...will start writing horrible monsters running hundreds and thousands of threads, and their creations will suffer from all other shortcomings of that decision.
R4-2 A public-private partnership should perfect and accelerate the adoption of more secure router technology and management, including out-of-band management. R4-3 Internet service providers, beginning with Tier 1 companies or R4-10 The private sector should consider including in near-term research and development priorities, programs for highly secure and trustworthy operating systems. If such systems are developed and successfully evaluated, the Federal government should accelerate procurement of such systems. in software code development, including processes and procedures that diminish the possibilities of erroneous code, malicious code, or trap doors that could be introduced during development. R4-17 The PCIPB s Awareness Committee, in cooperation with lead agencies,
They do realize that "trustworthy computing" name was originared by Microsoft, and has absolutely nothing to do with computer user's security and everything with software companies' "security" from the user, whoever he might be? Don't they?
IRC doesn't support a lot of the functionality of AIM, it's much less convenient for most of what I use AIM for
What functionality is in AIM and not in IRC? By IRC, of course, I mean "what can be reasonably done with it using existing clients and minimal scripting", not "what comes by default with the only IRC client I have seen 5 years ago".
and maybe I don't always want everyone to know my IP address?
Server doesn't have to disclose it, and you can proxy the connection if you are paranoid. Not that it matters for any reasonable purpose that financial institutions may have.
Financial companies should just give up "some fatass company should sell us things" and look how IRC is superior to all those closed and semi-closed "messengers".
There is no compatibility problems between chat systems just like there is no compatibility problems with email. It's just closed email systems already disappeared, and closed messaging system are still there -- but people who rely on them deserve to suffer from their closedness.
Re:Why China wants stake in Taiwan so bad
on
Upcoming Cyberwars
·
· Score: 2
But I think you're being overly pessimistic otherwise. Command of the air in Kosovo may not have led directly to the toppling of Milosevic, but let's remember that it was used from 16,000+ feet (out of the range of Serbian surface-to-air missiles), largely out of the understandable American desire to avoid unnecessary casualties.
That was a typical use of military force in the time of peace, when civilians' life is valuable -- it works until a real war starts. And, of course, it only supported Milosevich by providing a convenient foreign enemy, ultimately delaying his removal from power by then disorganized opposition, that couldn't use situation either way at the moment. Same with some later events and another weak leader that used foreign "air attacks" against civilians in his country as a crutch for his political career. He is doing fine so far.
I think a lot of the pessimism over the value of air superiority otherwise is a hangover from Vietnam, where major air offensives are generally believed to have been ineffective. What people tend to forget is that the Vietcong had an incredibly effective sigint network, and - from evesdropping on unsecured American communications, and particularly the ground- communications of air Force maintenance personel - often had quite advance warning on the location and timing of air assaults.
When the US got its act together and began to secure its communications, the effectiveness of bombing in Vietnam (and Cambodia) increased immesurably.
Vietnam was a special case in a lot of areas, so I wouldn't read much into that.
Early detection of airplanes flying from remote locations however doesn't depend on interception of communications now -- radars and even satellites do it much better. What is more important, when you are not completely out of resources, it simply doesn't matter much -- just defend what is worth being defended, and accept minor losses caused by whatever will slip through those defenses, or in the areas that are defended poorly. When you are at war you can't guard every building and civilian, but if losses from bombing are higher than, say, potential losses from enemy saboteurs (that are nearly impossible to prevent completely but never can be too high either), enemy will prefer bombing, and you will prefer tying up and wasting enemy's resources in actual battles rather than wasting yours ones on defending large areas that a bomber can reach.
Now I know that if someone will try to bring a hacksaw in San Francisco and cut the truss on the Golden Gate Bridge, he will be stopped at the airport when he will be flying to SFO!
Re:Why China wants stake in Taiwan so bad
on
Upcoming Cyberwars
·
· Score: 2
In any event -- its likely that air superiority will continue to be the decisive factor in contemporary military conflict -- and China doesn't have remarkably good aerospace airforce and knows it.
"Air superioruty"'s overblown value is a doctrine originated by Germans (in their pathetic attempts to attack Britain from the air alone) and picked up by US as a way to sell themselves an idea of having a war with extremely low casualties on their side. It's bullshit. Any airplane outside of the area of a battle on the ground or sea is nothing but a very large house fly -- an annoying target. The damage that it can cause to anything military is negligible, the possibilities to turn it away by all kinds of attacks, even without actually destroying, are endless, and the best it can do is to bomb cities full of civilians to at best demoralize and usually merely annoy the population.
