Welcome to the United States. For hysterical raisins, our government is divided into many parts: there is the Federal government, and there are 50 state governments, and you're probably covered by a county and a municipality, too. They are all separate entities. There is no particular requirement for their policies to be similar.
Because they think there's more than one, duh. Little do they know that the One True Math will damn them to the firey pit of Sociology for their blasphemy!
if you need <table> to lay out your page (and not for actual tables), you're not using CSS fully;
there's no good reason not to bash out an alt attribute every time you bash out an img tag; and
there's no good reason not to specify the language you're using.
These are small things in that they're relatively easy to work around, but treating that as the important thing is missing the basic point: all of the crap of the browser wars era came about because the spec was not simple, precise, and demanding, so people used all the wiggle room they had and created an enormous, unparseable, inconsistent mess. Common use of language makes it seem like demanding that people comply exactly with standards rather than just "mostly" is demanding a lot, but it really isn't in this case: the standard is there, it's clear, it's easy to follow, and there are freely available tools that will show you instantly what you're getting wrong. And the consequences of letting it slip (which always, always, always leads via human nature straight to habitual errors) are really ugly.
Your argument is against the ADA's facilities requirements as a whole, then, and that's fine. I even agree with you: as a libertarian, I think the law ought to stay out of how you choose to build your building or make your web site. I read your initial post to be asserting that you agreed with the court's decision, which is that the web is out of the scope of the ADA, on the grounds that it's unreasonable to ask someone to redo a web site (but, by omission, reasonable to ask them to build a wheelchair ramp).
That said, he who writes a web site usable by blind people is a member of that class technically referred to as "inconsiderate assholes". It's Not That Hard.
Please explain how the needs of disabled people are harder to accomodate on the web, or for that matter anything other than simple and predictable. Blind people generally use text readers or braille displays to access the web, so they can use your web site just fine as long as it is properly written. Problems appear when you do stupid shit like omitting ALT tags for significant images, misusing HTML tags for visual formatting instead of specifying document structure, and so on. Fixing these problems takes at most a few minutes per page. The cost of fixing either a small website or a database-driven site like an airline's is pretty trivial, and in any case much less than the equivalent costs in the physical world (elevators, ramps, extra bathrooms, and so on).
This is a fallacy. Saudia Arabia, source of the WTC terrorists, is far from a poor country. Poor people have historically been pretty ineffectual except when they banded together in large groups with lots of help from rich people.
Plenty of other sites have the news. If the Times hides theirs, they just get fewer eyeballs. (I only buy them for the crossword puzzle anyway. News comes from Google.
Enforcing software licensing can't work unless it's impossible to run software that isn't signed. Some of us would like to run software that does not come from Microsoft and their friends.
its impossible to estimate at least 4 of the numbers in the equation
I disagree that it is "impossible to estimate". Those numbers come naturally out of scientifically developed models of how planets evolve and how life evolves. Even though we are only in close contact with one planet, we have an increasing amount of long-distance observation of other planets and enough control over our environment that we can effectively simulate a lot of what might happen on planets that evolved differently from ours. The Drake equation does not hand us an answer, but it is a much more useful tool for exploring possible answers than just "guessing based on a single sample" as you seemed to be saying.
The Drake equation doesn't give us a probability of anything. It just kind of states what we would need to know before we could take a guess.
Well, that and what I said: it lets us mix controlled, limited, and hopefully educated guesses with solid facts to get something that is better than a single wild-ass guess.
If you think the Drake equation is about "predict[ing] anything from a sample of 1", you don't understand it. The Drake equation lets us estimate the probability that we will run into intelligent life based on several other numbers that are easier to estimate well. That's the whole point: right now, we have only one world that we've explored thoroughly, so we want to figure out what else besides that sample of 1 we can use to predict things.
Yes, I failed to think about general macro lossage the first time around. I'm still going to claim that it's the sanest way of doing without inlines, since I've never seen a preprocessor that will inline __inline__ functions for you, and writing your own is unlikely to be worth the trouble. Go ahead and use macros, just beware of the pitfalls.
Re:My first warwalk...
on
Wartrapping?
·
· Score: 2
Certainly you don't want to pick a fight by nmapping, and sometimes there's no reasonable way to find out, but it's likely that you can find out who to talk to by sshing out to some machine you control and then looking at your reverse DNS or the whois for your IP address.
