May I ask you how FORTRAN is a) integrated with it's OS (whatever that really means in the case of FORTRAN), or b) integrating the development environment with the software under development?
Actually, Fortran's version of the problem is too-close integration with hardware features of the 70x/70xx instruction set. The three-way branch is a prime example. The restrictions on arithmetic in subscripts and iteration variables (corresponding to the index-register operations) through at least Fortran II is another. Fortran managed to abstract this away and carry on after the life of the platform. But it did this largely on the legacy of its codebase, accumulated since the time it was the first (and thus the only) compiler-level language. Fortran started to show its age during the "structured programming" flap of the late '60s and early '70s, (though standards orginizations were still kicking it around into the '90s.)
Interestingly, Lisp's CAR and CDR are also a legacy of that instruction set. There were about a dozen index register instructions that contained two address-sized fields, along with convenient instructions for modifying just those fields while operating on the instruction as a data element. Lisp used these "address" and "decrement" fields (the A and D of CAR and CDR) and their manipulation instructions as a convenient way to build compact data structures. But the two-pointer abstraction was sufficiently removed from its implementation that it wasn't a barrier to portability.
That same instruction set dependence was what killed MAD. The Fortran calling convention was to use a TSX (Transfer and Set Index) instruction to jump and save the return address in an index register, followed by a NOP for each argument with the argument's address in the NOP's address field. Return was to the first NOP. MAD substituted one of the index-register operations for the NOP (several of them became two-address NOPs if no index register was selected). The argument's address was in the address field as usual, but the decrement field was often used also - pointing to the "dope vector" describing the geometry of matricies, the other end of a "block argument" (a through b, expressed as "a...b") and so on. So MAD could take advantage of the copious Fortran libraries as well as its own native code, while Fortran could call MAD subroutines that didn't use the extensions in the argument-passing convention.
But when IBM end-of-lifed the 70x and replaced it with the 360, the new Fortran calling convention didn't have a convenient slot for a hidden second address. And the second address was necessary for several of MAD's key features.
Meanwhile IBM's TSS time-sharing system project had hit a snag, and the University of Michigan was committed to supporting its own MTS - a grad-student's hack that had grown into the Computing Center's core infrastructure while they were waiting. The Comp Center's budget wasn't up to supporting MTS AND porting MAD AND porting and supporting a native equivalent of the whole Fortran subroutine legacy - while still supporting Fortran so the engineering students could find work. So MAD was allowed to die.
The Operating System is called BRiX, and it uses a language called Crush, which is woven tightly into the core of the OS.
And thus is the same class of mistake as were made in lisp, mad, smalltalk, fortran, forth, and a number of others made once more.
Integrating the language and the OS kills portability, robustness, and security. Integrating the development enviornment with the software under development risks breaking the environment as you develop your target application and sucking the whole environment, bugs and all, into the target.
The languages I named had one or both of those problems. Sometimes it was useful, or "elegant". But always it was an albatross around the neck. I don't know if this new pair has the environment/target confusion. But the anonymous poster brags about combining the OS and language. So (if he's not just mischaracterizing an interpreter/P-code compiler) it certainly has that problem.
The key to successful programming is isolation. Single-mindedly chopping the problem into tiny pieces and walling them off from each other, then putting a few tiny holes in their individual prisons to let in and out ONLY the things they need to know and manipulate.
"Modularity". "Data Hiding". "Strong type-checking". "Interface Abstraction". The list of buzzwords is long. But the battle is constant. The number of interactions goes up with the FACTORIAL of the number of pieces interacting, while a programmer can only keep track of about six things at a time. The more connected the compiler, OS, and target program, the bigger the ball of hair a programmer has to unsnarl to get the program working. One of the things that was key to the success of the C language was the isolation of the runtime environment behind the subroutine interface.
Let us hope it's the characterization, and not the implementation, which has the problem.
It strikes me as being very plausible that the laws on the books could be interpreted as making something as simple as turning on a laptop running Windows with a wireless LAN card in the area of somebody else's wireless network a crime, particularly if it is argued that warchalkers are doing this with the specific purpose of determining whether or not it is possible to use a network that doesn't belong to them.
That is indeed one possible interpretation. And the FBI hints that they think it might be interpreted that way in their letter on the subject.
But there is another interpretation possible.
For decades - essentially since permission systems came into being - many computer users have treated file and account permissions not just as technological means of protection (like locking a door), but as expressions of intent (like latching a screen door to indicate that permission to enter is required).
Even before there were laws and court decisions on the subject there were often company policies. And in the absense of company policies there was courteousy and custom. Oversimplified: If a file was read-any it was OK for anybody to look at it, without prior permission and without notice afterward. If it was read-user-only, even a system administrator would normally ask before "breaking the lock" and examining the contents.
Now with the WEP encryption scheme totally cracked, its usefulness as a technological means of protecting data is over. (That will have to be done with firewall configurations.) But its usefulness as an EXPRESSION OF INTENT is unimpaired.
And many owners of wireless gateways - commercial or otherwise - may wish to allow them to be used as a convenience by users in the area. Some reasons a business might do this are:
- To allow visitors (business partners, salesmen, job applicants, etc.) to use their laptops. (Use an encrypted tunnel from the laptop to the home office for business, surf the web or check mail while waiting, etc.) This is in the same category as providing a phone in the waiting room, a drinking fountain, an outlet for laptop power, and not charging a fare to use the elevator.
- To provide internet access to passers by in the hope that others will do the same, thus making it available to THEIR employees in the field.
- To attract customers (i.e. coffee shops).
- To "be a good neighbor" or "make a statement" about internet freedom, by letting nearby residents and passers-by access the net through their link. (There are a number of companies who do this.)
So it's not unreasonable to assume that an open wireless LAN might be deliberate.
A reasonable interpretation of 802.11 and firewall configurations as expressions of intent might be:
- WEP enabled: Ask for permission.
- WEP disabled, DHCP enabled, packets routed to/from the internet: It's OK for anyone to use this port as a convenience. (The sysadmin has INSTRUCTED the system to ACTIVELY ASSIST anyone trying to connect - or has at least not turned it off if it came out-of-the-box that way.) Please don't abuse our hospitality by cracking our machines, soaking up enough bandwidth to impair the business functions (like streaming video during business hours) or getting the company in legal trouble (like launching DoS attacks, cracking.mil sites, or downloading MP3s).
Now if it's "WEP disabled, no firewall between the port and the LAN machines" the message is "clueless system administrator" - a professional behaving in an unprofessional manner. (The implied intent would be "our business is wide open for you to review" - and that's not a reasonable expectation.)
For one we can ask all the big major backbones (whom the RIAA is sueing) to block their website at the routing points:)
That will be especially interesting when/if the RIAA gets congressional authorization to DOS P2Pers.
"Those guys were flooding our network with packets and DoSing some of our customers' customers. That's against our acceptable use policies and it was chewing up our backbone bandwidth. So we had to cut them off. No, it's just a coincidence that they were already suing us..."
Actually what Al Gore did was promote legislation giving commercial users unrestricted access to the Internet. Up until then it had been a government, educational-insitiution, and suppliers-to-them sort of thing, and commercial speech was largely banned.
