost (unregistered)
    (taken on 2008-01-06 from a building in São Paulo)

    My friend probably has a time machine, because he has published this picture on December 18th, 2007. Before it was taken!

  • (cs)

    Isn't the malpractice one from a lawyer's firm or something? Surely that's not much of a WTF. I can't see anything on the advert that means it's from a doctor's surgery, if that's the intended WTF...

    I think someone's pulling your leg with the junction of fail, too. I've seen that around for a while.

  • (cs) in reply to vt_mruhlin
    vt_mruhlin:
    ...gridlock!

    ????

    Profit!

  • jpk (unregistered)

    Comcast has an inept billing system? I never knew. I stopped using the "auto" bill pay because there was nothing auto about it. Unless by "auto" they mean that I am supposed to automatically call them every month and speak to a rep and make the payment over the phone because their "system" doesn't "automatically" charge my credit card. That must be what they meant.

  • Johnny Awkward (unregistered)

    On my installation of Linux, '/bin/true --version' is reporting version 5.2.1. And it's 12K of code.

  • (cs)

    Hmm. On my Solaris box, /bin/true is an executable.

  • (cs) in reply to Claxon
    Claxon:
    <screenshot>
    You win this round...
  • Jeremy (unregistered)

    Linux:/bin$ ll true -rwxr-xr-x 1 root root 12K 2007-09-29 08:51 true Linux:/bin$ file true true: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 2.6.8, dynamically linked (uses shared libs), stripped

  • notme (unregistered)

    You should have a look at true.c from the GNU coreutils some time. You'll be amazed...

  • (cs) in reply to Eduardo Habkost
    Eduardo Habkost:
    (taken on 2008-01-06 from a building in São Paulo)

    My friend probably has a time machine, because he has published this picture on December 18th, 2007. Before it was taken!

    The earlier date would justify the Santas

  • (cs) in reply to Johnny Awkward
    Johnny Awkward:
    On my installation of Linux, '/bin/true --version' is reporting version 5.2.1. And it's 12K of code.
    Reading comments like this, I had to check what my OS does. 
    $ file `which true`
/usr/bin/true: Bourne shell script text executable
$ cat `which true`
#! /bin/sh
exit 0
$
    Thank you, NetBSD.
    hallo.amt:
    You know, there is also a program called false
    There's also a programming language called FALSE. (Imagine how compact you could make "true" and "false" in that!)
  • adhominem (unregistered) in reply to Johnny Awkward
    Johnny Awkward:
    On my installation of Linux, '/bin/true --version' is reporting version 5.2.1. And it's 12K of code.

    On mine, it's version 6.9, and 19k. Feature creep, probably.

  • (cs) in reply to BlueKnot
    BlueKnot:
    MET:
    I blame the drivers. The junctions are clearly boxed (the yellow diamonds) so they should not have pulled out until there was somewhere to get out. Only a moron blindly pulls forward without thinking just because the lights told them to.
    Which describes the majority of drivers today.

    Which actually describes the majority of drivers at any given epoch.

  • (cs) in reply to jpk
    jpk:
    Comcast has an inept billing system? I never knew. I stopped using the "auto" bill pay because there was nothing auto about it. Unless by "auto" they mean that I am supposed to automatically call them every month and speak to a rep and make the payment over the phone because their "system" doesn't "automatically" charge my credit card. That must be what they meant.
    Better than British Gas, who automatically direct debit a fixed amount that has nothing to do with the amount on your (estimated) bill, and send a letter to the previous occupant 2 months after you move in saying his gas will be cut off unless he pays, even though they know you live there now and have even sent bills to you and had them paid.
  • Gonzalo (unregistered)

    What's the deal with black cars in Sao Paulo? Half of them are black!

  • Bob (unregistered)

    Due to budget constraints, the Civil Engineers were forced to implement a square roundabout...

  • jtl (unregistered) in reply to Bob
    Bob:
    Due to budget constraints, the Civil Engineers were forced to implement a square roundabout...

    I think you mean a boxabout.

  • Lunkwill (unregistered)

    A v1.6 shell-sript "true" is a WTF indeed. An even bigger WTF however is how everybody is laughing at binary implementations of the same without even benchmarking it:

    mb@aldous ~ $ file /bin/true /bin/true: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), for GNU/Linux 2.6.9, dynamically linked (uses shared libs), stripped mb@aldous ~ $ cat true.sh #!/bin/sh exit 0 mb@aldous ~ $ time for (( i=0; i<10000; ++i )); do ./true.sh; done

    real 0m35.886s user 0m4.432s sys 0m12.373s mb@aldous ~ $ time for (( i=0; i<10000; ++i )); do /bin/true; done

    real 0m25.883s user 0m1.648s sys 0m8.637s

    There you go, NetBSD. Considering what a monster /bin/sh is in comparison to some "int main{return 0;}" plus libc gruft, it's pretty obvious why that is. Of course a dynamically linked "true" is a minor WTF in itself:

    mb@aldous ~ $ diet gcc -Os true.c -o true; strip true; ls -l true; cat true.c -rwxr-x--- 1 mb users 1248 2008-02-22 17:55 true int main(void){return 0;} mb@aldous ~ $ time for (( i=0; i<10000; ++i )); do ./true; done

    real 0m6.122s user 0m1.008s sys 0m4.072s

  • (cs)

    DON'T BLOCK THE BOX.

    It is a simple driving rule. If you can't make it across the intersection, you do not enter the intersection. It drives me nuts when people are selfish like this. It ruins many well thought out intersections.

    I love the people that panic and enter when the light turns yellow. I mean, if they don't go then, they might NEVER make it across. Jackasses.