...the poster is already talking about ways to transform the text of the code, which almost certainly won't pay attention to type checking anyway. I'd rather have a macro that gets transformed when the code is compiled than run a Perl script to mangle the code once and have to maintain it thereafter. (Yes, I forgot about type checking when I wrote my reply.)
Oh, those warwalkers are so evil! They're actually accepting our offer to communicate! Let's demonstrate our moral virtue by creating false invitations to use our network backed with bogus MAC addresses so they don't catch on! Bleh.
Re:My first warwalk...
on
Wartrapping?
·
· Score: 2
The right thing to do is call the company and ask to speak to their IT person. Say something like "I noticed you have a wireless access point without any access controls on it. Are you intentionally sharing your network with passers-by?" If they are, thank them and ask nicely if you can chalk their sidewalk, and if they aren't, give them the gist of how to secure it (WEP for a "No Trespassing" sign, IPSec for real security).
From warwalking on my way home from the train station in my city, I know that there are at least a dozen wireless networks in use, and the ones you'd expect to be secured (banks and other paranoid corporations) are, and the ones you wouldn't expect (mine, the library's) aren't, and that's more or less the way it should be. In my experience, this idea that there's all these evil pirates taking advantage of clueless people with open networks is bogus, and it's a real shame that it's leading to things like honeypots and jammers that stomp on intentionally shared networks.
"Crooks", houses, and wireless
on
Wartrapping?
·
· Score: 5, Insightful
Using weak metaphors to argue about computer security gets really old. A closed door, locked or not, is an indication that you're not supposed to go in unless the owner wants you there. Likewise, a WEP-protected network may be easy to get into, but the use of WEP is a sign that you're not wanted there. And just like a house with an Open House sign on the front, my wireless network has no such "go away" signal because I want people to use it. (Of course, just like an Open House sign does not mean "please burn my house down", my 802.11b base station is not an invitation to abuse my network, just an opportunity.)
Unless I'm on crack, inlining all calls to a function is pretty much the same as replacing the function with a macro. The main difference between inlining and macros would be that the compiler can decide not to inline a function if it doesn't make sense (e.g., if the function is recursive and the compiler isn't up to iterative-izing it), whereas with macros you just can't do recursion safely.
"If you don't have anything to say, say it in a condescending tone." I'd be happy to continue the conversation about practical system administration as soon as you're ready to rejoin it. Since I had the last post with actual content, it's up to you now. I won't bother to reply to any more personal attacks from you.
I completely fail to be impressed with the importance of a security fix whose primary function is to turn the "convince a user to connect to an external SMB server, then sniff their password" attack into the "convince a user to connect to an external SMB server, then sniff and crack their password" attack. Users, in the main, have terrible passwords. Giving away a hash is little better than giving away the password
Kerberos provides some sort of authentication infrastructure, but it sure wasn't Public Key Intrastructure last I checked, but rather pre-shared keys for DES (in the case of K4) and other symmetric cyphers (with K5). SSL and IPSec, on the other hand, do work quite nicely with public keys, even if most people don't seem to want to bother using client certificates or even verifying server certificates.
I replied to a comment that said, in essense, "it's not Microsoft that's lame for breaking Samba compatibility, it's Samba that's lame for not keeping up with Microsoft's l33t security updates" by saying, in essense, "Microsoft's security update is not so l33t and has the added effect of breaking stuff, so Microsoft is lame after all". You have failed to explain how this is off topic or, for that matter, wrong.
A low UID is not a license to be rude, and snide comments are especially out of place when they are misinformed. I suggest a calmative, followed by a period of quiet reflection.
Encrypting SMB passwords is the wrong place to applys security. There is a gain that is not entirely trivial, but bad guys who can sniff your network can still capture your data, run dictionary attacks on your passwords, and so on. If you don't trust your network's physical security, there is no solution that makes a big difference except encryption at or below the transport layer (i.e., SSL/TLS or IPSec). Enforcing a weak security solution while breaking functionality is a double lose.
But you can still read your video card's framebuffer (admittedly more slowly than you can write to it) even if the video is never in the OS's window buffers.