Al's legislation largely spiked attempts by the users and operators of the Internet to block commercial speech. At first this just gave people posting ads to email lists and news groups an excuse. But eventually the first spammers began harvesting addresses and sending automated email (apparently starting with an ad for such email software B-( at least that's the first spam I got) and the flood was on.
Sheit... in California, you don't even need to show IDENTIFICATION to vote (this probably has to do with that case someone else mentioned in a different thread).
Just so long as the name and address you provide match up with a registered voter, you are good to go.
Worse than that: You can register at the polls and vote immediately. (This proposal wouldn't keep them from doing that, by the way. They could just seal the votes of the newly-registered voters until the ID had been checked, then count such ballots later in the final tally.)
Of course thanks to the "motor-voter" law you can also register by mailing in postpaid bingo cards that you can pick up at most government offices (and at supermarkets, and...). You can do this over and over again, making up names as you go. You can vote absentee on all of 'em. (One address in Berkeley had over 4,000 voters registered. "We're operating a maildrop for the homeless" - yeah, right!)
A guy down the street from us is not a US citizen, but brags about how he has over twenty registrations - and votes 'em all.
The girl next door has been trying for years to get her deceased mother off the voting rolls. Clerk keeps putting her back on "because she's still voting".
Vans stuffed with people go from polling place to polling place, with the people voting at EACH of them.
I have changed my party affiliation several times and found myself double-registered as a result twice - because the same form is used for add, change affiliation, and change address and the clerk typoed. (I made DAMNED sure nobody voted my extra registration until I got it canceled.)
Turnout in one district dropped over 80% when a (false) rumor went around that the INS would be checking voters for ID and deporting non-citizens.
I could go on.
But there's SO much corruption in the elections in California that I really wonder how much of the vote is the actual population and how much is the political machine.
This has got to stop!
Because when the elections become so corrupt that they don't actually represent the will of the people, and the people REALIZE it, they stop performing their real function. And that function is to convince the losers that they can't reverse the decision by force. Corrupt elections destabilize governments and lead to civil war.
Now there are two things I don't like:
- Corrupt elections.
- Government ID cards.
By tying them together I hope to get ONE of them stopped.
Maybe it will be the ID cards - because some of the politicions currently in power might think some of their votes were faked up by the party machine back home and the ID cards might cost them their seat. Or maybe we'll lose on the ID card issue but at least get improved elections out of it.
But I'll be DAMNED if I sit around silently while the machine-politicians ram through an ID card and still leave the machine intact.
All elections are state matters. You are voting for your state's reperentitives. Picking of congressmen, senators, and presidential electors are to be conducted in a manor perscribed by the state legistlature. Why don't people understand this?
Because it isn't true. The Federal government has overriding authority in elections where federal officials or issues are voted on. (They used this authority to enforce the civil rights voting act, for example.)
The Constitution also mandates that the Fed insure that each state has a Republican form of government. (Meaning elections of representative officials, not the Republican Party. B-) ) This lets the Fed diddle in elections of state officials and the general form of the election - especially if the state(s) in question have a record of using procedures that deny representation to their residents or a subset of them.
The Constitution also gave the Fed a mandate (after a period - long since over - when the states could still make the decision) to decide who is a legal immigrant and thus eligible to vote.
This really points to the disturbing trend (Palladium anyone?) that says you have to connect to the internet to even use your computer.
A bomber the FBI was hunting recently discovered something similar about his cell phone.
He had driven halfway across the country from the area where he had been planting bombs in people's mailboxes. Somewhere in Nevada he powered up his cellphone. And when the cellphone identified itself to the network, the new "locate the 911 call" system (which actually tracks the phone any time it's on) reported his location to the cops (who had already notified the phone company to look out for him). They had him captured within half an hour.
Of course the first time the general population heard about this capability was when it was mentioned in a news story about the capture. (If the cops hadn't told the reporter it had been used, even those of us who knew it was possible wouldn't have known it was already deployed.)
This digital rights management registration has the same properties, but for any type of line:
Turn on your computer while it's attached to the internet and it "phones home" to check your licenses, which are identified to you personally.
This identifies the IP number you're currently using.
The IP number - even if it's dynamic - identifies the ISP, and the port within it.
The ISP can track the port to a physical connection - either hardwired or dialup - and can do this either in real-time or from logs after the connection is dropped.
The location can be identified immediately for hardwired connectinos. For dialups the phone company or companies handling the call can track it - again either real-time or from logs. (Both the ISP and the phone companies can tie this to your name, bank account, and so on.)
The entire process CAN be automated (if it has not been already), much like Carnivore, giving the FBI or others instant access to the information.
This may already have been authorized by the Patriot act. It's directed at enemy non-citizens and intended to be used by the "intelligence community" and so claims to escape many civil-rights safeguards (such as the need to get a warrant before using it), much like the incarceration without recourse to courts used against Johnny Walker Lindh and others associated with the Taliban.
Of course if this facility is used to capture an actual bomber and save lives, that's good. But if it's used to capture somebody some law-enforcement or spy agency THINKS is the bomber, it's not so good. And if it's used to harass opposition political figures, anybody some bureaucrat or cop doesn't like, or random citizens, it's called "a police state".
Please don't tell me "It can't happen here." Because it DID happen here. Repeatedly. (Look up COINTELPRO - or the general history of the FBI - for examples within the computer era.) And don't tell me it USED to happen but doesn't anymore, either. It takes decades for this stuff to come to light, so the recent stuff is still not general knowledge. (I remember people saying it doesn't happen anymore when COINTELPRO was happeneing.)
But the "digital rights management" hook is just the last straw, tying your personal identity to your computer's identity in advance. The bulk of this has already been deployed - at least in Microsoft systems and possibly in others.
Microsoft system installs attempt to configure your network connection. If they succeed, it's "PC Phone Home". They have your Software Product Key (a unique identifier for the software distribution), the serial number of your CPU if it exposes one, the MAC address of any ethernet cards (which can serve as a hardware unique identifier if your CPU doesn't expose a serial number), and any info you entered during the setup - like to sign up for network service. Of course the connection itself gives them your call trace information.
A few years ago Microsoft found a new use for spam: They sent out a series of "developer conference" adds. The remove-me email address would bounce. But the remove-me URL would load a mix of HTML, Javascirpt, and VBscript which would construct a URL containing your registry information and use it to query register.microsoft.com. (The registry contains your Software Product Key, ethernet card MAC address, etc.)
Some of the file formats used by Microsoft tools embed identifying information in files they store or exchange - which can also get it into email. An example is Microsoft Word, and the identifying information has already been use to arrest at least one macro-virus author.
Of course, damn near everyone I know at work also gets spam e-mail to buy toner supplies for our d getting them two months after being hired. (Yeah, like the most recent hire automatically has purchasing control. *plonk*)
There was a phone scam something like that maybe 20 years ago - back when the long-distance rate wars were just starting.
Company on the west coast named itself "The Supply Room". In the morning (when night rates were still in effect on the west coast) they'd war-dial the office phones of businesses in the Eastern time zone. If they got a human (no doubt groggy in the morning) they'd say something like:
"This is 'The Supply Room'. Do you need any supplies? Xerox paper? Toner? Pens?"