In a battle on the ground or sea the air force can do a lot of assistance and may even determine the outcome of the battle, but don't kid yourselves -- there must be a battle there in the first place, and a lot of people will have to die there, with or without airplanes flying over them.
War sucks, and if you go there, you are most likely to die. No doctrine changes that.
Then try raising a family. You don't have the luxury of morals when you have children crying themselves to sleep because they are hungry. I'd stick a knife in the next person if it meant the difference between feeding my family and upholding my morals, and I don't apologize for it.
Solution: kill yourself and your spouse if you have any. Then your kids will be orphans, and whatever food they will get in an orphanage will be far better than whatever you will get them by killing a random person.
And if your situation is not as desperate to do that it's quite natural to hold you to the same moral standards as the rest of humankind.
C _is_ the state of the art of procedural languages for over 30 years. It's not like such a simple thing has a lot of room for improvement, there are other areas where new languages can be created, but honestly how many ways are there to do things like preprocessor, functions, variables, etc.?
OO languages' authors may feel that they are doing something more "advanced" but in fact they are working on a completely ortogonal area of development. And most of them are far from C elegance (ex: Stroustrup doesn't even understand C design properly, so C++ is even more inconsistent than what its origin would suggest, and I don't even consider a rotting pile of shit that its "standard" libraty is, to be a part of language), or are simply badly designed (ex: Java), or are not languages but eclectic messes made by including specific librariers' and object models design into the language itself (ex: C#).
This is the area where we can use a lot of progress until it will reach the state where we can keep call the same thing "state of the art" for 30 years, but I won't hold my breath -- OO language design is dead, everyone is just making "OO" languages as various vehicles to promote their narrow-minded ideas. So I won't be surprised if Stroustrup's mess will remain the most useful semi-OO language for the next 30 years, too (but those libraries HAVE TO GO, and so should the attitude that students should learn that atrocity without studying C first).
Can I have neo-nazi propaganda quoted at length by Reuters or AP? What about an article "Church of Scientology said:", quote of their self-promoting speech, then "some people disagree".
The problem is not that journalists don't offer _their_ opinions, it's that they choose material that already is a complete bullshit (come on, Microsoft and Hollywood are trying to justify a creation of monopolistic cartel, what _is_ legitimate here?), and just relay it without even looking for something that will give a reader an idea if it's actually something valid. Only tabloids operate like this, but the difference is that tabloids intentionally look for bullshit to publish and have a reputation for doing so.
The reporting about the number of people attending the Klan meeting is reporting about fact. You can't change the facts, they just exist. However the article we are talking about doesn't deal with any facts that happened, it reports about arguments and opinions. And it heavily promotes the opinions of industry leaders, making everything looks like a nothing out of the ordinary bargain -- Hollywood pays with content, gets broadband, tech industry pays with DRM, gets Hollywood making content available in new and exciting proprietary format, legislators are doing fine job mediating the process, everyone is happy but tech industry gets a bit shorter end of the stick.
This is, of course, a lie, and real issue is that "Hollywood" is trying to ptessure technology industry into oppressing the users and basically demands to sacrifice all development in computer technology outside of the walls of large companies that can form a DRM cartel. But if that was said, article would be "biased" because it will display movie industry as complete assholes. So nothing is said about that. Everything is fine. Don't forget to buy new and exciting products, with time bombs inside.
You seem to suffer from the mistaken notion that 'fair and balanced' must yield an appearance of equal legitimacy, or must include equal promotion of both sides of an issue.
In modern journalism "fair and balanced" means exactly that -- if one side appears to be less legitimate, journalist must write an article heavily skewed toward it, so it will be "equally promoted". With KKK this is a taboo because of "PC" problem, attempt to whitewash them will cause immediate outrage, but with industry crooks it isn't -- general population knows nothing about all this, it sees things through the eyes of the imaginary "balanced" person that journalist creates for his story. And that "person" happens to quote industry heavyweights much more than, say, consumer advocates -- for obvious reason that without this "fairness" industries' "leaders" come out as evil, self-serving companies using consumers as hostages and shutting out the competition.
The problem is, trying to be write something "balanced" when talking about something that evil ends up being just a mouthpiece for them. There is nothing at all that can justify "Palladium" (or KKK if that matters) without saying a lie.
