Comment On Elevator Failure

Hey you - do you live in or around Pittsburgh? Wanna meet TDWTF staffers? If you answered yes and yes (or maybe) then forget about what you were going to do tonight and head on down to the Diamond Market Grill in Market Square tonight for TDWTF Pittsburgh Meetup - Part 2. Things kick off at 6pm...stop by and say "hi"! [expand full text]
« PrevPage 1 | Page 2Next »

Re: Elevator Failure

2012-12-14 14:26 • by Lil Jon (unregistered)
397241 in reply to 397164
TRWTF is FoxPro

Re: Elevator Failure

2012-12-14 14:30 • by Brian (unregistered)
>> "Oh, Visual Firefox, I wish I knew how to quit you! Literally,"

Visual Firefox?

Re: Elevator Failure

2012-12-14 14:46 • by Jack 27 (unregistered)
397247 in reply to 397162
AH:
"YouTube think it knows..."?

If "YouTube" is singular, "think" should be "thinks".

If "YouTube" is plural (which it's not), "it knows" should be "they know".

Ugh.


It's a British thing to treat company names as if they were plural.

"Microsoft have come out with a new operating system."

Re: Elevator Failure

2012-12-14 14:49 • by urza9814 (unregistered)
397248 in reply to 397242
Brian:
>> "Oh, Visual Firefox, I wish I knew how to quit you! Literally,"

Visual Firefox?


Yes, not to be confused with non-visual Firefox, which uses a text-to-speech engine to read the raw HTML code to you. It's not very well known, as it can only be invoked though voice recognition software running on a headless server.

Re: Elevator Failure

2012-12-14 14:51 • by 71 (unregistered)
397249 in reply to 397176
faoileag:
portablejim:
I guess 2+5=9 for very large values of 2 and 5.

Nope. 2.99999 + 5.99999 will never be 9 irrespective of how many 9s you might throw at the two numbers. Close, but never really equal.


True for any finite number of 9's. But if they are non-terminating decimals (the 9's go on forever) then 2.9[...] + 5.9[...] is strictly equal to 9.

Re: Elevator Failure

2012-12-14 15:02 • by ubersoldat
397252 in reply to 397241
Lil Jon:
TRWTF is FoxPro


+1

And yet, it's still being used and some apps actively maintained.

Re: Elevator Failure

2012-12-14 15:18 • by Capitalist (unregistered)
397254 in reply to 397200
Bas B.:
"I want to know why there's still 41% of the laminating sheets left unclaimed with that awesome -21% savings."

61% was already claimed. According my math only 39% is left unclaimed.
That's the math version of Muphry's law.

Re: Elevator Failure

2012-12-14 15:20 • by Paul Neumann (unregistered)
397255 in reply to 397249
71:
True for any finite number of 9's. But if they are non-terminating decimals (the 9's go on forever) then 2.9[...] + 5.9[...] is strictly equal to 9.
Except, strictly speaking 2.9-non-terminating is 3 and 5.9-non-terminating is 6. If we are trunking the numbers, or performing any other conforming non-integral manipulation on the numbers than the only other solution's are
2 + 5 = { 6, 7, 8 }

Re: Elevator Failure

2012-12-14 15:36 • by Quietust
Personally, I'd like to think that the "2+5=9" is just a clever way to filter out spambots that try to evaluate mathematical equation CAPTCHAs - a human would enter 7, get the error "must be at least 9", then enter 9 and successfully sign up, but a bot would always enter 7, not knowing to parse the error message for the correct value.

Re: Elevator Failure

2012-12-14 16:03 • by Zapp Brannigan (unregistered)
397259 in reply to 397229
Ken B:
faoileag:
portablejim:
I guess 2+5=9 for very large values of 2 and 5.

Nope. 2.99999 + 5.99999 will never be 9 irrespective of how many 9s you might throw at the two numbers. Close, but never really equal.
(I hope this formats correctly.)

  _     _

2.9 + 5.9 == 9
(That's supposed to be overlines above the ".9"s.