If the poor sap at his desk said yes they'd ship some stuff to his office, and a large bill to the company's accounts-payable. (I hear they recorded the call as proof that the stuff had been "ordered", too.)
(I got one of those calls when consulting at an auto company's process automation department. Told 'em they were talking to the wrong person. Found out what they were when the company sent warnings around a few days later.)
And you can always take the stone out and have it put in a new or different setting.
---
Watch out if your jewler wants to cut a diffraction pattern in a used ring to make it new - for free. He gets to keep the gold removed by the cutting process - more than enough to pay for his efforts - and fine surface patterns in gold wear smooth quickly so the effect lasts only a year or less.
Asking a few thousand of his closest friends for suggestions, before going to her for that discussion, is a very useful thing to do.
But note that a gentleman will NOT discuss the actual relationship itself - either during or after - in a way that will reveal a lady's secrets or reflect badly on her.
Seriously. If you can't talk to each other about this, then you are fools to even consider marriage.
Asking a few thousand of his closest friends for suggestions, before going to her for that discussion, is a very useful thing to do. Perhaps one of them will come up with something brilliant.
But what makes you think he DIDN'T already talk to her about it? Or that she doesn't read slashdot? Or that SHE doesn't talk about things with HER friends before talking to HIM?
Pair bonding is a very important part of life. And hard to get right. (Fortunately you only have to get it right once. B-) ) Most people try to make every bit of the interaction as pleasant as they can for their partner. Getting information and opinion before trying something in the real relationship is one part of doing that.
To use your reasoning, I should expect modulated IF to come out of the phono sockets on the rear of my VHF "tuner", unless it is referred to in the brocure as a "receiver", which should then have loudspeakers, otherwise it's not doing it's job...?!?!?!?!
Right. Sorry. Yes, in at least the case of hi-fi audio "tuner" and "receiver" have been used respectively for devices that deliver receive RF and deliver low-level audio signals from those that also include amplifiers suitable for driving loudspeakers.
However, at least back when I was repairing them for college money, television systems separated into separate receivers and monitors weren't common outside a studio (or even inside one). "Tuner" applied to a television was a term-of-art for the heavily-shielded subassemblies that took antenna signals, power feeds, and possibly remote-control signals, and delivered IF over a coax to the IF amplifier.
In this context a "digital TV tuner" would be one that selected the station digitally, not one that was involved in receiving a digitally-modulated signal.
So when I saw the article I wondered why the HELL the Fed would want to mandate that all future televisions specify the channels digitally, rather than have continuous tuning adjustments. It wasn't until I had read a few responses that I realized what they were talking about.
I'll be in a plane over the Atlantic during the height of the showers... does that mean I'll get a kick-ass view?
Depends on the direction you're flying.
- If perseus is behind you you'll get a kick-ass view.
- If it's in front of you you'll get an in-your-face view.
- If it's beside you you'll get an in-your-ear view.
B-)
But figuratively:
You'll get a good view if it's behind or in front - because you'll be looking to the sides of the aircraft (unless you're in the cockpit). Looking directly at the radiant or directly away from it is not too impressive. Looking 90 degrees from it gives you lots of pretty streaks to watch.
By the way - it ought to be GREAT this year. I was in Nevada over the weekend - about 5000 feet above sea level in a dark area - and there were already quite a few bright ones showing.
What possible motivation would any future society have to thaw these people out?
I can think of a few who might be interested.
- History departments. (Benjamin Franklin wanted to be pickled in a wine barrel after death and revivied in a century or three to see how things had come out. Wouldn't you like to interview HIM? Or see Jefferson's reaction to what the Democratic Party has become? B-) There's been a lot of history since then and eyewitnesses can help sort it out.)
- Techie version of above: Anyone trying to fix a bug in a frozen programmer's code. B-)
- Political splinter groups of many sorts.
- Charities. (If you will donate to save a random starving child in Africa, would you donate to revive someone you knew or had heard of from your own history?)
- The entertainment industry. (LOTS of possibilities there...)
- Hobbiests. (Imagine the science-fiction convention you could have with every currently-dead author and fan in attendence... B-) Now do the same with civil-war recreationists, yachtsmen, skiers, archers. Want Karate lessons from an old master?)
- Previous revivees. (History department revives historical figure, who revives his wife and children, who revive their fellow cryonics club members...)
- Anybody with a bit of money and a bee in his bonnet. Do you have any idea how RICH (by current standards) the poorest of the poor would be when tech is up to reviving people frozen by current techniques? Try this: Think of the standard of living of a current welfare recipient - food - including imported fruit virtually year-round, medical care, recorded music, cable TV, electricity, etc. Now imagine how rich someone in 1812 would have to be to afford the equivalent. (Remember: No penicillin, no refrigeration, entertainment is live and rare for anyone less than a king,...)
and of course:
- CURRENT cryonicists, who will revive PAST cryonicists in the hope that FUTURE cryonicists will revive THEM. (Just because they can repair somebody who died of cancer in the naughties doesn't mean that they'll be able to keep people from dying from Arcturian Whooping Sneeze in the '80s. So there will likely still be cryonicists.)
Why would we need more people, especially those who can't accept their own mortality?
"... can't accept their own motality."? Sounds like you're believing pro-death propaganda.
We know damned well we're mortal. But that's no reason not to "Rage at the dying of the light" - and then see about repairing or replacing the lightbulb - as many times as possible.
(Not really a response to the previous article, other than to the fact that it's continuinuing to copy the mislabeling that yahoo (or somebody further upstream) started.)
Darn it, they're talking about DIGITAL RECEIVERS.
A DIGITAL TUNER is a device that selects an analog signal using a digital specification of the desired frequency. Doesn't matter if the signal itself is analog, digital, or whatever. A TUNER doesn't even DEMODULATE it. It just shifts its frequency to that of the IF amplifier (perhaps also amplifying it a bit and starting the process of filtering out nearby signals by attenuating those that are more than a few megahertz away from the desired signal.)
A DIGITAL RECIEVER takes a DIGITAL SIGNAL and extracts the modulation.
If this is "News for Nerds" let's get the terminology right.
I'm hesitant to mod down a post that I might disagree with even though I still might find it interesting. I.E. - INTERESTING+1, DISAGREE+1.
You're not SUPPOSED to mod down a post because you disagree with it. Whether it's wrong is for every reader to decide for themselves.
You're supposed to moderate on whether other people might want to see it, to direct their attention to interesting/insightful/funny stuff and away from flamebate/troll/redundant/etc.
(It might be interesting to allow a separate right/wrong moderation system. Perhaps with slightly different rules: No mod points to consume, probablility of being allowed to moderate moderately high for newbies, 100% for those with a significant preponderance of agree on right/wrong metamoderation.)
(I posted some of this in another reply. But it is also applicable here and I'm going a slightly different direction with it.)
But, if a large amount of the mirrors defected, quit accepting updates, etc, then they could gain control. Of course this would create a situation where you would randomly get two different results for the same lookup. If enough defected though, it may not matter.
Assigning domain names is the same issue as deciding who is the owner of a particular plot of land - precicely the issue that led to the creation of governments. And because domain names are a resource composed of unique items of considerable value you have the same potentials for conflict as with land.