They either are against us, or are ready to sell our rights as a piece of some bargain. Consumer advocates, free/open software developers and civil rights groups must do everything to affect the outcome of this because it's their interests that are at stake. Both Intel and Disney can kiss our asses and have Enron-style bankruptcy for all I care.
Agent Smith may be an extreme misanthrope, but in this case, his expression applies pretty well. There are things where people don't WANT to deal with people on the other end of the line -- those things are detrtministic, and unless something is horribly broken, a simple machine easily gives all the necessary information, performs simple transactions, etc.
What is my account's balance? Is some payment past due? When will the package arrive? Is there an outage on my service? How much does this thing cost? Will you reimburse me for this? Do you block ports? Where are you located? Transfer my money to another account. Send this to my address. Process this application. Get a credit card payment.
You don't need a human to answer those things. Actually you DO need a human if you want him to answer or perform them wrong, to pretend that numbers are not what they are, that there is no problems when there definitely are some, to claim that they don't do something, to blame someone else, to pretend to be annoyed, to throw a lot of irrelevant offers, to have problems with handset, or (my favorite) not understanding someone's accent. Human ingenuity is a great help when simple answer exists, but should be hidden at all costs.
For all those simple things one doesn't need humans, he needs a web site. Or, for people that don't use computers, or are away from them, a simple phone menu. And it will serve its purpose much better than a bunch of minimum-wage drones with headsets ever would.
Humans are needed for other things, to answer questions that are not asked 65537 times a day, to explain meanings of obscure things that someone's customer may or may not know but it's hard to list all of them on a web site or especially in a phone menu. Humans can make decisions, ask questions based on things that are hard to place in a menu, but those things are far beyond what "AI" can do anyway. It would be foolish to try to replace humans there -- in fact anything that needs "AI" actually needs a human because at this level of technology development AI simply doesn't exist as anything usable. In those cases a proposal to use a machine to do a human's job would be so insane, only Harvard MBA would ever think of such a thing.
But for the rest of things, I would really prefer a machine.
Have you ever tripped over a sprinker head? It happens. When it does, who replaces it?
Whoever I'll call from a company that does sprinklers repair. I don't think that $100/mo is a package that both "web-enables" sprinklers and gives an owner free repairs while "not web-enabled" owners have to pay for repairs.
A security alarm is great, but wouldn't it be better if a security officer actually came out to your house and tried to catch the burglar if you weren't home?
Remote monitoring of security systems is a separate service, and it doesn't cost $100/mo either.
If so, how will I see the difference between 256Kbps and anything above? Unless, of course, I constantly download (or watch streamed) movies, in which case the peak bandwidth means nothing, and average bandwidth over a long time means everything?
It's easy to make an over-expensive setup (that the users will pay for), then charge them for it through the nose (including mandatory service package) yet make sure that their actual use will be the same as with half-decent DSL (so you just buy piddly T-1 for each 80-100 users, with actual cost per user at most $15 and call it broadband).
While the maximum throughput can easily be that fast, the total bandwidth they are getting through those lines can't be more than usual 10-30Kbps/user in most of shared systems. They pay $135/mo for that plus digital cable TV + phone, but phone and cable TV are dirt cheap, so they pay $60-80/mo for the network connection -- comparable with high-end DSL, but this is a shared environment, it's supposed to be cheaper just because they buy the bandwidth for everyone at once. And what are the limitations -- can they run servers, do they have mandatory proxies on that?
Also $100/mo just to "maintain" security and web-controlled sprinklers is insane -- those things are just devices, they run themselves, why the monthly fee?
I doubt that good HOA (if it's HOA maintaining that and not just some company that is getting a hefty profit from that) will jack up the fees that much.
Now we will need to have full Unicode implementation on a handheld just to declare compliance with this protocol.
While the base idea is sane, I don't really see much of a point of doing that -- user interface for single appliance can be just kept on the appliance with some simple definition, but people would most likely use multi-device or scriptable user interfaces that combine them by controls' names.
Including a scripting language into protocol seems to be pointless -- scripting should be done in some central place that definitely knows all scripts that are running because information about active scripts is just as important as information about devices' state.
IMNSHO devices' interfaces must be very, very simple, or devices will get way too expensive, and user interface's definition in the protocol should be just a little more complex, or it will be a portability hell. Everything beyond those things should not be within this system, it just should be aware that there are scripts written in something else, that can see states and process requests.
Slashdot Mirror
...will start writing horrible monsters running hundreds and thousands of threads, and their creations will suffer from all other shortcomings of that decision.
I mean, it's got to hurt being successfully invaded by the French.