Not "close, but never really equal", but "exactly equal".


OMG! You are wrong. 2.99999... is an infinite series and it is equal to 3. Look it up.

Re: Elevator Failure

2012-12-14 16:15 • by foo (unregistered)
397262 in reply to 397259
Zapp Brannigan:
Ken B:
faoileag:
portablejim:
I guess 2+5=9 for very large values of 2 and 5.

Nope. 2.99999 + 5.99999 will never be 9 irrespective of how many 9s you might throw at the two numbers. Close, but never really equal.
(I hope this formats correctly.)

  _     _

2.9 + 5.9 == 9
(That's supposed to be overlines above the ".9"s.

Not "close, but never really equal", but "exactly equal".


OMG! You are wrong. 2.99999... is an infinite series and it is equal to 3. Look it up.
So, he's wrong because he's right!?

Re: Elevator Failure

2012-12-14 16:17 • by foo (unregistered)
397263 in reply to 397258
Quietust:
Personally, I'd like to think that the "2+5=9" is just a clever way to filter out spambots that try to evaluate mathematical equation CAPTCHAs - a human would enter 7, get the error "must be at least 9", then enter 9 and successfully sign up, but a bot would always enter 7, not knowing to parse the error message for the correct value.
Wow, this will certainly stop spammers for another day or two. Parsing error messages is so much harder than recognizing characters in an image.

Re: Elevator Failure

2012-12-14 17:44 • by Harrow (unregistered)
397271 in reply to 397198
Hank:
We were doing properly atomic database transactions with reliable locking, commits, rollbacks, before and after images, "deadly embrace" handling, all that stuff, back in 1980 on a Honeywell mainframe with COBOL.

Then came the "microcomputer" (aka PC) people: "You guys make everything too hard! This will be so easy!"

We looked into the technology and saw they'd blown off virtually all the principles of computer science and practical knowledge acquired from experience. We warned it wouldn't be reliable. There will be problems.

Nobody cared. All they wanted was easy. And we still struggle from that laziness today.
Apparently the thinking went something like "If a computer costs $9,000 instead of $900,000 then it obviously needs only 1/100 the amount of hard-won practical computer engineering knowledge to program it."

It was especially infuriating to us minicomputer (small stripped-down mainframe) programmers, who had labored for decades to ensure that our tiny charges were just as mathematically sound as (if a bit slower than) their big brothers.

-Harrow.

Re: Elevator Failure

2012-12-14 18:42 • by Turtlepie (unregistered)
Visual Firefox?

Re: Elevator Failure

2012-12-14 18:57 • by Silverhill
397274 in reply to 397181
Lockwood:
Sometimes they are nine. Sometimes they are six. Sometimes they are all of them at once.
This was obviously a premonition of quantum computing.

Re: Elevator Failure

2012-12-14 19:20 • by pedant (unregistered)
397276 in reply to 397193
Zapp Brannigan:
Grammar police are the the buzz killers of the the internet.


Your doing it wrong, if you comment about grammar on the internet, you have to make a grammatical error.

Re: Elevator Failure

2012-12-14 20:37 • by da Doctah
397280 in reply to 397177
Ho Miscreant!:
The value of "What is the sound of one hand clapping is wrong. Please meditate on this further"
What is the sound of Slow Hand Clapton?

Re: Elevator Failure

2012-12-14 20:52 • by Friedrice the Great (unregistered)
397281 in reply to 397174
portablejim:
I guess 2+5=9 for very large values of 2 and 5.


Unless you've overloaded the "=" operator to automatically add 2 ...

Re: Elevator Failure

2012-12-14 21:21 • by Anon (unregistered)
397282 in reply to 397167
chris:
John Jackson:
The FoxPro one isn't that uncommon... When you're running a foxpro program, it doesn't like being interrupted (it could corrupt tables if it exits prematurely)

Great. I just tried to look up how old the concept of db transaction logs is - I couldn't find a direct answer, but did see the top of a paper on the "history" of it, written in 1993.