If you have:
two (or more) authorities issuing two diverging interpretations of who owns which parts of a unique divisible resource
the general population divided into factions, with only the opportunity to switch factions or start their own and recruit
the leader of each faction (root server operator) chosing which authority to obey
you have a political situation. If the faction leaders for most of the population stick with the old leader you have a king or "benevolent dictator". If they desert to a new one you have a peaceful transition to a new king. If they split you have a war.
In this case it would be a "virtual world war", because the entire virtual world would, of necessity, be participating. And because it would be a fight over virtual things of real value, the potential for "collateral damage" is very real.
Earlier on you might have gotten away with a root-server defection. But now that the government has blessed ICANN, and large companies have registered valuable domain names with ICANN-blessed registries, or bought them from others for large sums, it's a new ballgame.
There's no point to going to a new root database unless it's different from the old one. That means some of the domain names in the new one will be assigned to different people than in the old. So if a root server switches, or an ISP's mail server switches to a new root server, some of the people owning domain names in the old root's registry would bring suit - against the owners in the new registry for "domain squatting", and against the root servers and ISPs as well. (Ditto if they ALL swtich at once, only more so.) That's how the virtual war becomes mapped into real-world money and real-world cops-with-guns enforcing real-world government court rulings.
Now inviting such trouble is NOT a profitable business practice. So don't expect ISPs to switch to root servers that point to a new registry or stick with existing root servers that switch. And don't expect operators of existing root servers to switch.
The same applies for essentially any attempt at a change that involves grass-roots abandonment rather than government action, now that the government has blessed ICANN. ICANN, its associates,and their customers won't let go lightly. Be prepared to face them in court.
Allowing this to be a monopoly is a mistake. Making it a monopoly is the easy choice, but it's the wrong one.
A natural monopoly is something like building city streets. Not selecting a server. A natural monopoly has a small cost of providing the service, and a very large start-up cost, that keeps competitors out. If you need government regulations to keep out the competition, then you aren't a natural monopoly.
You are confusing economics and politics vs mathematics.
Domain name assignment is not a possible "natural monopoly" because of any economic or political issues.
It is a possible "natural monopoly" becuase (in the absense of a new solution to the distributed update problem - a fundamental computer science conundrum) the process of:
determining that a particular domain name is unassigned and
assigning it to a SINGLE applicant is indivisible.
This means:
it must occur in a single database
the operator of that database becomes an "authority" over whatever part of the namespace his database assigns and records.
the only known multiple-authority solution that is "decicive" is to divide the namespace into regions and have a separate single authority for each region.
This is the same issue as deciding who is the owner of a particular plot of land - precicely the issue that led to the creation of governments. And because domain names are a resource composed of unique items of considerable value you have the same potentials for conflict as with land.
If you have (as a poster has suggested elsewhere):
two (or more) authorities issuing two diverging interpretations of who owns which parts of a unique divisible resource
the general population divided into factions, with only the opportunity to switch factions or start their own and recruit
the leader of each faction (root server operator) chosing which authority to obey you have a political situation. If the faction leaders for most of the population stick with the old leader you have a king or "benevolent dictator". If they desert to a new one you have a peaceful transition to a new king. If they split you have a war.
In this case it would be a "virtual world war", because the entire virtual world would, of necessity, be participating. And because it would be a fight over virtual things of real value, the potential for "collateral damage" is very real.
A peaceful, decicive, solution will thus have single authorities over each chunk of domain name space. That's what I mean by "natural monopoly".
Also, even natural monopolies need to be given a hard look. Frequently even something that would naturally tend toward becoming a monopoly can be channelled away from that with only a slight change in the operating requirements.
By all means let's try to find a better solution. If we do, we can map it directly onto the allocation of other resources and settlement of disputes over them. And thus replace the current forms of government with something better.
But I won't hold my breath - or hold up reforming the current monopoly decision-making system - waiting for it to be invented.
But ICANN doesn't seem, to me, to come close to being a natural monopoly. They needed a special law from the government to keep out the competitors.
That's just the current government (last-resort monopoly on dispute-resolution - natural or otherwise) exercising its prerogative of setting policy about something it thinks it created. With its usual efficiency. B-) The fact that ICANN's monopoly is currently propped up by the government doesn't say jack about whether it's "natural" and the government just picked the winner or whether it's imposed. It just says that it's currently imposed.
The US government selected our current "assigned names and numbers czar" and imposed the monopoly situation. And right now it's about the only group of people in a position to change it.
Now a grass-roots migration to an alternative registry might work for domain names - until a few companies whose names got "squatted" in the alternative registry file suit. But that won't work for IP number ranges. The whole backbone and all the ISPs have to agree on those because if any are in dispute their packets get lost.... natural or not, they've been an abusive monopoly.
If our government had more than about 3 brain cells dedicated to this problem, we wouldn't even have a monopoly in the first place.
If there's one thing that's a "natural monopoly" it's insuring there are no collisions in a global name space. (It's probably more of one than being the court for people who can't agree on an arbitrator.)
If we're going to continue with the current domains I think we'll have to bite the bullet on this one, let a monopoly have it, and ride herd on them to keep them from being oppressive.
But IMHO "oppressive" includes charging an ongoing fee in the tens of US dollars annually for each name, rather than a (much smaller) one-time fee for assigning or transferring a unique name. (Imagine if you had to rent your personal name on the same basis.) It doesn't cost THAT much to maintain a database for assignments. The root servers can be maintained by ISPs as a (trivially-expensive) part of the service, if nobody (like MIL, universities, clubs, etc.) volunteer.
Now an alternative for domain names would be to establish a bunch of new TLDs, one for each competing registry, and let them compete. Country domains could go wherever the country in question wants. IP numbers are another can of worms - but at least with IPV6 you have so many you could hand off BIG blocks and never feel a pinch.
(By the way: I've NEVER understood why.us and/or the domain system in general was handed off to ICANN rather than the patent and trademark office, and protocol numbers to ANSI, NIST, ITU, etc. That's what they DO for a living, after all. A domain name, for instance, is one of the best examples of a service mark I've ever seen.)
Not true, and if it was, it would be a really bad idea to have them all in the same place. RFC 2010 [faqs.org] gives the standard requirements for the servers.
I think you're confusing two issues.
- There is one canonical root database. This is where the decisions about what is registered and what is not (at the root level, the TLDs, and the significant [.com,.org,.net] SLDs) are made. If it's lost it can be restarted from a backup or mirror. But changes made since the last backup or flush will be lost.
- There are a number of root servers. These are all effectively mirrors of the contents of the root database as of the last snapshot.
The issue is who maintains the canonical database, which provides the data for the servers, not the servers themselves.
Resistance is futile. You too will want a cellphone one day (your kids certainly do).
I've wanted one (or a radio phone) since at least 1967.
At least that's the earliest I remember looking at whether anything less solvent than a government agency could acquire enough miniature parts to hack a transciever into a Princess (tm) phone handset and use it to talk to an automated base station.
Slashdot Mirror
May I ask you how FORTRAN is a) integrated with it's OS (whatever that really means in the case of FORTRAN), or b) integrating the development environment with the software under development?