Probably about as much as it was in England.
R4-2 A public-private partnership should perfect and accelerate the adoption of more secure router technology and management, including out-of-band management. R4-3 Internet service providers, beginning with Tier 1 companies or R4-10 The private sector should consider including in near-term research and development priorities, programs for highly secure and trustworthy operating systems. If such systems are developed and successfully evaluated, the Federal government should accelerate procurement of such systems. in software code development, including processes and procedures that diminish the possibilities of erroneous code, malicious code, or trap doors that could be introduced during development. R4-17 The PCIPB s Awareness Committee, in cooperation with lead agencies,
They do realize that "trustworthy computing" name was originared by Microsoft, and has absolutely nothing to do with computer user's security and everything with software companies' "security" from the user, whoever he might be? Don't they?
Never try to check if you are being attacked by anything other than trivial DoS -- the attack that will succeed is unlikely to be seen.
As for being taken over, just read the bug description. Or, better, patch the system before the exploit comes out.
He rotates his head, sings a single tune, his head is empty, and he is trying to avoid scrutiny...
And a plural of "troll" is "slashdot", but "trolls" is still acceptable.
IRC doesn't support a lot of the functionality of AIM, it's much less convenient for most of what I use AIM for
What functionality is in AIM and not in IRC? By IRC, of course, I mean "what can be reasonably done with it using existing clients and minimal scripting", not "what comes by default with the only IRC client I have seen 5 years ago".
and maybe I don't always want everyone to know my IP address?
Server doesn't have to disclose it, and you can proxy the connection if you are paranoid. Not that it matters for any reasonable purpose that financial institutions may have.
Financial companies should just give up "some fatass company should sell us things" and look how IRC is superior to all those closed and semi-closed "messengers".
There is no compatibility problems between chat systems just like there is no compatibility problems with email. It's just closed email systems already disappeared, and closed messaging system are still there -- but people who rely on them deserve to suffer from their closedness.
But I think you're being overly pessimistic otherwise. Command of the air in Kosovo may not have led directly to the toppling of Milosevic, but let's remember that it was used from 16,000+ feet (out of the range of Serbian surface-to-air missiles), largely out of the understandable American desire to avoid unnecessary casualties.
That was a typical use of military force in the time of peace, when civilians' life is valuable -- it works until a real war starts. And, of course, it only supported Milosevich by providing a convenient foreign enemy, ultimately delaying his removal from power by then disorganized opposition, that couldn't use situation either way at the moment. Same with some later events and another weak leader that used foreign "air attacks" against civilians in his country as a crutch for his political career. He is doing fine so far.
I think a lot of the pessimism over the value of air superiority otherwise is a hangover from Vietnam, where major air offensives are generally believed to have been ineffective. What people tend to forget is that the Vietcong had an incredibly effective sigint network, and - from evesdropping on unsecured American communications, and particularly the ground- communications of air Force maintenance personel - often had quite advance warning on the location and timing of air assaults.
When the US got its act together and began to secure its communications, the effectiveness of bombing in Vietnam (and Cambodia) increased immesurably.
Vietnam was a special case in a lot of areas, so I wouldn't read much into that.
Early detection of airplanes flying from remote locations however doesn't depend on interception of communications now -- radars and even satellites do it much better. What is more important, when you are not completely out of resources, it simply doesn't matter much -- just defend what is worth being defended, and accept minor losses caused by whatever will slip through those defenses, or in the areas that are defended poorly. When you are at war you can't guard every building and civilian, but if losses from bombing are higher than, say, potential losses from enemy saboteurs (that are nearly impossible to prevent completely but never can be too high either), enemy will prefer bombing, and you will prefer tying up and wasting enemy's resources in actual battles rather than wasting yours ones on defending large areas that a bomber can reach.
Now I know that if someone will try to bring a hacksaw in San Francisco and cut the truss on the Golden Gate Bridge, he will be stopped at the airport when he will be flying to SFO!
In any event -- its likely that air superiority will continue to be the decisive factor in contemporary military conflict -- and China doesn't have remarkably good aerospace airforce and knows it.
"Air superioruty"'s overblown value is a doctrine originated by Germans (in their pathetic attempts to attack Britain from the air alone) and picked up by US as a way to sell themselves an idea of having a war with extremely low casualties on their side. It's bullshit. Any airplane outside of the area of a battle on the ground or sea is nothing but a very large house fly -- an annoying target. The damage that it can cause to anything military is negligible, the possibilities to turn it away by all kinds of attacks, even without actually destroying, are endless, and the best it can do is to bomb cities full of civilians to at best demoralize and usually merely annoy the population.