There's NOTHING that can't corrupt a FoxPro table. Especially with the retarded way they "lock" a file by writing data to an invalid part of your drive. Because that was never going to be fixed in the OS.

CAPTCHA: eros
Say no more.

Re: Elevator Failure

2012-12-15 00:36 • by Norman Diamond (unregistered)
397283 in reply to 397171
Larry Garfield:
but it leaves me wondering - If my elevator has no operating system, how is it moving???
Unfortunately a non-funny answer is highly relevant. It's a big WTF and not a bit funny.

In recent years two Schindler elevators have moved while the doors were open, killing people who were halfway in.

Elevators move by not having safety devices. Just like programs, stripping out debugging checks when moving from development mode to release mode. You can be an arrogrant pr*ck so sure that no problems remain in your product even if your product isn't an operating system.

Re: Elevator Failure

2012-12-15 03:42 • by Otis (unregistered)
397284 in reply to 397185
MindChild:
My father is an elevator mechanic for over 20 years, so I can shed some light on the elevator moving. That panel has a missing operating system, not the elevator. An elevator is almost completely mechanical, driven by controllers that have designs dating back 80+ years. Some newer elevators (Otis most notably) use microcontrollers, but these tend to be undesirable because they require cooling in the machine room.


I'm a computer programmer/sys admin, and I can shed even more light: it doesn't say that the panel has a missing operating system, just that the Intel Boot Agent didn't find one. I get the same message from the Boot Agent every time I reboot my computer -- there is a BIOS error on my computer that prevents me from disabling the Boot Agent, so I have to wait for the network boot agent to fail before booting from disk.

Re: Elevator Failure

2012-12-15 07:17 • by powpowpow (unregistered)
397286 in reply to 397242
Brian:
>> "Oh, Visual Firefox, I wish I knew how to quit you! Literally,"

Visual Firefox?


to be fair, Visual firefox has the same problem

"firefox is already running"

Re: Elevator Failure

2012-12-15 16:06 • by Hired Mind (unregistered)
397289 in reply to 397283
Norman Diamond:
Larry Garfield:
but it leaves me wondering - If my elevator has no operating system, how is it moving???
Unfortunately a non-funny answer is highly relevant. It's a big WTF and not a bit funny.

In recent years two Schindler elevators have moved while the doors were open, killing people who were halfway in.

Elevators move by not having safety devices. Just like programs, stripping out debugging checks when moving from development mode to release mode. You can be an arrogrant pr*ck so sure that no problems remain in your product even if your product isn't an operating system.


Or maybe, just maybe, you're human. And maybe you make mistakes.

Elevators have physical interlocks to prevent the car from moving unless the doors are closed, precisely because the developers are human. And sometimes those fail as well, perhaps due to improper maintenance (because building managers are human as well).

Re: Elevator Failure

2012-12-15 16:57 • by PiisAWheeL
397290 in reply to 397183
Lockwood:
"Do you remember," he went on, "writing in your diary, "Freedom is the freedom to say that two plus five make seven"?"
"Yes," said Alexander.
The Registration Form held up his right hand, all fingers extended, then his left hand, its back towards Alexander, with the thumb, ring and little fingers hidden and two fingers extended.
"How many fingers am I holding up, Alexander?"
"Seven"
"And if the party says it is not seven but nine, then how many?"
"Seven"
"How many fingers, Alexander?"
"Seven! Seven! What else can I say? Seven!"
"How many fingers, Alexander?"
"Nine! Nine! Nine!"
"No, Winston, that is no use. You are lying. You still think there are seven. How many fingers, please?"
"How can I help seeing what is in front of my eyes? Two and five are seven."
"Sometimes, Alexander. Sometimes they are nine. Sometimes they are six. Sometimes they are all of them at once."
Somebody get that man a trophy! George Orwell ftw.