Actually, Fortran's version of the problem is too-close integration with hardware features of the 70x/70xx instruction set. The three-way branch is a prime example. The restrictions on arithmetic in subscripts and iteration variables (corresponding to the index-register operations) through at least Fortran II is another. Fortran managed to abstract this away and carry on after the life of the platform. But it did this largely on the legacy of its codebase, accumulated since the time it was the first (and thus the only) compiler-level language. Fortran started to show its age during the "structured programming" flap of the late '60s and early '70s, (though standards orginizations were still kicking it around into the '90s.)
Interestingly, Lisp's CAR and CDR are also a legacy of that instruction set. There were about a dozen index register instructions that contained two address-sized fields, along with convenient instructions for modifying just those fields while operating on the instruction as a data element. Lisp used these "address" and "decrement" fields (the A and D of CAR and CDR) and their manipulation instructions as a convenient way to build compact data structures. But the two-pointer abstraction was sufficiently removed from its implementation that it wasn't a barrier to portability.
That same instruction set dependence was what killed MAD. The Fortran calling convention was to use a TSX (Transfer and Set Index) instruction to jump and save the return address in an index register, followed by a NOP for each argument with the argument's address in the NOP's address field. Return was to the first NOP. MAD substituted one of the index-register operations for the NOP (several of them became two-address NOPs if no index register was selected). The argument's address was in the address field as usual, but the decrement field was often used also - pointing to the "dope vector" describing the geometry of matricies, the other end of a "block argument" (a through b, expressed as "a...b") and so on. So MAD could take advantage of the copious Fortran libraries as well as its own native code, while Fortran could call MAD subroutines that didn't use the extensions in the argument-passing convention.
But when IBM end-of-lifed the 70x and replaced it with the 360, the new Fortran calling convention didn't have a convenient slot for a hidden second address. And the second address was necessary for several of MAD's key features.
Meanwhile IBM's TSS time-sharing system project had hit a snag, and the University of Michigan was committed to supporting its own MTS - a grad-student's hack that had grown into the Computing Center's core infrastructure while they were waiting. The Comp Center's budget wasn't up to supporting MTS AND porting MAD AND porting and supporting a native equivalent of the whole Fortran subroutine legacy - while still supporting Fortran so the engineering students could find work. So MAD was allowed to die.
The Operating System is called BRiX, and it uses a language called Crush, which is woven tightly into the core of the OS.
And thus is the same class of mistake as were made in lisp, mad, smalltalk, fortran, forth, and a number of others made once more.
Integrating the language and the OS kills portability, robustness, and security. Integrating the development enviornment with the software under development risks breaking the environment as you develop your target application and sucking the whole environment, bugs and all, into the target.
The languages I named had one or both of those problems. Sometimes it was useful, or "elegant". But always it was an albatross around the neck. I don't know if this new pair has the environment/target confusion. But the anonymous poster brags about combining the OS and language. So (if he's not just mischaracterizing an interpreter/P-code compiler) it certainly has that problem.
The key to successful programming is isolation. Single-mindedly chopping the problem into tiny pieces and walling them off from each other, then putting a few tiny holes in their individual prisons to let in and out ONLY the things they need to know and manipulate.
"Modularity". "Data Hiding". "Strong type-checking". "Interface Abstraction". The list of buzzwords is long. But the battle is constant. The number of interactions goes up with the FACTORIAL of the number of pieces interacting, while a programmer can only keep track of about six things at a time. The more connected the compiler, OS, and target program, the bigger the ball of hair a programmer has to unsnarl to get the program working. One of the things that was key to the success of the C language was the isolation of the runtime environment behind the subroutine interface.
Let us hope it's the characterization, and not the implementation, which has the problem.
It strikes me as being very plausible that the laws on the books could be interpreted as making something as simple as turning on a laptop running Windows with a wireless LAN card in the area of somebody else's wireless network a crime, particularly if it is argued that warchalkers are doing this with the specific purpose of determining whether or not it is possible to use a network that doesn't belong to them.
.mil sites, or downloading MP3s).
That is indeed one possible interpretation. And the FBI hints that they think it might be interpreted that way in their letter on the subject.
But there is another interpretation possible.
For decades - essentially since permission systems came into being - many computer users have treated file and account permissions not just as technological means of protection (like locking a door), but as expressions of intent (like latching a screen door to indicate that permission to enter is required).
Even before there were laws and court decisions on the subject there were often company policies. And in the absense of company policies there was courteousy and custom. Oversimplified: If a file was read-any it was OK for anybody to look at it, without prior permission and without notice afterward. If it was read-user-only, even a system administrator would normally ask before "breaking the lock" and examining the contents.
Now with the WEP encryption scheme totally cracked, its usefulness as a technological means of protecting data is over. (That will have to be done with firewall configurations.) But its usefulness as an EXPRESSION OF INTENT is unimpaired.
And many owners of wireless gateways - commercial or otherwise - may wish to allow them to be used as a convenience by users in the area. Some reasons a business might do this are:
- To allow visitors (business partners, salesmen, job applicants, etc.) to use their laptops. (Use an encrypted tunnel from the laptop to the home office for business, surf the web or check mail while waiting, etc.) This is in the same category as providing a phone in the waiting room, a drinking fountain, an outlet for laptop power, and not charging a fare to use the elevator.
- To provide internet access to passers by in the hope that others will do the same, thus making it available to THEIR employees in the field.
- To attract customers (i.e. coffee shops).
- To "be a good neighbor" or "make a statement" about internet freedom, by letting nearby residents and passers-by access the net through their link. (There are a number of companies who do this.)
So it's not unreasonable to assume that an open wireless LAN might be deliberate.
A reasonable interpretation of 802.11 and firewall configurations as expressions of intent might be:
- WEP enabled: Ask for permission.
- WEP disabled, DHCP enabled, packets routed to/from the internet: It's OK for anyone to use this port as a convenience. (The sysadmin has INSTRUCTED the system to ACTIVELY ASSIST anyone trying to connect - or has at least not turned it off if it came out-of-the-box that way.) Please don't abuse our hospitality by cracking our machines, soaking up enough bandwidth to impair the business functions (like streaming video during business hours) or getting the company in legal trouble (like launching DoS attacks, cracking
Now if it's "WEP disabled, no firewall between the port and the LAN machines" the message is "clueless system administrator" - a professional behaving in an unprofessional manner. (The implied intent would be "our business is wide open for you to review" - and that's not a reasonable expectation.)
IBM got in trouble for chalking ads
I thought they got in trouble for PAINTING ads. Is that incorrect?
For one we can ask all the big major backbones (whom the RIAA is sueing) to block their website at the routing points :)
That will be especially interesting when/if the RIAA gets congressional authorization to DOS P2Pers.
"Those guys were flooding our network with packets and DoSing some of our customers' customers. That's against our acceptable use policies and it was chewing up our backbone bandwidth. So we had to cut them off. No, it's just a coincidence that they were already suing us..."
Since Al Gore invented the internet.
Actually what Al Gore did was promote legislation giving commercial users unrestricted access to the Internet. Up until then it had been a government, educational-insitiution, and suppliers-to-them sort of thing, and commercial speech was largely banned.
