I've got analog cable, and no plans to upgrade. Does this mean that I need to use an analog card like the Hauppage, and that if I had digital cable I'd need a digital card instead?
If there's a standard around that does what you need to do, it's probably worth using it, at least if it's usable. There's a lot of application design these days that's too minor for a standards committee to bother with, and it's usually more important to get creative and interesting stuff out there than to talk people into thinking your work is going to be sufficiently creative and interesting that they should form a standards committee for stuff like yours.
However, you should still do so openly - build interfaces that people can use, and document them so people can figure out how to use them, and if you're lucky, people will use them for things you've never thought of, so try not to prevent that.
The important overall effect is that the performance or price-performance of computers doubles every N years, where N is about 1-2. The specific formulation in terms of transistor density on a 2-D grid is less important, especially because there are other ways to get performance increases - going to 3-D many-layer designs instead of flat chips, or getting faster communication paths to support larger chip areas, or using faster materials, or whatever it takes to get speed or capacity.
Increasing density is valuable, not only because it lets you put more transistors per chip, but also because the smaller transistors mean shorter travel distances for electons/holes, and therefore faster computation and higher clock rates. But if you could do the mythical true 3D design, instead of just a few layers, then you'd have *lots* of transistors a few microns or tens of microns away from each other instead of hundreds to thousands of microns away on the other side of a chip. Maybe you can't do that easily in silicon, or maybe a coarse-3D approach will help (e.g. 10-20 layers of chip stacked on top of each other, as opposed to hundreds of layers), or maybe carbon nanotubes or buckyballs or nano-unobtainium or whatever will be more flexible. Or maybe holographic memories could be useful. Who knows?
It's hypothetically possible that the Quantum Computing people might make some breakthrough that lets some kinds of problems be solved in small-polynomial time instead of exponential time, with some usable probability of a correct answer, so you'll have to start filling those liquid-cooling systems with liquid helium. (As a cryptographer, I'd find this very annoying, because most or all of the currently useful public-key technology would get trashed, but as a combinatoric mathematician, I'd find it to be really really cool:-) That's definitely not a Moore's Law approach to computers - it's major theoretical breakthroughs as opposed to continual rapid improvement due to technical investment and lots of minor theoretical breakthroughs.
It looks like BigPond's real concern is that their DNS architecture isn't well-designed or scalable enough to handle the load, so they're dealing with it by identifying heavy users instead of fixing it - but that works because most of the heavy users are zombies. So that's an interesting new development; most ISPs deal with the problem by either blocking or rate-limiting Port 25 or spam-filtering email or both.
The reason zombies are heavy DNS users is that they're trying to send a million emails a day, so they need to look up probably 10K-100K recipient's domains, depending on how much reuse they can get away with, and even if they're running on a PC with a caching DNS server, they're going to blow out the cache if they don't have them all sorted (and they probably *don't* have them all sorted, because they're trying to evade spam detection on the recipients' ISPs.) Also, the spamware probably doesn't have very intelligent DNS handling in it - if it did, it would probably go to some other DNS server or do something else to evade detection, though using the ISP's server does scale well if the ISP is competent.
An intermediate step they could take would be to put heavy DNS users on different DNS servers than the light users. Most PCs get their DNS server addresses from DHCP configuration, so they can do fancy things at DHCP lease renewal time like load-balance their server assignments, but they can also concentrate the heavy users. That might be a good way to put their email connections through more thorough spam-filtering than the average user gets as well.
First of all, most ISPs that "Block Port 25" don't block it for connections to their own mail server - only for connections that don't use their servers, either because they're going directly to the recipient or because they're going to some other mail server. If you're using the ISP's outgoing mail server, then they've got a handle for rate-limiting your mail (so they can detect or at least inhibit spammers, and possibly even spam-filter email), and they can provide whatever quality of email administration they want. For dialup users, this is often useful, because mailers benefit from being directly connected to the net in case the recipient can't handle their mail immediately (an especially frequent problem due to grey-listing.) But for broadband Linux users, it's often annoying, because the cable companies especially are often not very good at it. Some ISPs, mostly cable, used to be really obnoxious and not only block non-port-25 email but also require your From: address to be an address on their mail server. Fortunately, most of them have been beaten into submission by the market.