Re: Elevator Failure

2012-12-15 17:41 • by F (unregistered)
397292 in reply to 397262
foo:
Zapp Brannigan:
Ken B:
faoileag:
portablejim:
I guess 2+5=9 for very large values of 2 and 5.

Nope. 2.99999 + 5.99999 will never be 9 irrespective of how many 9s you might throw at the two numbers. Close, but never really equal.
(I hope this formats correctly.)

  _     _

2.9 + 5.9 == 9
(That's supposed to be overlines above the ".9"s.

Not "close, but never really equal", but "exactly equal".


OMG! You are wrong. 2.99999... is an infinite series and it is equal to 3. Look it up.
So, he's wrong because he's right!?

Of course. This *is* TDWTF, after all.

Re: Elevator Failure

2012-12-15 20:53 • by Norman Diamond (unregistered)
397294 in reply to 397289
Hired Mind:
Norman Diamond:
Larry Garfield:
but it leaves me wondering - If my elevator has no operating system, how is it moving???
Unfortunately a non-funny answer is highly relevant. It's a big WTF and not a bit funny.

In recent years two Schindler elevators have moved while the doors were open, killing people who were halfway in.

Elevators move by not having safety devices. Just like programs, stripping out debugging checks when moving from development mode to release mode. You can be an arrogrant pr*ck so sure that no problems remain in your product even if your product isn't an operating system.
Or maybe, just maybe, you're human. And maybe you make mistakes.
Yes I am human, yes I make mistakes, and yes I believe that is why debugging checks SHOULD NOT BE REMOVED when moving from test mode to release mode.
Hired Mind:
Elevators have physical interlocks to prevent the car from moving unless the doors are closed,
No they do not, which is exactly why those two elevators killed people.
Hired Mind:
precisely because the developers are human.
No, elevators NEED physical interlocks because developers are human, but elevators don't have physical interlocks because developers are arrogant pr*cks.

Improper maintenance was also a problem, but the intentional omission of safety devices is unconscionable.

Re: Elevator Failure

2012-12-16 16:42 • by FRed (unregistered)
2+5

Seems an effective way to get rid of SPAM bots - give the answer required in the error message rather than being the actual answer to the captcha....

Sort of reminds me of the Quizmania/The Mint.....Stupid questions with nonsensical answers....

Re: Elevator Failure

2012-12-16 18:59 • by Coyne
397301 in reply to 397197
Ben Jammin:
faoileag:
portablejim:
I guess 2+5=9 for very large values of 2 and 5.

Nope. 2.99999 + 5.99999 will never be 9 irrespective of how many 9s you might throw at the two numbers. Close, but never really equal.

I believe that:
1) 2.9 and 5.9 are just small values of 3 and 6 respectively and
2) if you're gonna be a math nazi, you should get it right.

p.s. I still like the original joke



No! NO!! NOOOO!!!! Anything but the 1 does not equal .99999999... argument again! It's been settled! The mathematicians won! Please, no more of this torture!


Seriously, does it really bother you that 3.00000... and 3 are the same number? No? Then why does it bother you that 2.99999... and 3 are the same number? That if you repeat .99999... to infinity, the difference between that number and 1 is zero, proving they are equal?

Re: Elevator Failure

2012-12-16 20:25 • by Dirk (unregistered)
I made come changes to this comment. Sorry about the incontinence.

Re: Elevator Failure

2012-12-16 21:11 • by moz (unregistered)
397303 in reply to 397189
Mike Nightstky:

Winnipeg minor basketball Association
We have made come changes to your schedule on December 8th we apologize for the incontinence.


First i thought it was the association of kindergarden kids. And the "come" changes were because of the incontinence of some of the members of the minor league.

Either that, or because of the incontinence of one or more of the staff, and at the suggestion of the police.

