I'm amazed that there are people here arguing AGAINST this. If you do not separate things then you make things so much harder to fix and change - even if you're the only person who'll ever look at the code.
There's also a proof-of-concept called PWNtcha (http://sam.zoy.org/pwntcha/ which can automatically work out a large number of common CAPTCHAs (including PHPbb and vBulletin's standard ones ) with well over 90% correct.
CAPTCHAs are NOT the best solution - they're just a band-aid, and they make your site harder to use ( especially for low vision people ). Personally I prefer web-server level blocking of dodgy UA's, IP ranges, POST payloads with something like the wonderful mod_security for Apache, coupled with word filtering ( Spamassassin or just a black list ).
Don't just adblock it! that way the companies never learn that you disagree with them, the ads are still served, they can still report x thousand ad views, and advertising companies keep shitting all over the web.
Vote with your feet, and send them an email telling them why you're changing. This way, at least there's a chance they'll stop doing it.
You do realise that the assholes would find a way around it, don't you? Then we're right back where we started, except now we have the added benefit of assholes carrying free punch cards.
The "one and only" method is the "comparative method" and this does NOT just rely on lexicon and morphological items, it also takes into account typological/structural/grammatical features too.
Next, why *aren't* these characteristics suitable? Many researchers ( Joanna Nichols being the poster scientist ) think that grammatical characters are much more stable than lexicon based characteristics which other computational phylogenetic studies of language evolution have relied on ( including my own work ). Also - Dunn, Terrill & Reesink are very highly regarded Austronesian linguists, so they do have some idea of what they're talking about.
This research is interesting precisely because it uses many data types, and appears to show some signal at MUCH greater time depths than most historical linguists would have predicted.
I think that linguisitics has a major hang-up about "magic bullets" i.e. one or two characteristics proving or disproving the rule. Surely the better approach is to take a statistical analysis. In fact, these methods (phylogenetic techniques from evolutionary biology) that they ( & our lab ) are using are very accurate and capable of detecting these false similarities, and providing an understanding based on ALL the evidence and not one or two examples plucked from a hat.
Anyway, if anyone's interested I wrote a paper earlier this year on these methods applied to the Austronesian language family and early human settlement of the Pacific here (Sorry - PDF! )
As for "being answered" in the literature many times and "sexing up an unreliable methodology by involving computers" - not really no. It's NOT lexicostatistics or glottochronology or any of those 1950s era techniques which are crap, these methods are far more realistic and sensible (and used by evolutionary biologists every day). The computers are not just used because they're sexy, but because analysing all that data by hand is hard.
--Simon
Re:Naming Worms - Virii's pride
on
Name That Worm
·
· Score: 1
It's not going to work - sure the official name will be CME-1234etc but it will still have some k3wl script kiddy name. This means the media will use THAT term instead of the boring CME name. Face it - the next worm to break into the 15 seconds of media attention won't be CME-xyz but CME-xyz AKA "secret elite name that sounds scary and weird to average people".
CMSMatrix.org is another place to go - it doesn't limit the CMSes to PHP/MySQL ones, and allows you to narrow down the list by your required features. Very useful.
Once you've got a shortlist, then drop by opensourcecms, or the website for the actual app, and try out their demo.
Seriously - I've been looking for a good RPG for months now, I tried the demo of DS2 yesterday, and it's slightly better than 1, but it's still just a screensaver with swords.
Where's a GOOD RPG with a storyline? Character Development?
Where's this years System Shock 2? NWN? Eye of the Beholder? Morrowind? Fallout? Planescape Torment? BG2? Anything?
I've tried Sacred, DS2, and about 10 others that just sucked hard, either because of major bugs, or dull gameplay ( click-click-slash-slash-drink healing potion-repeat ).
Any suggestions? please? Or am I just going to have to wait for the Morrowind followup (Oblivion) and reinstall Nethack?
If he had imagination he'd be busy writing the next operating system, or building the next google, or doing pretty much ANYTHING else other than f**king up other peoples lives.