Al's legislation largely spiked attempts by the users and operators of the Internet to block commercial speech. At first this just gave people posting ads to email lists and news groups an excuse. But eventually the first spammers began harvesting addresses and sending automated email (apparently starting with an ad for such email software B-( at least that's the first spam I got) and the flood was on.
Sheit ... in California, you don't even need to show IDENTIFICATION to vote (this probably has to do with that case someone else mentioned in a different thread).
...). You can do this over and over again, making up names as you go. You can vote absentee on all of 'em. (One address in Berkeley had over 4,000 voters registered. "We're operating a maildrop for the homeless" - yeah, right!)
Just so long as the name and address you provide match up with a registered voter, you are good to go.
Worse than that: You can register at the polls and vote immediately. (This proposal wouldn't keep them from doing that, by the way. They could just seal the votes of the newly-registered voters until the ID had been checked, then count such ballots later in the final tally.)
Of course thanks to the "motor-voter" law you can also register by mailing in postpaid bingo cards that you can pick up at most government offices (and at supermarkets, and
A guy down the street from us is not a US citizen, but brags about how he has over twenty registrations - and votes 'em all.
The girl next door has been trying for years to get her deceased mother off the voting rolls. Clerk keeps putting her back on "because she's still voting".
Vans stuffed with people go from polling place to polling place, with the people voting at EACH of them.
I have changed my party affiliation several times and found myself double-registered as a result twice - because the same form is used for add, change affiliation, and change address and the clerk typoed. (I made DAMNED sure nobody voted my extra registration until I got it canceled.)
Turnout in one district dropped over 80% when a (false) rumor went around that the INS would be checking voters for ID and deporting non-citizens.
I could go on.
But there's SO much corruption in the elections in California that I really wonder how much of the vote is the actual population and how much is the political machine.
This has got to stop!
Because when the elections become so corrupt that they don't actually represent the will of the people, and the people REALIZE it, they stop performing their real function. And that function is to convince the losers that they can't reverse the decision by force. Corrupt elections destabilize governments and lead to civil war.
Now there are two things I don't like:
- Corrupt elections.
- Government ID cards.
By tying them together I hope to get ONE of them stopped.
Maybe it will be the ID cards - because some of the politicions currently in power might think some of their votes were faked up by the party machine back home and the ID cards might cost them their seat. Or maybe we'll lose on the ID card issue but at least get improved elections out of it.
But I'll be DAMNED if I sit around silently while the machine-politicians ram through an ID card and still leave the machine intact.
All elections are state matters. You are voting for your state's reperentitives. Picking of congressmen, senators, and presidential electors are to be conducted in a manor perscribed by the state legistlature. Why don't people understand this?
Because it isn't true. The Federal government has overriding authority in elections where federal officials or issues are voted on. (They used this authority to enforce the civil rights voting act, for example.)
The Constitution also mandates that the Fed insure that each state has a Republican form of government. (Meaning elections of representative officials, not the Republican Party. B-) ) This lets the Fed diddle in elections of state officials and the general form of the election - especially if the state(s) in question have a record of using procedures that deny representation to their residents or a subset of them.
The Constitution also gave the Fed a mandate (after a period - long since over - when the states could still make the decision) to decide who is a legal immigrant and thus eligible to vote.
If they ever do institute national ID cards there's one thing I want them used for:
Voting.
To insure that any person who votes in any national election is ELIGIBLE to vote in a national election and ONLY VOTES ONCE.
A bomber the FBI was hunting recently discovered something similar about his cell phone.
He had driven halfway across the country from the area where he had been planting bombs in people's mailboxes. Somewhere in Nevada he powered up his cellphone. And when the cellphone identified itself to the network, the new "locate the 911 call" system (which actually tracks the phone any time it's on) reported his location to the cops (who had already notified the phone company to look out for him). They had him captured within half an hour.
Of course the first time the general population heard about this capability was when it was mentioned in a news story about the capture. (If the cops hadn't told the reporter it had been used, even those of us who knew it was possible wouldn't have known it was already deployed.)
This digital rights management registration has the same properties, but for any type of line:
Turn on your computer while it's attached to the internet and it "phones home" to check your licenses, which are identified to you personally.
This identifies the IP number you're currently using.
The IP number - even if it's dynamic - identifies the ISP, and the port within it.
The ISP can track the port to a physical connection - either hardwired or dialup - and can do this either in real-time or from logs after the connection is dropped.
The location can be identified immediately for hardwired connectinos. For dialups the phone company or companies handling the call can track it - again either real-time or from logs. (Both the ISP and the phone companies can tie this to your name, bank account, and so on.)
The entire process CAN be automated (if it has not been already), much like Carnivore, giving the FBI or others instant access to the information.
This may already have been authorized by the Patriot act. It's directed at enemy non-citizens and intended to be used by the "intelligence community" and so claims to escape many civil-rights safeguards (such as the need to get a warrant before using it), much like the incarceration without recourse to courts used against Johnny Walker Lindh and others associated with the Taliban.
Of course if this facility is used to capture an actual bomber and save lives, that's good. But if it's used to capture somebody some law-enforcement or spy agency THINKS is the bomber, it's not so good. And if it's used to harass opposition political figures, anybody some bureaucrat or cop doesn't like, or random citizens, it's called "a police state".
Please don't tell me "It can't happen here." Because it DID happen here. Repeatedly. (Look up COINTELPRO - or the general history of the FBI - for examples within the computer era.) And don't tell me it USED to happen but doesn't anymore, either. It takes decades for this stuff to come to light, so the recent stuff is still not general knowledge. (I remember people saying it doesn't happen anymore when COINTELPRO was happeneing.)
But the "digital rights management" hook is just the last straw, tying your personal identity to your computer's identity in advance. The bulk of this has already been deployed - at least in Microsoft systems and possibly in others.
Microsoft system installs attempt to configure your network connection. If they succeed, it's "PC Phone Home". They have your Software Product Key (a unique identifier for the software distribution), the serial number of your CPU if it exposes one, the MAC address of any ethernet cards (which can serve as a hardware unique identifier if your CPU doesn't expose a serial number), and any info you entered during the setup - like to sign up for network service. Of course the connection itself gives them your call trace information.
A few years ago Microsoft found a new use for spam: They sent out a series of "developer conference" adds. The remove-me email address would bounce. But the remove-me URL would load a mix of HTML, Javascirpt, and VBscript which would construct a URL containing your registry information and use it to query register.microsoft.com. (The registry contains your Software Product Key, ethernet card MAC address, etc.)
Some of the file formats used by Microsoft tools embed identifying information in files they store or exchange - which can also get it into email. An example is Microsoft Word, and the identifying information has already been use to arrest at least one macro-virus author.
Of course, damn near everyone I know at work also gets spam e-mail to buy toner supplies for our d getting them two months after being hired. (Yeah, like the most recent hire automatically has purchasing control. *plonk*)
There was a phone scam something like that maybe 20 years ago - back when the long-distance rate wars were just starting.
Company on the west coast named itself "The Supply Room". In the morning (when night rates were still in effect on the west coast) they'd war-dial the office phones of businesses in the Eastern time zone. If they got a human (no doubt groggy in the morning) they'd say something like:
"This is 'The Supply Room'. Do you need any supplies? Xerox paper? Toner? Pens?"
