Sure, using YoureAllWrong(at)yahoo(dot)com is trivial to detect, but there are an infinite number of schemata that can be used. Just use your imagination.
What a fantastic project. 80 million people is an awful lot of people to be ignored, but exactly what one would expect from a profit-driven corporation like Microsoft. Computing should not be about filling some fat american's wallet, it should be about communication, exchange of ideas and increased freedoms and abilities to learn. I look forward to seeing the linux counter clock up to 80 million over time. (And more, as you say, as you're not the only sector marginalised by the almighty dollar.) YAW.
"Configure my printer" is not a verb, and is inconsistent with the claimed interface design. In fact almost all "verbs" are not verbs, and the noun phrases that are attached to them should be selected from other columns - e.g. "my printer" is in the rightmost devices column.
Money has changed hands. I'd stake my reputation on it. I'd just like to follow the dollar upstream and know who felt threatened? Anyone know of any just-about-to-be-cancelled contracts?
However, I think that an official pro-OSS stance is no better than any other kind of so called "positive discrimination".
OSS source software is on the whole less of a risk as for example the OSS community has been forced to implement interoperability with proprietory standards, and the big commercial names like MS have been desparate to obfuscate their file formats in order to try and evolve away from the OSS equivalents' compatibilities. Therefore they're much less of a lock-in risk. You can take your data with you as you leave, you can't with proprietory formats, you only need to replace the binaries. (Oversimplified of course.)
MS.ICO (icon) files may contain more than one actual bitmap. They may be at various different sizes, and at different colour depths (so that it's possible to have the graphic designer cope best with the limitations of 16 colours or 256 colours, rather than have the OS try and second guess).
And how do you think OS X does it?
Don't inflate yourself so much, you'd be less of a target to puncture. That kind of behaviour can give Mac users a bad image on fora such as slashdot.
If the document is not "in the dock", which I hold as manifestly true, then it is effectively an alias/shortcut for the document. Therefore dragging the alias off the dock and onto the desktop should surely _move_ the alias onto the desktop. I.e. it should not create a new alias as you've suggested -- that in my mind is not as sensible as you suggest, and not intuitive. Clone-on-drag, or copy-on-drag as a default drag behaviour is a very strange paradigm indeed (it doesn't fit in with any desktop metaphor, for a start).
I think eventually it will just be a childish "fuck you, if you won't let me spam you I'm going to make your e-mail useless". It's nearly there. Some of the stuff that passes my home-brew procmail anti-spam rules is so mangled that I, a supposedly intelligent human, can't even work out what it's trying to say! I.e. it's no more than an inbox denial-of-service.
Fixed word lists like this are not the solution to the problem. For every word, there could be thousands of obfuscations. You'd end up with a multi-megabyte word list. It's better to heuristically detect simply whether some kid of obfuscation has been performed. e.g. if you see a word with an '@' in it, does the corresponding word with the '@' replaced by an 'a' match one of your block words. If so, you've immediately blocked all of 'viagr@', 'vi@gra', 'vi@gr@', 'x@nax', 'xan@x', 'x@n@x', and 'v@lium'.
Base64's a good indicator, I use::0 H: * Content-Type: text/(html|plain) * Content-Transfer-Encoding: base64 ${CRAP}
And similarly, I check to see if people are using quoted-printable in order to disguise plain characters. Likewise, the use of HTML entities to encode plain characters is a dead giveaway::0 B: * &#([3-9][0-9]|1[0-2][0-9]); ${CRAP}_1_6
(That one's still experimental, as in theory it has legit uses.)
The clue is not to look at what's being displayed or hidden, it's to look at whether something's trying to be hidden.
Bacause that's useless against mails which contain source-code snippets, or.procmailrc snippets, or unix command line examples, or... . All of those fail a spell-checker miserably.
I blocked a mate's mail just yesterday as it had
$ command < something > something_else because my blocker reckognised that <something> is not a valid HTML tag. Too clever for its own good...
Re:How long before people start gaming the system?
on
IBM vs. Content Chaos
·
· Score: 1
"isn't this only possible with the cooperations of thousands of bloggers?"
hundreds, probably, but nevertheless - there _are_ hundreds, thousands even, of bloggers, and therefore the system can be beaten.
Coordinated bloggers are like a swarm of locusts, or of termites.
Look at slashdot - could you imagine what would happen if 1000 people decided tomorrow to post random nonsense to every thread. Slashdot would be rendered absolutely useless immmediately. Never underestimate the power of numbers.
If I can't decide what format to put structured data into I now just fall back onto what would be XML if only I were to codify the DTD I have in mind.
People who don't care about structuring their data 99% of the moronic web-loggers, are just not going to be interested in XML at all.