In a battle on the ground or sea the air force can do a lot of assistance and may even determine the outcome of the battle, but don't kid yourselves -- there must be a battle there in the first place, and a lot of people will have to die there, with or without airplanes flying over them.
War sucks, and if you go there, you are most likely to die. No doctrine changes that.
Right. Now show me even a single genuinely happy person in this country.
Then try raising a family. You don't have the luxury of morals when you have children crying themselves to sleep because they are hungry. I'd stick a knife in the next person if it meant the difference between feeding my family and upholding my morals, and I don't apologize for it.
Solution: kill yourself and your spouse if you have any. Then your kids will be orphans, and whatever food they will get in an orphanage will be far better than whatever you will get them by killing a random person.
And if your situation is not as desperate to do that it's quite natural to hold you to the same moral standards as the rest of humankind.
C _is_ the state of the art of procedural languages for over 30 years. It's not like such a simple thing has a lot of room for improvement, there are other areas where new languages can be created, but honestly how many ways are there to do things like preprocessor, functions, variables, etc.?
OO languages' authors may feel that they are doing something more "advanced" but in fact they are working on a completely ortogonal area of development. And most of them are far from C elegance (ex: Stroustrup doesn't even understand C design properly, so C++ is even more inconsistent than what its origin would suggest, and I don't even consider a rotting pile of shit that its "standard" libraty is, to be a part of language), or are simply badly designed (ex: Java), or are not languages but eclectic messes made by including specific librariers' and object models design into the language itself (ex: C#).
This is the area where we can use a lot of progress until it will reach the state where we can keep call the same thing "state of the art" for 30 years, but I won't hold my breath -- OO language design is dead, everyone is just making "OO" languages as various vehicles to promote their narrow-minded ideas. So I won't be surprised if Stroustrup's mess will remain the most useful semi-OO language for the next 30 years, too (but those libraries HAVE TO GO, and so should the attitude that students should learn that atrocity without studying C first).
Can I have neo-nazi propaganda quoted at length by Reuters or AP? What about an article "Church of Scientology said:", quote of their self-promoting speech, then "some people disagree".
The problem is not that journalists don't offer _their_ opinions, it's that they choose material that already is a complete bullshit (come on, Microsoft and Hollywood are trying to justify a creation of monopolistic cartel, what _is_ legitimate here?), and just relay it without even looking for something that will give a reader an idea if it's actually something valid. Only tabloids operate like this, but the difference is that tabloids intentionally look for bullshit to publish and have a reputation for doing so.
The reporting about the number of people attending the Klan meeting is reporting about fact. You can't change the facts, they just exist. However the article we are talking about doesn't deal with any facts that happened, it reports about arguments and opinions. And it heavily promotes the opinions of industry leaders, making everything looks like a nothing out of the ordinary bargain -- Hollywood pays with content, gets broadband, tech industry pays with DRM, gets Hollywood making content available in new and exciting proprietary format, legislators are doing fine job mediating the process, everyone is happy but tech industry gets a bit shorter end of the stick.
This is, of course, a lie, and real issue is that "Hollywood" is trying to ptessure technology industry into oppressing the users and basically demands to sacrifice all development in computer technology outside of the walls of large companies that can form a DRM cartel. But if that was said, article would be "biased" because it will display movie industry as complete assholes. So nothing is said about that. Everything is fine. Don't forget to buy new and exciting products, with time bombs inside.
You seem to suffer from the mistaken notion that 'fair and balanced' must yield an appearance of equal legitimacy, or must include equal promotion of both sides of an issue.
In modern journalism "fair and balanced" means exactly that -- if one side appears to be less legitimate, journalist must write an article heavily skewed toward it, so it will be "equally promoted". With KKK this is a taboo because of "PC" problem, attempt to whitewash them will cause immediate outrage, but with industry crooks it isn't -- general population knows nothing about all this, it sees things through the eyes of the imaginary "balanced" person that journalist creates for his story. And that "person" happens to quote industry heavyweights much more than, say, consumer advocates -- for obvious reason that without this "fairness" industries' "leaders" come out as evil, self-serving companies using consumers as hostages and shutting out the competition.
The problem is, trying to be write something "balanced" when talking about something that evil ends up being just a mouthpiece for them. There is nothing at all that can justify "Palladium" (or KKK if that matters) without saying a lie.