Re: Elevator Failure

2012-12-17 03:47 • by Iain (unregistered)
397305 in reply to 397263
foo:
Quietust:
Personally, I'd like to think that the "2+5=9" is just a clever way to filter out spambots that try to evaluate mathematical equation CAPTCHAs - a human would enter 7, get the error "must be at least 9", then enter 9 and successfully sign up, but a bot would always enter 7, not knowing to parse the error message for the correct value.
Wow, this will certainly stop spammers for another day or two. Parsing error messages is so much harder than recognizing characters in an image.


You might be on to something. A program to parse an error message in an expected format is trivial but if the error message is, say, a paragraph of unstructured data that contains the hint of the answer then that's going to be a lot harder for programs without Watson level natural language interpretation.
It will also likely be harder for non-native speakers of the language the message is displayed in, so such a system would come at the price that localization would have to be done very well; "all your base belong to us" style translation wouldn't cut it.

Re: Elevator Failure

2012-12-17 04:13 • by tin
397307 in reply to 397195
Martin E:
The Hyper-V submission isn't actually a bug.
When you configure a VM in Hyper-V 2012 to use dynamic RAM it defaults to a max memory of 1048576 Mb (or 1 Tb).
If the VM is running, as this one is, you can't lower the max memory, only raise it and if the VM was configured with the default memory config you can't enter anything but the max amount of memory - which the tooltip will inform you about.


Still a WTF... Like WTF could they not simply have the tooltip tell you that? Or WTF do they default RAM to a freaking terabyte in the first place?

Oh wait - this is a Microsoft product. It's meant to have stupidly worded dialogs and retarded defaults. How else would they sell the new versions in 12 months?

Re: Elevator Failure

2012-12-17 05:09 • by LoremIpsumDolorSitAmet
397308 in reply to 397286
powpowpow:
Brian:
>> "Oh, Visual Firefox, I wish I knew how to quit you! Literally,"

Visual Firefox?

to be fair, Visual firefox has the same problem

"firefox is already running"

Run:

firefox --no-remote -profilemanager

Re: Elevator Failure

2012-12-17 11:21 • by NotHere
397335 in reply to 397307
tin:
Martin E:
The Hyper-V submission isn't actually a bug.
When you configure a VM in Hyper-V 2012 to use dynamic RAM it defaults to a max memory of 1048576 Mb (or 1 Tb).
If the VM is running, as this one is, you can't lower the max memory, only raise it and if the VM was configured with the default memory config you can't enter anything but the max amount of memory - which the tooltip will inform you about.


Still a WTF... Like WTF could they not simply have the tooltip tell you that? Or WTF do they default RAM to a freaking terabyte in the first place?

Oh wait - this is a Microsoft product. It's meant to have stupidly worded dialogs and retarded defaults. How else would they sell the new versions in 12 months?


Or, WTF couldn't they simply fill in the amount for you with a note about how you can't change it.

Re: Elevator Failure

2012-12-17 13:26 • by jay (unregistered)
397348 in reply to 397176
faoileag:
portablejim:
I guess 2+5=9 for very large values of 2 and 5.

Nope. 2.99999 + 5.99999 will never be 9 irrespective of how many 9s you might throw at the two numbers. Close, but never really equal.


Umm, I'm pretty sure that references to "very large values of 2" is a joke. Maybe I'm missing the meta-humor, but you seem to be (a) taking this joke seriously, and (b) somehow leaping to the conclusion that it means 2.99999. Who says that a "large value of 2" is 2.99999? 4 or 5 would also be a large value of 2, no?

Re: Elevator Failure

2012-12-17 13:59 • by Mirza Baig (unregistered)
397351 in reply to 397184
Heh that elevator is definitely in DePaul's CDM building. It's been like that for at least a year.

Re: Elevator Failure

2012-12-17 15:31 • by Anon. (unregistered)
397355 in reply to 397185
Most of the readers here would be surprised at how much can be achieved without an OS. Even with elevators that use microcontrollers, they probably don't have OSes anyway.

Re: Elevator Failure

2012-12-17 16:48 • by sdfh (unregistered)
397356 in reply to 397301
Coyne:
Ben Jammin:
faoileag:
portablejim:
I guess 2+5=9 for very large values of 2 and 5.