However, some do care, typically the ones that do more with their web-presence than just being another blogger on some handle-turning point-and-click blog site. However, whether this minority would adopt such a formal approach is still far from certain.
<dt>, and <dd> are certainly highly semantic, <a> even more so.
Sure, 99% of webpages use the tags for nothing but presentation, but that doesn't mean that their only, or even intended, use is presentation. <table> being the perfect example of something that started off as a semantic structure, but got almost entirely hijacked to do layout instead.
Even more originally called the "Screaming Abdabs" while they were in their formative years at architectural college in London? Several of my relatives delight in telling me about concerts by such bands when they were younger. Bastards.
Mostly right, but the 'scales better' isn't totally true. It doesn't scale so well under load compared to various recent linux kernels due to its use of linear time algoriths rather than constant time or log time ones in code to do with resource management. Too many processes, or too many ports open, and both big BSDs begin to slow down rather. 2.0 and 2.2 linux had similar problems, but 2.4 began to fix them, and 2.6 has them almost all nailed now I believe. Scalability over many hosts is a different issue, of course.
I'm a linux desktop man, but my main world-facing server is FBSD 4, and I have no plans to change either of those OS choices.
Total bullshit. It's such bullshit that it's/not even wrong/, it's just completely meaningless. "Linux" is not a company. "Linux", therefore, does not have a "product". "Linux", therefore cannot charge for said product.
I run 4 different 'Free' OSes, and all of them cost me $0.00.
That's why he said "convicted monopoly", do distinguish them from those, who by neverhaving bee convicted of abusing a monopoly position, are not convicted monopolies.
Slashdot Mirror
Sure, using YoureAllWrong(at)yahoo(dot)com is trivial to detect, but there are an infinite number of schemata that can be used. Just use your imagination.
YAW.
"becoming a full-fledged portal,"
Don't rely on them. If they are going to become a portal, then tey're fucked, just likehe rest of them. Did no-one learn from the last 5 years?
YAW.
Couldn't he just learn basic HTML and edit that from fullscreen DOS using any old editor? (And I mean old, Q or B (boxer) would do fine.)
YAW.
What a fantastic project. 80 million people is an awful lot of people to be ignored, but exactly what one would expect from a profit-driven corporation like Microsoft. Computing should not be about filling some fat american's wallet, it should be about communication, exchange of ideas and increased freedoms and abilities to learn.
I look forward to seeing the linux counter clock up to 80 million over time.
(And more, as you say, as you're not the only sector marginalised by the almighty dollar.)
YAW.
Totally broken.
"Configure my printer" is not a verb, and is inconsistent with the claimed interface design. In fact almost all "verbs" are not verbs, and the noun phrases that are attached to them should be selected from other columns - e.g. "my printer" is in the rightmost devices column.
YAW.
Money has changed hands. I'd stake my reputation on it.
I'd just like to follow the dollar upstream and know who felt threatened? Anyone know of any just-about-to-be-cancelled contracts?
However, I think that an official pro-OSS stance is no better than any other kind of so called "positive discrimination".
OSS source software is on the whole less of a risk as for example the OSS community has been forced to implement interoperability with proprietory standards, and the big commercial names like MS have been desparate to obfuscate their file formats in order to try and evolve away from the OSS equivalents' compatibilities. Therefore they're much less of a lock-in risk. You can take your data with you as you leave, you can't with proprietory formats, you only need to replace the binaries. (Oversimplified of course.)
YAW.
MS .ICO (icon) files may contain more than one actual bitmap.
They may be at various different sizes, and at different colour
depths (so that it's possible to have the graphic designer cope
best with the limitations of 16 colours or 256 colours, rather
than have the OS try and second guess).
And how do you think OS X does it?
Don't inflate yourself so much, you'd be less of a target to puncture. That kind of behaviour can give Mac users a bad image
on fora such as slashdot.
YAW.
If the document is not "in the dock", which I hold as manifestly true, then it is effectively an alias/shortcut for the document. Therefore dragging the alias off the dock and onto the desktop should surely _move_ the alias onto the desktop. I.e. it should not create a new alias as you've suggested -- that in my mind is not as sensible as you suggest, and not intuitive. Clone-on-drag, or copy-on-drag as a default drag behaviour is a very strange paradigm indeed (it doesn't fit in with any desktop metaphor, for a start).
YAW.
It's not funny; it's the reason they're doing it.
YAW.
I think eventually it will just be a childish "fuck you, if you won't let me spam you I'm going to make your e-mail useless". It's nearly there. Some of the stuff that passes my home-brew procmail anti-spam rules is so mangled that I, a supposedly intelligent human, can't even work out what it's trying to say! I.e. it's no more than an inbox denial-of-service.
YAW.
Fixed word lists like this are not the solution to the problem.