  • BJ Upton (unregistered) in reply to Claxon

    screenshot

    Easily the funniest thing I've seen on the WTF.

    Just awesome

  • Chris Eldredge (unregistered)

    On the last one, the revisions probably reflect changes to the copyright years.

  • Jarrod (unregistered)

    One of my colleagues has just pointed out that the linux version of true can return false:

    $ /bin/true --help >&-
/bin/true: write error: Bad file descriptor
$ echo $?
1

    umm....

  • AC (unregistered) in reply to adhominem
    adhominem:
    Johnny Awkward:
    On my installation of Linux, '/bin/true --version' is reporting version 5.2.1. And it's 12K of code.

    On mine, it's version 6.9, and 19k. Feature creep, probably.

    I have version 6.9 of core utils and my version of true is 16k (my distro is optimized though) and does not accept command-line arguments (not even --version or --help). The GNU Core Utils versions of true are most likely version numbered after core utils and not what changes have occured.

  • (cs) in reply to DeLos
    DeLos:
    I love the people that panic and enter when the light turns yellow. I mean, if they don't go then, they might NEVER make it across. Jackasses.
    I have to admit to doing this at least a few times, particularly one particular intersection. Although only after sitting through a whole phase or two, and realising that if I don't go, I *will never* make it through. (Until either 10am or 8pm, at least...)

    It's just plain frustrating to sit through a green light at a blocked intersection, have the light turn red, the intersection clear only to block up again just before the light changes...

  • (cs) in reply to Mel
    Mel:
    It's just plain frustrating to sit through a green light at a blocked intersection, have the light turn red, the intersection clear only to block up again just before the light changes...

    Completely agreed. Being polite only works when everyone else agrees to be polite. If you're the only one being polite, you never get your turn and the guy stuck behind you gets out and punches you in the face. You just go when the light is green and keep inching forward.

  • eric76 (unregistered) in reply to MET
    MET:
    I blame the drivers. The junctions are clearly boxed (the yellow diamonds) so they should not have pulled out until there was somewhere to get out. Only a moron blindly pulls forward without thinking just because the lights told them to.
    When I'm in a city wanting to turn left, I'll pull out into the intersection and then wait for oncoming traffic to get past or to stop when the light turns red and then complete my turn.

    Any more, I rarely go to any big city. My current office is at a traffic light in such a lighly travelled city that you can walk diagonally across the intersection at 5 pm and not obstruct traffic.

  • (cs) in reply to Claxon
    Claxon:
    http://www.hst-sc.com/misc/therealwtf.gif
    Well done, sir, well done.
  • The Doctor What (unregistered)

    Even better, it's version 1.6, contains no uncommented code, AND HAS A BUG!

    The bug is that if you use this /bin/true from within your shell init scripts it'll cause an infinite loop!

    Not to mention a bug in your shell init scripts can cause /bin/true to return false!

    Re: version number, I suspect it was just to update the legalese. I seem to recall that the original version was just an empty file.

    Ciao!

  • (cs) in reply to MET
    MET:
    I blame the drivers. The junctions are clearly boxed (the yellow diamonds) so they should not have pulled out until there was somewhere to get out. Only a moron blindly pulls forward without thinking just because the lights told them to.

    Dunno 'bout other countries, but over here (the Netherlands) the rules are such that you HAVE TO start driving when the light turns green.

    That said, only a moron blindly obeys moronic rules.

    BTW: the santas are indeed cool, but I also like the bus blocking all lanes towards the bottom right.

  • La Monte H.P. Yarroll (unregistered) in reply to ThePants999

    The posted version of /bin/true has a bug, at least if /usr/bin/sh is bash or another "advanced" shell.

    This is a bug I originally reported in NextStep 0.8 quite a few years ago.

    On NextStep /bin/true was an empty file. /bin/sh was actually bash.

    What does bash do when it starts up? It loads ~/.bashrc. What happens if you call /bin/true from ~/.bashrc?

  • Leo (unregistered) in reply to Lunkwill

    Wow, you must have an incredibly slow 64 bit machine to take 25 seconds for the default true binary..

    @debian:~/Desktop$ file /bin/true /bin/true: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 2.6.8, dynamically linked (uses shared libs), stripped

    @debian:~/Desktop$ time for (( i=0; i<10000; ++i )); do /bin/true; done

    real 0m10.954s user 0m3.280s sys 0m7.504s

  • mare (unregistered)

    If I understand this correctly, those yellow diamonds in the intersection mean that you're not allowed to stop in the middle of it? And you have to wait before the intersection for the oncoming traffic to clear? (I'm serious, I've never seen them before and the first thing that turns up on google for "yellow diamonds intersection" is this very page :D)

  • Fant (unregistered)

    The GNU/Linux /bin/true as true.c: http://www.koders.com/c/fidDF83F838AA5D91F9B07D35DD6353BCA20BD2B48B.aspx?s=md5

    I find the Solaris solution rather elegant :-)

  • SomeCoder (unregistered) in reply to mare
    mare:
    If I understand this correctly, those yellow diamonds in the intersection mean that you're not allowed to stop in the middle of it? And you have to wait before the intersection for the oncoming traffic to clear? (I'm serious, I've never seen them before and the first thing that turns up on google for "yellow diamonds intersection" is this very page :D)

    In America, you aren't allowed to enter an intersection unless you can clear it, period. The left turn yield on green means that you can enter it and wait for traffic or until the light turns yellow to clear but you still must be able to clear it.

    99% of Americans (and especially in the state where I live) don't understand that you aren't supposed to enter an intersection that you can't clear. It's really irritating.

Leave a comment on “The Comcast "Bill", A Buggy Traffic Light, and More”

Log In or post as a guest

Replying to comment #:

Log Out

« Return to Article