Nope. 2.99999 + 5.99999 will never be 9 irrespective of how many 9s you might throw at the two numbers. Close, but never really equal.

I believe that:
1) 2.9 and 5.9 are just small values of 3 and 6 respectively and
2) if you're gonna be a math nazi, you should get it right.

p.s. I still like the original joke



No! NO!! NOOOO!!!! Anything but the 1 does not equal .99999999... argument again! It's been settled! The mathematicians won! Please, no more of this torture!


Seriously, does it really bother you that 3.00000... and 3 are the same number? No? Then why does it bother you that 2.99999... and 3 are the same number? That if you repeat .99999... to infinity, the difference between that number and 1 is zero, proving they are equal?

Oh yeah....
x=0.9999[...]
=> 10x=9.9999[...]
=> 10x-x=9.9999[...]-0.9999[...]
=> 9x=9
=> x=1
QED

Re: Elevator Failure

2012-12-17 16:49 • by Julie (unregistered)
397357 in reply to 397305
Iain:
foo:
Quietust:
Personally, I'd like to think that the "2+5=9" is just a clever way to filter out spambots that try to evaluate mathematical equation CAPTCHAs - a human would enter 7, get the error "must be at least 9", then enter 9 and successfully sign up, but a bot would always enter 7, not knowing to parse the error message for the correct value.
Wow, this will certainly stop spammers for another day or two. Parsing error messages is so much harder than recognizing characters in an image.


You might be on to something. A program to parse an error message in an expected format is trivial but if the error message is, say, a paragraph of unstructured data that contains the hint of the answer then that's going to be a lot harder for programs without Watson level natural language interpretation.
It will also likely be harder for non-native speakers of the language the message is displayed in, so such a system would come at the price that localization would have to be done very well; "all your base belong to us" style translation wouldn't cut it.
Uhm....not to mention how many sites have captcha down the bottom and wierd error messages up the top (and often not sufficiently highlighted to be noticed).

How many times do you fill in a form and keep having it rejected before you finally spot a message that says "Phone number is wrong" (or something)

Re: Elevator Failure

2012-12-19 05:02 • by Neil (unregistered)
I don't have Visual FoxPro (OK, well, we might have had a copy lying around at some point but it was never opened) but I have used FoxPro for Windows, and the web suggests that the issue is similar in that version.

In order for a custom FoxPro window to interact with the user in any way it is generally required to use some sort of READ command, however attempting to quit FoxPro while the READ is active causes the aforementioned error message.

A workaround is to define a custom close handler (such as ON SHUTDOWN QUIT) although then the problem is that the window's Close button becomes disabled. (The Close item on the system menu still works because it's a custom system menu.)

Re: Elevator Failure

2012-12-21 12:37 • by Kuba
397836 in reply to 397249
71:
faoileag:
portablejim:
I guess 2+5=9 for very large values of 2 and 5.
Nope. 2.99999 + 5.99999 will never be 9 irrespective of how many 9s you might throw at the two numbers. Close, but never really equal.
True for any finite number of 9's. But if they are non-terminating decimals (the 9's go on forever) then 2.9[...] + 5.9[...] is strictly equal to 9.
Exactly. Way too many people have not learned in elementary school that the decimal positional system is a particular way of representing numbers. See, on one hand there's that abstract concept of a number. On the other hand, there's a practical way of representing those abstract numbers using a positional sytem.

It so happens that every number out there that has a finite length positional representation in some base, also happens to have an infinite length positional representation in the same base. That's at least two positional representations for one number. The illiterate fools out there have a hard time getting over it, somehow.

2.9[...] + 5.9[...] is strictly equal to 9, because you're adding the integers 3 and 6. That shouldn't be news to anyone.