If the poor sap at his desk said yes they'd ship some stuff to his office, and a large bill to the company's accounts-payable. (I hear they recorded the call as proof that the stuff had been "ordered", too.)
(I got one of those calls when consulting at an auto company's process automation department. Told 'em they were talking to the wrong person. Found out what they were when the company sent warnings around a few days later.)
Pawn shop.
Genuine antiques without the markup.
And you can always take the stone out and have it put in a new or different setting.
---
Watch out if your jewler wants to cut a diffraction pattern in a used ring to make it new - for free. He gets to keep the gold removed by the cutting process - more than enough to pay for his efforts - and fine surface patterns in gold wear smooth quickly so the effect lasts only a year or less.
Asking a few thousand of his closest friends for suggestions, before going to her for that discussion, is a very useful thing to do.
But note that a gentleman will NOT discuss the actual relationship itself - either during or after - in a way that will reveal a lady's secrets or reflect badly on her.
Seriously. If you can't talk to each other about this, then you are fools to even consider marriage.
Asking a few thousand of his closest friends for suggestions, before going to her for that discussion, is a very useful thing to do. Perhaps one of them will come up with something brilliant.
But what makes you think he DIDN'T already talk to her about it? Or that she doesn't read slashdot? Or that SHE doesn't talk about things with HER friends before talking to HIM?
Pair bonding is a very important part of life. And hard to get right. (Fortunately you only have to get it right once. B-) ) Most people try to make every bit of the interaction as pleasant as they can for their partner. Getting information and opinion before trying something in the real relationship is one part of doing that.
To use your reasoning, I should expect modulated IF to come out of the phono sockets on the rear of my VHF "tuner", unless it is referred to in the brocure as a "receiver", which should then have loudspeakers, otherwise it's not doing it's job...?!?!?!?!
Right. Sorry. Yes, in at least the case of hi-fi audio "tuner" and "receiver" have been used respectively for devices that deliver receive RF and deliver low-level audio signals from those that also include amplifiers suitable for driving loudspeakers.
However, at least back when I was repairing them for college money, television systems separated into separate receivers and monitors weren't common outside a studio (or even inside one). "Tuner" applied to a television was a term-of-art for the heavily-shielded subassemblies that took antenna signals, power feeds, and possibly remote-control signals, and delivered IF over a coax to the IF amplifier.
In this context a "digital TV tuner" would be one that selected the station digitally, not one that was involved in receiving a digitally-modulated signal.
So when I saw the article I wondered why the HELL the Fed would want to mandate that all future televisions specify the channels digitally, rather than have continuous tuning adjustments. It wasn't until I had read a few responses that I realized what they were talking about.
... Texas Republican Rep. Lamar Smith, said that FBI should not go for casual users but but instead to go after operators of "network "nodes" ...
Oh, dear.
Looks like I finally have to rename my machine.
Darn. Had that name since the entire list of domain names fit on three printed pages, too...
I'll be in a plane over the Atlantic during the height of the showers... does that mean I'll get a kick-ass view?
Depends on the direction you're flying.
- If perseus is behind you you'll get a kick-ass view.
- If it's in front of you you'll get an in-your-face view.
- If it's beside you you'll get an in-your-ear view.
B-)
But figuratively:
You'll get a good view if it's behind or in front - because you'll be looking to the sides of the aircraft (unless you're in the cockpit). Looking directly at the radiant or directly away from it is not too impressive. Looking 90 degrees from it gives you lots of pretty streaks to watch.
By the way - it ought to be GREAT this year. I was in Nevada over the weekend - about 5000 feet above sea level in a dark area - and there were already quite a few bright ones showing.
What possible motivation would any future society have to thaw these people out?
...)
I can think of a few who might be interested.
- History departments. (Benjamin Franklin wanted to be pickled in a wine barrel after death and revivied in a century or three to see how things had come out. Wouldn't you like to interview HIM? Or see Jefferson's reaction to what the Democratic Party has become? B-) There's been a lot of history since then and eyewitnesses can help sort it out.)
- Techie version of above: Anyone trying to fix a bug in a frozen programmer's code. B-)
- Political splinter groups of many sorts.
- Charities. (If you will donate to save a random starving child in Africa, would you donate to revive someone you knew or had heard of from your own history?)
- The entertainment industry. (LOTS of possibilities there...)
- Hobbiests. (Imagine the science-fiction convention you could have with every currently-dead author and fan in attendence... B-) Now do the same with civil-war recreationists, yachtsmen, skiers, archers. Want Karate lessons from an old master?)
- Previous revivees. (History department revives historical figure, who revives his wife and children, who revive their fellow cryonics club members...)
- Anybody with a bit of money and a bee in his bonnet. Do you have any idea how RICH (by current standards) the poorest of the poor would be when tech is up to reviving people frozen by current techniques? Try this: Think of the standard of living of a current welfare recipient - food - including imported fruit virtually year-round, medical care, recorded music, cable TV, electricity, etc. Now imagine how rich someone in 1812 would have to be to afford the equivalent. (Remember: No penicillin, no refrigeration, entertainment is live and rare for anyone less than a king,
and of course:
- CURRENT cryonicists, who will revive PAST cryonicists in the hope that FUTURE cryonicists will revive THEM. (Just because they can repair somebody who died of cancer in the naughties doesn't mean that they'll be able to keep people from dying from Arcturian Whooping Sneeze in the '80s. So there will likely still be cryonicists.)
Why would we need more people, especially those who can't accept their own mortality?
"... can't accept their own motality."? Sounds like you're believing pro-death propaganda.
We know damned well we're mortal. But that's no reason not to "Rage at the dying of the light" - and then see about repairing or replacing the lightbulb - as many times as possible.
Do you WANT to die? You can ALWAYS arrange it.
(Not really a response to the previous article, other than to the fact that it's continuinuing to copy the mislabeling that yahoo (or somebody further upstream) started.)
Darn it, they're talking about DIGITAL RECEIVERS.
A DIGITAL TUNER is a device that selects an analog signal using a digital specification of the desired frequency. Doesn't matter if the signal itself is analog, digital, or whatever. A TUNER doesn't even DEMODULATE it. It just shifts its frequency to that of the IF amplifier (perhaps also amplifying it a bit and starting the process of filtering out nearby signals by attenuating those that are more than a few megahertz away from the desired signal.)
A DIGITAL RECIEVER takes a DIGITAL SIGNAL and extracts the modulation.
If this is "News for Nerds" let's get the terminology right.
I'm hesitant to mod down a post that I might disagree with even though I still might find it interesting. I.E. - INTERESTING+1, DISAGREE+1.
You're not SUPPOSED to mod down a post because you disagree with it. Whether it's wrong is for every reader to decide for themselves.
You're supposed to moderate on whether other people might want to see it, to direct their attention to interesting/insightful/funny stuff and away from flamebate/troll/redundant/etc.
(It might be interesting to allow a separate right/wrong moderation system. Perhaps with slightly different rules: No mod points to consume, probablility of being allowed to moderate moderately high for newbies, 100% for those with a significant preponderance of agree on right/wrong metamoderation.)