But there are several other protocols for sending email that don't look like Port 25 to the ISP. There are a couple of SMTP-submission protocols which let you set up a connection to a mail server where you have an account and do various kinds of authentication, including some that use SSL encryption. Alternatively, you can do SSH or IPSEC or other VPN tunnels to your email provider. And then for us old folks, there's always "login to a shell account":-) (Kids can use webmail instead.)
As far as email-over-telepathy goes, Dan Kaminsky recently demonstrated IP-over-DNS tunnelling at Codecon. It's really really evil - he was even able to do video-over-IP-over DNS by coopting about 25000 DNS servers. I'm pretty sure he was the guy who did a lot of the IP-over-HTTP tunnelling a couple of years back, and he;s done lots of other creative work with detailed protocol analysis.
Heh. Customers will be doing well to be permitted to use real systems at all - maybe Macintoshes have enough clout, but too many ISPs are bad at supporting Linux users already, doing things like PPPoE (useless and evil) or blocking Port 25 for everybody, and of course not being good at answering service questions when they don't have Windows desktops.
Also, "broken" printers *can* be a problem for desktop supporters. Sometimes the problem is the printer hardware, or the ink running out, or the coffee-cup holder on their machine breaking, but often the problem is something with Windows setup. I'm using a work-managed Windows 2000 laptop, and there's some sort of permissions problem that keeps me from using my USB printer at home (it can support it on a parallel port, but if I plug in non-storage USB devices it says I'm not allowed to do that.) My home PC supported the same printer ok until XP SP2 came along, and now it complains about drivers every time I reboot because I'm using the vendor's driver and not Microsoft's (it still works fine - *because* I'm using the vendor's driver and not Microsoft's:-)
Spyware isn't a problem for ISPs - it's just a problem for users. Sure, it creates a little traffic, not usually much, but the real problem is that it's an annoying invasion of privacy and may also slow down the user interface on the PC if it's badly implemented. Cleaning viruses is critical, because they affect other users. Cleaning spyware is just nice, not critical.
Also, it's a lot harder to tell what's really spyware - most of the spyware products I've used complain if they see cookies from Usual Suspects, but some of them have complained about the adware in Sponsored-Mode Eudora (yes, it's there, and it's the cost of using the free version of the software.) Some of it has even complained about things in found in my Windows Recycle Bin:-)
Joe Average User hits a lot of domains a day, from web surfing - if you count all the banner ads and similar trash, it adds up. But yeah, it's a lot slower than a spam zombie, or a large legitimate mailing list. And spammers are going to send a lot of spam to big ISPs, so there'll be a lot of concentration.
But a spammer *could* adapt to this by using DNS servers that aren't from the local ISP, or using spamware that downloads the victim's IP address along with their domain name. It's an arms race. Blocking Port 25 is more effective - Blocking it for everybody is a Bad Thing, but blocking it by default and letting users enable it themselves is fine. But until the spammers start working around DNS servers, they're a potentially useful hook for identifying problem customers.
My experience as a foreigner dealing with Telstra over the last decade and a half has been that they were always one of the most clueless telecom companies in the developed world. (There are worse companies like VSNL or most of the African PTTs, but among people who should know better, Telstra were always way down at the bottom of the list.) Maybe they were ok for voice phones; I haven't dealt with that. But for data, their cable modem people were the fools who came up with the idea of capping monthly downloads, and they've been trading those ideas with the US cable modem companies who are also terminally clueless. Even before cable modems, though, Telstra never did get the concept that somebody might want an E1 line because they wanted to pump ~2 Mbps in one fat channel, as opposed to 30 channels of 64 kbps (this was the early 90s, so 2 Mbps was still fairly fast, as opposed to today when it's pretty slow, but higher-end data users did want those speeds.) They've also tended to insist on traffic-usage-based billing for data circuits, even if they're just local connections across town in Sydney, rather than selling flat-rate access pipes (it's been a year or so since I've tried to get that; maybe they've seen the light since then.)
Real privatization usually has some market distortions, because the former monopoly is usually in a strong position, and it takes a while for competition to build up. But it's better than not doing it. On the other hand, bogus privatization, e.g. spinning off the company into a profit-making corporation but letting them keep monopoly power through regulatory mechanisms and owned by friends&relatives of the politicians in power is usualy worse. (Not always - even a quasi-monopoly can occasionally see that it makes more money if its customers like it and if it has higher-value services to offer.) A typical problem in places like Carribbean islands was that the monopoly PTT was Cable&Wireless with a few highly-paid positions for high politicians, and they'd insist on maintaining an expensive and antiquated wired telephone system because it "creates jobs" for lots of people, when it would be cheaper to replace the whole system with a couple of cell towers stuck on different sides of the mountain.
