Indeed. I've done the same, and as a result, I found out from my referrer logs that I'm the number one hit in Google for "fucking string functions". Apparently someone else was having a bad day too.
Hmm... It looks like some of my gripes may have actually been fixed since then. Since as of that writing the last update was three years old, I didn't think anyone actually gave a shit anymore.
I love finding comments like that in code I've inherited, because it means that I am not alone in Hell!. Someone has been here before and was in pain then, too.
Unless there have been some major, major, major advances in OCR that I've missed, I find it's faster just to type the damn things than OCR them and hunt for/correct the OCR Engrish pee.
For most books - lots of pages of nicely printed English text, the main bottleneck for me in OCRing them is the length of time the scanner head takes to return to the start position. If my scanner were smart enough to be bidirectional, it'd be the length of time taken to lift the book up, turn the page and plonk it down on the platten again.
A guy I know at one of the universities in Portugal mentioned that to archive vast quantities of printed media, they're using 15 or 22 megapixel digital backs attached to medium format cameras to record the information on a page, and then do severe hoo-haa to the resulting TIFF.
Is TIFF->plain text OCR any harder than TIFF->PDF/other binary formats? I imagine that if you have sane contrast and knowledge of the kerning, that it couldn't be too hard.
Perhaps the trick is to have the money to afford medium format backs and the endless number of manhours found in a university.
That's beautiful. *) Thank you. I'm so tempted to forward it to the professor who once dissed me for naming variables "Fred" and "Wilma" and never leaving comments to explain what they did (I didn't care because it was a school project and no one else was ever going to see the code). "See just how bad it could have been if I had left comments? Aren't you glad I didn't?"
If said professor ever named a variable "foo" or "bar", or, for that matter, anything AT ALL other than long_descriptive_and_unambigous_name, then he deserves to be shot.
I'm still having mixed feelings on switching to "iter" for my iterator objects instead of "i". Especially because then I end up with silly things like "jter" and "kter" in the inner loops (which I should probably be avoiding with ....but I digress).
I used to maintain a huge C/C++/Pseudo-LISP hybrid application which had a cross-platform GUI layer, including a Motif implementation. The vitriolic comments were LEGION.
We can't blame the laywers for that one -- those doing the sanitization really went overboard. Most of them tended to rip out anything that even remotely suggested that maybe the code was less than perfect.
I'm a total novice programmer.
It's nice to see that I'm not alone in the foulness of my source-code commentary.
Indeed. I've done the same, and as a result, I found out from my referrer logs that I'm the number one hit in Google for "fucking string functions". Apparently someone else was having a bad day too.
Hmm... It looks like some of my gripes may have actually been fixed since then. Since as of that writing the last update was three years old, I didn't think anyone actually gave a shit anymore.
I love finding comments like that in code I've inherited, because it means that I am not alone in Hell!. Someone has been here before and was in pain then, too.
Have a fun day typing?
It only took me about half an hour to type it in. I type fast.
I'd get bored after typing about 3 lines and take a nap.
No wonder you never get anything done, slacker.
OCR!
Spoken like someone who has never actually, you know, used OCR.
I never said it wouldn't have bugs! ;)
Unless there have been some major, major, major advances in OCR that I've missed, I find it's faster just to type the damn things than OCR them and hunt for/correct the OCR Engrish pee.
For most books - lots of pages of nicely printed English text, the main bottleneck for me in OCRing them is the length of time the scanner head takes to return to the start position. If my scanner were smart enough to be bidirectional, it'd be the length of time taken to lift the book up, turn the page and plonk it down on the platten again.
I wouldn't want to try it on source code though.
A guy I know at one of the universities in Portugal mentioned that to archive vast quantities of printed media, they're using 15 or 22 megapixel digital backs attached to medium format cameras to record the information on a page, and then do severe hoo-haa to the resulting TIFF.
Is TIFF->plain text OCR any harder than TIFF->PDF/other binary formats? I imagine that if you have sane contrast and knowledge of the kerning, that it couldn't be too hard.
Perhaps the trick is to have the money to afford medium format backs and the endless number of manhours found in a university.
He said it took him half an hour to type it in. I'm betting it would have taken roughly half a month to correct all the errors OCR would generate.
That's beautiful. *) Thank you. I'm so tempted to forward it to the professor who once dissed me for naming variables "Fred" and "Wilma" and never leaving comments to explain what they did (I didn't care because it was a school project and no one else was ever going to see the code). "See just how bad it could have been if I had left comments? Aren't you glad I didn't?"
If said professor ever named a variable "foo" or "bar", or, for that matter, anything AT ALL other than long_descriptive_and_unambigous_name, then he deserves to be shot.
Knowing her, probably not.
Oh well.
I'm still having mixed feelings on switching to "iter" for my iterator objects instead of "i". Especially because then I end up with silly things like "jter" and "kter" in the inner loops (which I should probably be avoiding with....but I digress).
The worst I could find in the company source tree was a few 'damn's. We fucking suck. :-(
(I thought I'd found a couple at first, but it turns out they were in the Boost and OpenSSL libraries.)
I checked in some code with "anal" in it (as in "Java is anal about exception handling.") and got chewed out.
Computer programming is such a mellowing profession. Sadly I think Open Source is killing the art of comment profanity.
Thank you for that, Jamie. You are a god among men. Or something.
Sniff. Makes me long for a glimpse of the toxic tower. Or at least a quick down-wind whiff of it. (Daddy needs a fourth eye and a third nut.)
Oh, yeah. Fuck you, too.
Warmest Regards,
Some Fuckchop
So tell us the trials you went through to get them to placate you with respect to your views on mail signatures.
And, for a dead pigeon, it sure managed to fart wetly for a long time!
The Windows and Mac weenies really, really didn't want to force .sigs to be preceeded by "-- \n". That took a lot of yelling.
At least Mozilla got that right.
Damn, I don't have enough vulgarity in my comments ...
... damn, I've got bugger-all comments to have vulgarity in.
My favorite.
-bZj
/* Wow, how's this for object fucking oriented? -- jwz. */
Everyone else is mumbling obscenities to themselves, and here you are making wisecracks and signing them.
They say great programmers use code as a medium for self-expression.
Not to mention self-preservation, self-hatred, and self-flaggelation.
Thank you so much for finding that, Jamie. My carefully preserved paper copy continues to hide from me.
I still haven't found my mind, but at least Macs have gotten somewhat better.
- Terry
Wow, I wonder who had so many problems with Motif.
Any of us who tried.
I used to maintain a huge C/C++/Pseudo-LISP hybrid application which had a cross-platform GUI layer, including a Motif implementation. The vitriolic comments were LEGION.
Link? ;-)
Commercial software :-|
I'm surprised that hack had to be sanitized. Or sucks.
We can't blame the laywers for that one -- those doing the sanitization really went overboard. Most of them tended to rip out anything that even remotely suggested that maybe the code was less than perfect.
Now I know where all of the slimy comments went that I've never written into JavaScript source.
I'm so fucking relieved.