Prices for copies go to zero, but the actual act of creation continues to be worth money. If you want something, do it; if you want something done, pay for it; and in either case, everyone gets the final benefit. In practice, this probably means that large groups of users would have to get together to pay for the software they want to be written and fixed. The compensation for this is that software never goes away: instead of having to start from scratch because you're not the company that "owns" the program, you can build on an existing program that's almost perfect.
Slashdot Mirror
Welcome to the United States. For hysterical raisins, our government is divided into many parts: there is the Federal government, and there are 50 state governments, and you're probably covered by a county and a municipality, too. They are all separate entities. There is no particular requirement for their policies to be similar.
That actually seems to support the contention that their web sites are more campaign ads than information for constituents.
Because they think there's more than one, duh. Little do they know that the One True Math will damn them to the firey pit of Sociology for their blasphemy!
I contend that
These are small things in that they're relatively easy to work around, but treating that as the important thing is missing the basic point: all of the crap of the browser wars era came about because the spec was not simple, precise, and demanding, so people used all the wiggle room they had and created an enormous, unparseable, inconsistent mess. Common use of language makes it seem like demanding that people comply exactly with standards rather than just "mostly" is demanding a lot, but it really isn't in this case: the standard is there, it's clear, it's easy to follow, and there are freely available tools that will show you instantly what you're getting wrong. And the consequences of letting it slip (which always, always, always leads via human nature straight to habitual errors) are really ugly.
[that is all]
Your argument is against the ADA's facilities requirements as a whole, then, and that's fine. I even agree with you: as a libertarian, I think the law ought to stay out of how you choose to build your building or make your web site. I read your initial post to be asserting that you agreed with the court's decision, which is that the web is out of the scope of the ADA, on the grounds that it's unreasonable to ask someone to redo a web site (but, by omission, reasonable to ask them to build a wheelchair ramp).
That said, he who writes a web site usable by blind people is a member of that class technically referred to as "inconsiderate assholes". It's Not That Hard.
Please explain how the needs of disabled people are harder to accomodate on the web, or for that matter anything other than simple and predictable. Blind people generally use text readers or braille displays to access the web, so they can use your web site just fine as long as it is properly written. Problems appear when you do stupid shit like omitting ALT tags for significant images, misusing HTML tags for visual formatting instead of specifying document structure, and so on. Fixing these problems takes at most a few minutes per page. The cost of fixing either a small website or a database-driven site like an airline's is pretty trivial, and in any case much less than the equivalent costs in the physical world (elevators, ramps, extra bathrooms, and so on).
This is a fallacy. Saudia Arabia, source of the WTC terrorists, is far from a poor country. Poor people have historically been pretty ineffectual except when they banded together in large groups with lots of help from rich people.
Plenty of other sites have the news. If the Times hides theirs, they just get fewer eyeballs. (I only buy them for the crossword puzzle anyway. News comes from Google.
I don't see why the cracked version couldn't have access to the same keys the non-cracked version has.
Enforcing software licensing can't work unless it's impossible to run software that isn't signed. Some of us would like to run software that does not come from Microsoft and their friends.
I disagree that it is "impossible to estimate". Those numbers come naturally out of scientifically developed models of how planets evolve and how life evolves. Even though we are only in close contact with one planet, we have an increasing amount of long-distance observation of other planets and enough control over our environment that we can effectively simulate a lot of what might happen on planets that evolved differently from ours. The Drake equation does not hand us an answer, but it is a much more useful tool for exploring possible answers than just "guessing based on a single sample" as you seemed to be saying.
Well, that and what I said: it lets us mix controlled, limited, and hopefully educated guesses with solid facts to get something that is better than a single wild-ass guess.If you think the Drake equation is about "predict[ing] anything from a sample of 1", you don't understand it. The Drake equation lets us estimate the probability that we will run into intelligent life based on several other numbers that are easier to estimate well. That's the whole point: right now, we have only one world that we've explored thoroughly, so we want to figure out what else besides that sample of 1 we can use to predict things.
Yes, I failed to think about general macro lossage the first time around. I'm still going to claim that it's the sanest way of doing without inlines, since I've never seen a preprocessor that will inline __inline__ functions for you, and writing your own is unlikely to be worth the trouble. Go ahead and use macros, just beware of the pitfalls.
Certainly you don't want to pick a fight by nmapping, and sometimes there's no reasonable way to find out, but it's likely that you can find out who to talk to by sshing out to some machine you control and then looking at your reverse DNS or the whois for your IP address.