Yes, they get metamoderated, though they're not used a lot so they don't show up often. Nothing cowardly about using overrated / underrated - they're really more of a way of disagreeing with the other moderators. For instance, somebody's posting might get moderated "+5 Funny", and you might think that "ok, it's trying to be funny, but it's not *that* funny - it's maybe a +2 amused smile, not a +5 rotflmao" - so you mark it overrated. Or something's marked "+5 Insightful" and you think it's really worth "+3 does appear to get the point but isn't that exciting."
It's a bit hard to moderate "overrated" well, because you'd really want to know what the rating was when the moderator moderated it.
For the last couple of decades, Telstra has been one of the most aggressively clueless telecom companies in the developed world. They've been the big pushers of traffic limits for their users, and unfortunately, the *next* most clueless telecom companies in the world, US cable modem providers, have been listening to them. Even before cable modems showed up, they couldn't deal with the concept that somebody might want an E1 line to send a full 2 Mbps of data on it as opposed to 30 channels of 64kbps.
I use BitTorrent to download and upload lots of music on my DSL line - all of it legal concert tapes from etree etc. or free downloads. And I also use it for Linux and other software distributions.
Even if you are or are not maintaining the website for her, the domain name ought to be hers, and if you're being stubborn about the $6-$35 "ownership" you might have in it, you're being a jerk about the wrong things and ought to put the energy into something else. And if she feels like pointing the domain name somewhere else and writing a blog or a rant about her ex-boyfriend, well, it goes with the territory:-)
(As I was saying about bad typing, I seem to have hit the Enter key instead of a quote-mark before I was done editing:-)
So Jump Domain's "About Us" page calls itself a legal disclaimer, not that it's particularly disclaiming anything, though it does indicate an unwillingness to provide accurate contact information, like the name of the human who operates it.
Jumpdomain can't even use correct English on their "About us" page, which doesn't give me great confidence about their ability to get little details right in domain registration either. ("it's" should be "its", and one hopes that the founder has a "Juris Doctor" degree from an actual law school as opposed to a "Jurist Doctor" degree from an illiterate mail-order diploma printer...)
Sure, spelling/grammar flames are normally tacky, but not only do compilers bitch at you for getting your programming-language spelling and syntax incorrect (thus geeks ought to be good at them), but your ability to get spelling and grammar correct does indicate something about your level of effort, and about your willingness to use automated tools (like spell-checkers) if you're not natively good at spelling, grammar, or (in my case) typing.
It turns out that for a whole range of applications, including VOIP, you can detect most of the packets that need QoS by simply picking small UDP packets. (In fact, almost ANY UDP packet, except for BitTorrent and file-sharing protocols, tends to be good prioritization material - and file-sharing shouldn't be going out over your WAN except through secure tunnels anyway.)
Some blocklist implementations tell the sender's MTA they've been blocked, so the MTA can give the sender some useful information about the problem. Some blocklist implementations trash the sender's email silently, because if the sender _is_ a spammer, that feedback would let them listwash and verify which addresses were valid and can be sold for more money.
A few blocklists really only block verified spammers and/or verified spam-abuse tools (open relays/proxies/zombies/whatever). Others do less verification, or deliberately cause collateral damage by blocking increasingly large address ranges around detected spammers, or don't respond to cleanups, or generally go off the wall. Both approaches are helpful - collateral damage from MAPS *was* what caused ISPs to take the open relay problem seriously, annoying as it was, and it does increase the number of not-yet-reported spammers that you can block, at the cost of some false positives, and some lists like Dialup / Consumer Broadband blockers mostly only annoy those unimportant people running their own Linux machines (:-) while blocking lots of zombies.
So what do you do? They're all useful as SpamAssassin weights, or for filters that decide which messages get the full-blast SpamAssassin treatment and which ones don't, because most of them do have some information about the likelihood of a given source being a spammer, even if you don't want to trust some of them not to get lots of false positives. MAPS is, IMHO, in this category.
They're also useful for greylists, at least until spamware authors figure out how to work around greylists. After all, a false positive isn't a big problem for a greylist, because real mailers will keep trying. They can also be useful for teergrubes, if you're running the kind that does eventually accept messages eventually as opposed to junking them entirely.