For every word, there could be thousands of obfuscations. You'd end up with a multi-megabyte word list. It's better to heuristically detect simply whether some kid of obfuscation has been performed.
e.g. if you see a word with an '@' in it, does the corresponding
word with the '@' replaced by an 'a' match one of your block words.
If so, you've immediately blocked all of 'viagr@', 'vi@gra', 'vi@gr@', 'x@nax', 'xan@x', 'x@n@x', and 'v@lium'.
Base64's a good indicator, I use: :0 H:
:0 B:
* Content-Type: text/(html|plain)
* Content-Transfer-Encoding: base64
${CRAP}
And similarly, I check to see if people are using quoted-printable in order to disguise plain characters. Likewise, the use of HTML entities to encode plain characters is a dead giveaway:
* &#([3-9][0-9]|1[0-2][0-9]);
${CRAP}_1_6
(That one's still experimental, as in theory it has legit uses.)
The clue is not to look at what's being displayed or hidden, it's to look at whether something's trying to be hidden.
YAW.
Bacause that's useless against mails which contain source-code snippets, or .procmailrc snippets, or unix command line examples, or ... . All of those fail a spell-checker miserably.
I blocked a mate's mail just yesterday as it had
$ command < something > something_else
because my blocker reckognised that <something> is not a valid HTML tag. Too clever for its own good...
YAW.
Why did slashcode remove the diaresis from Torbjorn's 2nd o, and the acute from his second e?
Anyway, I get loads of mails with valid non-ASCII.
The world's bigger than just the English-speaking parts of the USA, you know, se\~nor.
YAW.
"anything outside ascii 32 to 126 goes in the bin"
Which is useless if your mates have names like Torbjorn Velen.
YAW.
Yeah. Exactly.
Some companies have an explicit "don't be evil" principle.
(Google, for example, and at least one of the independent
music downloads sites too).
Verisign have an "it it's sleazy, we're up for it" approach instead.
They're contracted to do this, eh? I hope they fail and get their arses sued to hell and back.
YAW.
Obvious:3 .100.gif
http://www.astro.princeton.edu/~mjuric/universe/p
Noone around to see you, you see?
YAW.
"isn't this only possible with the cooperations of thousands of bloggers?"
hundreds, probably, but nevertheless - there _are_ hundreds, thousands even, of bloggers, and therefore the system can be
beaten.
Coordinated bloggers are like a swarm of locusts, or of termites.
Look at slashdot - could you imagine what would happen if 1000 people decided tomorrow to post random nonsense to every thread.
Slashdot would be rendered absolutely useless immmediately.
Never underestimate the power of numbers.
YAW.
If I can't decide what format to put structured data into I now just fall back onto what would be XML if only I were to codify the DTD I have in mind.
People who don't care about structuring their data 99% of the moronic web-loggers, are just not going to be interested in XML at all.
However, some do care, typically the ones that do more with their web-presence than just being another blogger on some handle-turning point-and-click blog site. However, whether this minority would adopt such a formal approach is still far from certain.
YAW.
"HTML tags for RENDERING, not semantics"
In theory, or in practice?
<dt>, and <dd> are certainly highly semantic, <a> even more so.
Sure, 99% of webpages use the tags for nothing but presentation, but that doesn't mean that their only, or even intended, use is presentation. <table> being the perfect example of something that started off as a semantic structure, but got almost entirely hijacked to do layout instead.
YAW.
"the first commercial use will be to track public opinion for companies."
Have they learnt _nothing_ from the google-bombs?
As soon as people find out what algoritm they use, there'll be someone coordinating abuse thereof.
YAW.
Even more originally called the "Screaming Abdabs" while they were in their formative years at architectural college in London?
Several of my relatives delight in telling me about concerts by such bands when they were younger. Bastards.
YAW.
Mostly right, but the 'scales better' isn't totally true. It doesn't scale so well under load compared to various recent linux kernels due to its use of linear time algoriths rather than constant time or log time ones in code to do with resource management. Too many processes, or too many ports open, and both big BSDs begin to slow down rather.
2.0 and 2.2 linux had similar problems, but 2.4 began to fix them, and 2.6 has them almost all nailed now I believe. Scalability over many hosts is a different issue, of course.
I'm a linux desktop man, but my main world-facing server is FBSD 4, and I have no plans to change either of those OS choices.
YAW.
"Linux ('Free OS') is charging for their product"
/not even wrong/, it's just completely meaningless. "Linux" is not a company. "Linux", therefore, does not have a "product". "Linux", therefore cannot charge for said product.
Total bullshit. It's such bullshit that it's
I run 4 different 'Free' OSes, and all of them cost me $0.00.
YAW.
That's why he said "convicted monopoly", do distinguish them from those, who by neverhaving bee convicted of abusing a monopoly position, are not convicted monopolies.
Simple, eh?
YAW.