Re: Elevator Failure

2012-12-21 12:42 • by Kuba
397839 in reply to 397356
sdfh:
Oh yeah....
x=0.9999[...]
=> 10x=9.9999[...]
=> 10x-x=9.9999[...]-0.9999[...]
=> 9x=9
=> x=1
QED
This is idiotic. Just because you saw it on internet, doesn't mean it's not bovine manure. Here's why. You see, the stuff you have up there deals with numbers. Numbers are abstract concepts. They do have some practical representations, one of them being the positional representation. So, when you deal with positional representation, you must always keep in mind what numbers you're talking about.

The first line says:
x=1.

That's it. The rest of it is entirely pointless. You're not proving anything. In line 2, you're multiplying both sides by 10, then rearranging things to remove the mess you made, and since you didn't make any mistakes, you get it back that x=1. Big deal, basic algebra still works. Just because something works doesn't make it useful, or fit for the purpose.

The proof that the positional representation of numbers has a certain property of non-uniqueness for some numbers is a bit more involved and doesn't exactly end up at 5th grade math knowledge.

Re: Elevator Failure

2012-12-21 12:45 • by Kuba
397840 in reply to 397284
Otis:
I'm a computer programmer/sys admin, and I can shed even more light: it doesn't say that the panel has a missing operating system, just that the Intel Boot Agent didn't find one. I get the same message from the Boot Agent every time I reboot my computer -- there is a BIOS error on my computer that prevents me from disabling the Boot Agent, so I have to wait for the network boot agent to fail before booting from disk.
Let's state the obvious: TRWTF is the Intel Boot Agent. Does anyone actually use it on the systems where it's most often idiotically installed (like desktop motherboards)?

Re: Elevator Failure

2012-12-21 15:51 • by JJ (unregistered)
397858 in reply to 397276
pedant:
Zapp Brannigan:
Grammar police are the the buzz killers of the the internet.


Your doing it wrong, if you comment about grammar on the internet, you have to make a grammatical error.

And you failed you look closely at the original post. I won't claim the OP did it on purpose, but it certainly fits the stereotypical grammar flame.

Re: Elevator Failure

2012-12-21 15:52 • by JJ (unregistered)
397859 in reply to 397858
JJ:
pedant:
Zapp Brannigan:
Grammar police are the the buzz killers of the the internet.


Your doing it wrong, if you comment about grammar on the internet, you have to make a grammatical error.

And you failed you look closely at the original post. I won't claim the OP did it on purpose, but it certainly fits the stereotypical grammar flame.

Wow. I just now noticed what I updated. Batting .500 here.

Re: Elevator Failure

2012-12-31 19:57 • by CigarDoug (unregistered)
398199 in reply to 397305
I am downright giddy to be the first to post an xkcd reference on this thread.

http://xkcd.com/810/

Re: Elevator Failure

2012-12-31 20:01 • by CigarDoug (unregistered)
398200 in reply to 397305
Iain:
foo:
Quietust:
Personally, I'd like to think that the "2+5=9" is just a clever way to filter out spambots that try to evaluate mathematical equation CAPTCHAs - a human would enter 7, get the error "must be at least 9", then enter 9 and successfully sign up, but a bot would always enter 7, not knowing to parse the error message for the correct value.
Wow, this will certainly stop spammers for another day or two. Parsing error messages is so much harder than recognizing characters in an image.


You might be on to something. A program to parse an error message in an expected format is trivial but if the error message is, say, a paragraph of unstructured data that contains the hint of the answer then that's going to be a lot harder for programs without Watson level natural language interpretation.
It will also likely be harder for non-native speakers of the language the message is displayed in, so such a system would come at the price that localization would have to be done very well; "all your base belong to us" style translation wouldn't cut it.



I am downright giddy to be the first to post an xkcd reference on this thread.

http://xkcd.com/810/

AND, I really feel like a long-time poster now because I mistook reply for quote! Thanks, forum software!

Re: Elevator Failure

2013-04-25 21:47 • by [insert name here] (unregistered)
406612 in reply to 397183
That was beautiful. *wipes tears from eyes*
« PrevPage 1 | Page 2Next »

Add Comment