...the poster is already talking about ways to transform the text of the code, which almost certainly won't pay attention to type checking anyway. I'd rather have a macro that gets transformed when the code is compiled than run a Perl script to mangle the code once and have to maintain it thereafter. (Yes, I forgot about type checking when I wrote my reply.)
Oh, those warwalkers are so evil! They're actually accepting our offer to communicate! Let's demonstrate our moral virtue by creating false invitations to use our network backed with bogus MAC addresses so they don't catch on! Bleh.
The right thing to do is call the company and ask to speak to their IT person. Say something like "I noticed you have a wireless access point without any access controls on it. Are you intentionally sharing your network with passers-by?" If they are, thank them and ask nicely if you can chalk their sidewalk, and if they aren't, give them the gist of how to secure it (WEP for a "No Trespassing" sign, IPSec for real security).
From warwalking on my way home from the train station in my city, I know that there are at least a dozen wireless networks in use, and the ones you'd expect to be secured (banks and other paranoid corporations) are, and the ones you wouldn't expect (mine, the library's) aren't, and that's more or less the way it should be. In my experience, this idea that there's all these evil pirates taking advantage of clueless people with open networks is bogus, and it's a real shame that it's leading to things like honeypots and jammers that stomp on intentionally shared networks.
Using weak metaphors to argue about computer security gets really old. A closed door, locked or not, is an indication that you're not supposed to go in unless the owner wants you there. Likewise, a WEP-protected network may be easy to get into, but the use of WEP is a sign that you're not wanted there. And just like a house with an Open House sign on the front, my wireless network has no such "go away" signal because I want people to use it. (Of course, just like an Open House sign does not mean "please burn my house down", my 802.11b base station is not an invitation to abuse my network, just an opportunity.)
Unless I'm on crack, inlining all calls to a function is pretty much the same as replacing the function with a macro. The main difference between inlining and macros would be that the compiler can decide not to inline a function if it doesn't make sense (e.g., if the function is recursive and the compiler isn't up to iterative-izing it), whereas with macros you just can't do recursion safely.
"If you don't have anything to say, say it in a condescending tone." I'd be happy to continue the conversation about practical system administration as soon as you're ready to rejoin it. Since I had the last post with actual content, it's up to you now. I won't bother to reply to any more personal attacks from you.
I completely fail to be impressed with the importance of a security fix whose primary function is to turn the "convince a user to connect to an external SMB server, then sniff their password" attack into the "convince a user to connect to an external SMB server, then sniff and crack their password" attack. Users, in the main, have terrible passwords. Giving away a hash is little better than giving away the password
Kerberos provides some sort of authentication infrastructure, but it sure wasn't Public Key Intrastructure last I checked, but rather pre-shared keys for DES (in the case of K4) and other symmetric cyphers (with K5). SSL and IPSec, on the other hand, do work quite nicely with public keys, even if most people don't seem to want to bother using client certificates or even verifying server certificates.
I replied to a comment that said, in essense, "it's not Microsoft that's lame for breaking Samba compatibility, it's Samba that's lame for not keeping up with Microsoft's l33t security updates" by saying, in essense, "Microsoft's security update is not so l33t and has the added effect of breaking stuff, so Microsoft is lame after all". You have failed to explain how this is off topic or, for that matter, wrong.
A low UID is not a license to be rude, and snide comments are especially out of place when they are misinformed. I suggest a calmative, followed by a period of quiet reflection.
Encrypting SMB passwords is the wrong place to applys security. There is a gain that is not entirely trivial, but bad guys who can sniff your network can still capture your data, run dictionary attacks on your passwords, and so on. If you don't trust your network's physical security, there is no solution that makes a big difference except encryption at or below the transport layer (i.e., SSL/TLS or IPSec). Enforcing a weak security solution while breaking functionality is a double lose.
But you can still read your video card's framebuffer (admittedly more slowly than you can write to it) even if the video is never in the OS's window buffers.
Prices for copies go to zero, but the actual act of creation continues to be worth money. If you want something, do it; if you want something done, pay for it; and in either case, everyone gets the final benefit. In practice, this probably means that large groups of users would have to get together to pay for the software they want to be written and fixed. The compensation for this is that software never goes away: instead of having to start from scratch because you're not the company that "owns" the program, you can build on an existing program that's almost perfect.