But, if a large amount of the mirrors defected, quit accepting updates, etc, then they could gain control. Of course this would create a situation where you would randomly get two different results for the same lookup. If enough defected though, it may not matter.
Assigning domain names is the same issue as deciding who is the owner of a particular plot of land - precicely the issue that led to the creation of governments. And because domain names are a resource composed of unique items of considerable value you have the same potentials for conflict as with land.
If you have:
two (or more) authorities issuing two diverging interpretations of who owns which parts of a unique divisible resource
the general population divided into factions, with only the opportunity to switch factions or start their own and recruit
the leader of each faction (root server operator) chosing which authority to obey
you have a political situation. If the faction leaders for most of the population stick with the old leader you have a king or "benevolent dictator". If they desert to a new one you have a peaceful transition to a new king. If they split you have a war.
In this case it would be a "virtual world war", because the entire virtual world would, of necessity, be participating. And because it would be a fight over virtual things of real value, the potential for "collateral damage" is very real.
Earlier on you might have gotten away with a root-server defection. But now that the government has blessed ICANN, and large companies have registered valuable domain names with ICANN-blessed registries, or bought them from others for large sums, it's a new ballgame.
There's no point to going to a new root database unless it's different from the old one. That means some of the domain names in the new one will be assigned to different people than in the old. So if a root server switches, or an ISP's mail server switches to a new root server, some of the people owning domain names in the old root's registry would bring suit - against the owners in the new registry for "domain squatting", and against the root servers and ISPs as well. (Ditto if they ALL swtich at once, only more so.) That's how the virtual war becomes mapped into real-world money and real-world cops-with-guns enforcing real-world government court rulings.
Now inviting such trouble is NOT a profitable business practice. So don't expect ISPs to switch to root servers that point to a new registry or stick with existing root servers that switch. And don't expect operators of existing root servers to switch.
The same applies for essentially any attempt at a change that involves grass-roots abandonment rather than government action, now that the government has blessed ICANN. ICANN, its associates,and their customers won't let go lightly. Be prepared to face them in court.
A natural monopoly is something like building city streets. Not selecting a server. A natural monopoly has a small cost of providing the service, and a very large start-up cost, that keeps competitors out. If you need government regulations to keep out the competition, then you aren't a natural monopoly.
You are confusing economics and politics vs mathematics.
Domain name assignment is not a possible "natural monopoly" because of any economic or political issues.
It is a possible "natural monopoly" becuase (in the absense of a new solution to the distributed update problem - a fundamental computer science conundrum) the process of:
determining that a particular domain name is unassigned and
assigning it to a SINGLE applicant
is indivisible.
This means:
it must occur in a single database
the operator of that database becomes an "authority" over whatever part of the namespace his database assigns and records.
the only known multiple-authority solution that is "decicive" is to divide the namespace into regions and have a separate single authority for each region.
This is the same issue as deciding who is the owner of a particular plot of land - precicely the issue that led to the creation of governments. And because domain names are a resource composed of unique items of considerable value you have the same potentials for conflict as with land.
If you have (as a poster has suggested elsewhere):
two (or more) authorities issuing two diverging interpretations of who owns which parts of a unique divisible resource
the general population divided into factions, with only the opportunity to switch factions or start their own and recruit
the leader of each faction (root server operator) chosing which authority to obey
... natural or not, they've been an abusive monopoly.
you have a political situation. If the faction leaders for most of the population stick with the old leader you have a king or "benevolent dictator". If they desert to a new one you have a peaceful transition to a new king. If they split you have a war.
In this case it would be a "virtual world war", because the entire virtual world would, of necessity, be participating. And because it would be a fight over virtual things of real value, the potential for "collateral damage" is very real.
A peaceful, decicive, solution will thus have single authorities over each chunk of domain name space. That's what I mean by "natural monopoly".
Also, even natural monopolies need to be given a hard look. Frequently even something that would naturally tend toward becoming a monopoly can be channelled away from that with only a slight change in the operating requirements.
By all means let's try to find a better solution. If we do, we can map it directly onto the allocation of other resources and settlement of disputes over them. And thus replace the current forms of government with something better.
But I won't hold my breath - or hold up reforming the current monopoly decision-making system - waiting for it to be invented.
But ICANN doesn't seem, to me, to come close to being a natural monopoly. They needed a special law from the government to keep out the competitors.
That's just the current government (last-resort monopoly on dispute-resolution - natural or otherwise) exercising its prerogative of setting policy about something it thinks it created. With its usual efficiency. B-) The fact that ICANN's monopoly is currently propped up by the government doesn't say jack about whether it's "natural" and the government just picked the winner or whether it's imposed. It just says that it's currently imposed.
The US government selected our current "assigned names and numbers czar" and imposed the monopoly situation. And right now it's about the only group of people in a position to change it.
Now a grass-roots migration to an alternative registry might work for domain names - until a few companies whose names got "squatted" in the alternative registry file suit. But that won't work for IP number ranges. The whole backbone and all the ISPs have to agree on those because if any are in dispute their packets get lost.
Absolutely!
If our government had more than about 3 brain cells dedicated to this problem, we wouldn't even have a monopoly in the first place.
.us and/or the domain system in general was handed off to ICANN rather than the patent and trademark office, and protocol numbers to ANSI, NIST, ITU, etc. That's what they DO for a living, after all. A domain name, for instance, is one of the best examples of a service mark I've ever seen.)
If there's one thing that's a "natural monopoly" it's insuring there are no collisions in a global name space. (It's probably more of one than being the court for people who can't agree on an arbitrator.)
If we're going to continue with the current domains I think we'll have to bite the bullet on this one, let a monopoly have it, and ride herd on them to keep them from being oppressive.
But IMHO "oppressive" includes charging an ongoing fee in the tens of US dollars annually for each name, rather than a (much smaller) one-time fee for assigning or transferring a unique name. (Imagine if you had to rent your personal name on the same basis.) It doesn't cost THAT much to maintain a database for assignments. The root servers can be maintained by ISPs as a (trivially-expensive) part of the service, if nobody (like MIL, universities, clubs, etc.) volunteer.
Now an alternative for domain names would be to establish a bunch of new TLDs, one for each competing registry, and let them compete. Country domains could go wherever the country in question wants. IP numbers are another can of worms - but at least with IPV6 you have so many you could hand off BIG blocks and never feel a pinch.
(By the way: I've NEVER understood why
Not true, and if it was, it would be a really bad idea to have them all in the same place. RFC 2010 [faqs.org] gives the standard requirements for the servers.
.org, .net] SLDs) are made. If it's lost it can be restarted from a backup or mirror. But changes made since the last backup or flush will be lost.
I think you're confusing two issues.
- There is one canonical root database. This is where the decisions about what is registered and what is not (at the root level, the TLDs, and the significant [.com,
- There are a number of root servers. These are all effectively mirrors of the contents of the root database as of the last snapshot.
The issue is who maintains the canonical database, which provides the data for the servers, not the servers themselves.
Resistance is futile. You too will want a cellphone one day (your kids certainly do).
I've wanted one (or a radio phone) since at least 1967.
At least that's the earliest I remember looking at whether anything less solvent than a government agency could acquire enough miniature parts to hack a transciever into a Princess (tm) phone handset and use it to talk to an automated base station.