It's like "fair abd balanced story" about, say, KKK. Some things just should be never encouraged.
They either are against us, or are ready to sell our rights as a piece of some bargain. Consumer advocates, free/open software developers and civil rights groups must do everything to affect the outcome of this because it's their interests that are at stake. Both Intel and Disney can kiss our asses and have Enron-style bankruptcy for all I care.
Agent Smith may be an extreme misanthrope, but in this case, his expression applies pretty well. There are things where people don't WANT to deal with people on the other end of the line -- those things are detrtministic, and unless something is horribly broken, a simple machine easily gives all the necessary information, performs simple transactions, etc.
What is my account's balance? Is some payment past due? When will the package arrive? Is there an outage on my service? How much does this thing cost? Will you reimburse me for this? Do you block ports? Where are you located? Transfer my money to another account. Send this to my address. Process this application. Get a credit card payment.
You don't need a human to answer those things. Actually you DO need a human if you want him to answer or perform them wrong, to pretend that numbers are not what they are, that there is no problems when there definitely are some, to claim that they don't do something, to blame someone else, to pretend to be annoyed, to throw a lot of irrelevant offers, to have problems with handset, or (my favorite) not understanding someone's accent. Human ingenuity is a great help when simple answer exists, but should be hidden at all costs.
For all those simple things one doesn't need humans, he needs a web site. Or, for people that don't use computers, or are away from them, a simple phone menu. And it will serve its purpose much better than a bunch of minimum-wage drones with headsets ever would.
Humans are needed for other things, to answer questions that are not asked 65537 times a day, to explain meanings of obscure things that someone's customer may or may not know but it's hard to list all of them on a web site or especially in a phone menu. Humans can make decisions, ask questions based on things that are hard to place in a menu, but those things are far beyond what "AI" can do anyway. It would be foolish to try to replace humans there -- in fact anything that needs "AI" actually needs a human because at this level of technology development AI simply doesn't exist as anything usable. In those cases a proposal to use a machine to do a human's job would be so insane, only Harvard MBA would ever think of such a thing.
But for the rest of things, I would really prefer a machine.
Have you ever tripped over a sprinker head? It happens. When it does, who replaces it?
Whoever I'll call from a company that does sprinklers repair. I don't think that $100/mo is a package that both "web-enables" sprinklers and gives an owner free repairs while "not web-enabled" owners have to pay for repairs.
A security alarm is great, but wouldn't it be better if a security officer actually came out to your house and tried to catch the burglar if you weren't home?
Remote monitoring of security systems is a separate service, and it doesn't cost $100/mo either.
If so, how will I see the difference between 256Kbps and anything above? Unless, of course, I constantly download (or watch streamed) movies, in which case the peak bandwidth means nothing, and average bandwidth over a long time means everything?
It's easy to make an over-expensive setup (that the users will pay for), then charge them for it through the nose (including mandatory service package) yet make sure that their actual use will be the same as with half-decent DSL (so you just buy piddly T-1 for each 80-100 users, with actual cost per user at most $15 and call it broadband).
For every user?
While the maximum throughput can easily be that fast, the total bandwidth they are getting through those lines can't be more than usual 10-30Kbps/user in most of shared systems. They pay $135/mo for that plus digital cable TV + phone, but phone and cable TV are dirt cheap, so they pay $60-80/mo for the network connection -- comparable with high-end DSL, but this is a shared environment, it's supposed to be cheaper just because they buy the bandwidth for everyone at once. And what are the limitations -- can they run servers, do they have mandatory proxies on that?
Also $100/mo just to "maintain" security and web-controlled sprinklers is insane -- those things are just devices, they run themselves, why the monthly fee?
I doubt that good HOA (if it's HOA maintaining that and not just some company that is getting a hefty profit from that) will jack up the fees that much.
Now we will need to have full Unicode implementation on a handheld just to declare compliance with this protocol.
While the base idea is sane, I don't really see much of a point of doing that -- user interface for single appliance can be just kept on the appliance with some simple definition, but people would most likely use multi-device or scriptable user interfaces that combine them by controls' names.
Including a scripting language into protocol seems to be pointless -- scripting should be done in some central place that definitely knows all scripts that are running because information about active scripts is just as important as information about devices' state.
IMNSHO devices' interfaces must be very, very simple, or devices will get way too expensive, and user interface's definition in the protocol should be just a little more complex, or it will be a portability hell. Everything beyond those things should not be within this system, it just should be aware that there are scripts written in something else, that can see states and process requests.