One correction: The Kea can fly. They've got a reputation for being nasty cheeky little buggers who'll quickly strip cars of anything shiny. http://www.nzbirds.com/Kea.html
The flightless bird you're thinking of is the Kakapo. These things weigh about 8 pounds and there's only about 60 of them left. http://www.kakapo.net/en/
There's a large group of us out there who can't run it at all: Install, ok, connect to steam, ok, update, ok, run game.....wait.....dvd drive spins up....wait.... nothing happens.
I've been using simplebackup for a few months now. It's the best free (libre & gratis) solution I've found.
It allows you to backup certain directories, ignore others, archive incrementally, differentially or completely in any one of a number of formats (zip, tar.gz, tar, rar of the top of my head). It also has volume spanning.
Downside - you do have to edit a text config file (so it's not for you average windows user), but it's fairly trivial to set up a few batch files ("back up my pictures now!", "back up my documents now!") etc.
It's written in perl, so you'll need to have active perl installed on windows, but it also works nicely on linux and osx (although you may want to use one of those tars which can deal with the resource forks mentioned earlier in the thread).
Slashdot Mirror
Seconded - & the database design running it all is dodgy too ( and barely documented last time I looked at it ).
A far superior wiki is DokuWiki
Amen.
I'm amazed that there are people here arguing AGAINST this. If you do not separate things then you make things so much harder to fix and change - even if you're the only person who'll ever look at the code.
the same thing we do every night, Pinky... try to take over the world!
There's also a proof-of-concept called PWNtcha (http://sam.zoy.org/pwntcha/ which can automatically work out a large number of common CAPTCHAs (including PHPbb and vBulletin's standard ones ) with well over 90% correct.
CAPTCHAs are NOT the best solution - they're just a band-aid, and they make your site harder to use ( especially for low vision people ). Personally I prefer web-server level blocking of dodgy UA's, IP ranges, POST payloads with something like the wonderful mod_security for Apache, coupled with word filtering ( Spamassassin or just a black list ).
No thanks, I still have nightmares about dying of cholera.
--Simon
bah, Second Reality was better. It's all been downhill since Assembly 93.
... if you work in advertising, please, kill yourself. Do it for the good of everyone.
Don't just adblock it! that way the companies never learn that you disagree with them, the ads are still served, they can still report x thousand ad views, and advertising companies keep shitting all over the web.
Vote with your feet, and send them an email telling them why you're changing. This way, at least there's a chance they'll stop doing it.
You do realise that the assholes would find a way around it, don't you? Then we're right back where we started, except now we have the added benefit of assholes carrying free punch cards.
...Ford was also criminally prosecuted for wreckless homicide in Indiana...
Huh, I thought the wrecks were the problem.
The best webcomic I've seen in years is Alien Loves Predator ( Guest stars Jesus too ).
Wonderful stuff.
The "one and only" method is the "comparative method" and this does NOT just rely on lexicon and morphological items, it also takes into account typological/structural/grammatical features too.
Next, why *aren't* these characteristics suitable? Many researchers ( Joanna Nichols being the poster scientist ) think that grammatical characters are much more stable than lexicon based characteristics which other computational phylogenetic studies of language evolution have relied on ( including my own work ). Also - Dunn, Terrill & Reesink are very highly regarded Austronesian linguists, so they do have some idea of what they're talking about.
This research is interesting precisely because it uses many data types, and appears to show some signal at MUCH greater time depths than most historical linguists would have predicted.
I think that linguisitics has a major hang-up about "magic bullets" i.e. one or two characteristics proving or disproving the rule. Surely the better approach is to take a statistical analysis. In fact, these methods (phylogenetic techniques from evolutionary biology) that they ( & our lab ) are using are very accurate and capable of detecting these false similarities, and providing an understanding based on ALL the evidence and not one or two examples plucked from a hat.
Anyway, if anyone's interested I wrote a paper earlier this year on these methods applied to the Austronesian language family and early human settlement of the Pacific here (Sorry - PDF! )
As for "being answered" in the literature many times and "sexing up an unreliable methodology by involving computers" - not really no. It's NOT lexicostatistics or glottochronology or any of those 1950s era techniques which are crap, these methods are far more realistic and sensible (and used by evolutionary biologists every day). The computers are not just used because they're sexy, but because analysing all that data by hand is hard.
--Simon
It's not going to work - sure the official name will be CME-1234etc but it will still have some k3wl script kiddy name. This means the media will use THAT term instead of the boring CME name. Face it - the next worm to break into the 15 seconds of media attention won't be CME-xyz but CME-xyz AKA "secret elite name that sounds scary and weird to average people".
CMSMatrix.org is another place to go - it doesn't limit the CMSes to PHP/MySQL ones, and allows you to narrow down the list by your required features. Very useful.
Once you've got a shortlist, then drop by opensourcecms, or the website for the actual app, and try out their demo.
--Simon
Seriously - I've been looking for a good RPG for months now, I tried the demo of DS2 yesterday, and it's slightly better than 1, but it's still just a screensaver with swords.
Where's a GOOD RPG with a storyline? Character Development?
Where's this years System Shock 2? NWN? Eye of the Beholder? Morrowind? Fallout? Planescape Torment? BG2? Anything?
I've tried Sacred, DS2, and about 10 others that just sucked hard, either because of major bugs, or dull gameplay ( click-click-slash-slash-drink healing potion-repeat ).
Any suggestions? please? Or am I just going to have to wait for the Morrowind followup (Oblivion) and reinstall Nethack?
Of course he doesn't, he's a virus 'Coder'.
If he had imagination he'd be busy writing the next operating system, or building the next google, or doing pretty much ANYTHING else other than f**king up other peoples lives.
--Simon
The major one is "Magic Quotes" which automatically escapes GET/POST/Cookie data, in the hope of protecting against SQL Injections.
For example - if I entered this into the form field 'name':
Hack! "
Php would store it as:
Hack! \"
However, most professional/good PHP programmers I know make a point of turning this off, as it's just damn annoying if you know what you're doing.
--Simon
Ah crap. Godwin's law.
THREAD CLOSED - Nothing to see here
One Tree Hill, Greenlane, Royal Oak and Three Kings aren't beaches. They're suburbs of Auckland, New Zealand, where a certain Mr. Goodger comes from.
(and me!)
Ki ora, mate.
One correction: The Kea can fly. They've got a reputation for being nasty cheeky little buggers who'll quickly strip cars of anything shiny. http://www.nzbirds.com/Kea.html
The flightless bird you're thinking of is the Kakapo. These things weigh about 8 pounds and there's only about 60 of them left. http://www.kakapo.net/en/
Gravity's only a theory too. This is why I'm posting this whilst floating in mid air.
I don't think 'theory' means what you think it means.
at least you can play it.
? s=&threadid=176998&perpage=15&pagenumber=1 or http://www.halflife2.net/forums/showthread.php?s=6 9940b3af0c4026c73f52b06585fe6ba&t=52983
There's a large group of us out there who can't run it at all: Install, ok, connect to steam, ok, update, ok, run game.....wait.....dvd drive spins up....wait.... nothing happens.
See http://www.steampowered.com/forums/showthread.php
Seems to be a problem with DVD+R/RWs.
Or, for that matter - Edgar A. Poe's Ballon Hoax (1844) which preceded Verne by a good 20 years (From Earth to the Moon wasn't published until IIRC 1865).
Not sure if Verne had read Poe, by Poe did have a very good reputation in Europe, so it's highly likely.
I've been using simplebackup for a few months now. It's the best free (libre & gratis) solution I've found.
It allows you to backup certain directories, ignore others, archive incrementally, differentially or completely in any one of a number of formats (zip, tar.gz, tar, rar of the top of my head). It also has volume spanning.
Downside - you do have to edit a text config file (so it's not for you average windows user), but it's fairly trivial to set up a few batch files ("back up my pictures now!", "back up my documents now!") etc.
It's written in perl, so you'll need to have active perl installed on windows, but it also works nicely on linux and osx (although you may want to use one of those tars which can deal with the resource forks mentioned earlier in the thread).
Hell no! I tried this a few months ago (wasn't thinking & just double clicked the icon...
Result: open...close...open...close... repeat...
No escape until I hard rebooted.