As far as 4) goes, I've been on Usenet since 1981, and mainly using one email address for some noisy mailing lists for almost a decade. Waaaayyyy too late for that one:-)
Any time the legitimate business uses a term for a while, the spammers starting using it to pretend they're legitimate. Then the legitimate people have to go find another term. Then the spammers start using that. At some point most legitimate people get bored with keeping up with the Joneses.
After all, the term "opt-in" once meant that the recipient had actually opted in to the list. Then it meant "we're lying about the user having opted in to the list."
Wargames, yes, Nine Billion Names of God, no
on
The Baby Bootstrap?
·
· Score: 1
The computer in WarGames was a learning system, and teaching it examples of not doing stupid things so it wouldn't do the one particular stupid thing was part of the plot.
But the Nine Billion Names Of God wasn't a learning system - the monks had figured out that just reciting all the names of God would do the trick, and printing them out on paper using an appropriate Tibetan font would do the trick. It turns out that Tibetan typography is actually rather more complicated than in the story, so simply making an X-Windows counter that runs in the background isn't going to do the job very well:-)
There are a lot of folks out there who don't like free speech, and while the Right Wing are much more enthusiastic about enforcing political correctness, the left wingers aren't always much better.
If you're talking about pr0n on the Internet, they'll tell you that the First Amendment is designed to protect political speech, and that it doesn't apply to indecent or obscene speech.
If you're talking about tobacco advertising, they'll tell you that the First Amendment doesn't apply to commercial speech, only non-commercial speech, especially political speech, and that we have to ban tobacco advertising to Protect the Children.
But if you're talking about campaign financing, well, no, no, Elections are WAYYYY too important to allow unrestricted political speech, especially speech that people pay to print or broadcast.
And anonymous political speech? We certainly can't have that at all! Why, people could say Bad Things about our Politicians, and the public wouldn't know not to believe them!
And anonymous non-political speech? Unnamed Administration Sources say that that technology is too dangerous, because somebody might use it to talk about politics without filling out campaign finance papers, or to express terrorist opinions. Oh, or pr0n.
It's bad enough to hear this kind of demand for censorship coming from sources like "Right-Wingers For A Politically Correct Tomorrow!" But this is San Francisco, a city that's famous for hippies, rock&roll, beatniks reading subversive poetry, underground newspapers, anarchist bookstores, and proximity to Berkeley, where the Free Speech Movement advocated the right to stand on top of a car with a bullhorn and call the [expletive-deleted] University Administration and the [expletive-deleted] Governor Ronald Reagan a bunch of [Expletive deleted][expletive-deleted]s.
And if the Supes can't figure out that the First Amendment applies to political speech in all forms, well then [expletive deleted], it's time to start a blog just to advocate throwing the [expletive deleted] bums out.
And you don't get your stinkin' papers, either - only journalists working for established journalism businesses get press passes (except for the White House, where the Department of Toady Affairs can arrange them.) And by the way, if you're planning to enter the US and you're hoping that calling yourself a Journalist will get you better treatment, it turns out that foreign journalists have technically needed special journalist visas since the McCarthy witch-hunt days...
Some people probably use these boards to actually talk to the RS232, or to talk parallel port to an older printer or whatever. I suspect the real issue is that they're using an older SouthBridge chip that has all those ports on it, so they figure they might as well populate them.
Personally I'd rather have a couple of extra USB2 ports and an extra network port (though you can add the network port via USB, or you can add a wireless connection via USB since there's no PCMCIA.)
Sitefinder did a lot more than hijack unclaimed URLs - it did so without discussing the process with the community first, and it did so in a way that broke DNS for services other than Port 80 http.
For instance, if you tried to telnet/ssh to missspelled-example.com, the right response is to return a "missspelled-example.com not found" packet, not to return sitefinder's IP address and then not be running telnet/ssh there.
Similarly, if you tried to connect to https://missspelled-example.com on port 443, the right thing is to return that the DNS doesn't resolve, not to return sitefinder's address and hijack the connection.
I don't know what it did to firewall-penetration applications that use Port 80 for something other than HTTP, e.g. some of Skype's games. It's one thing to say that tunnel-servers-over-http are evil and deserve to be broken occasionally (:-), but they still should be broken correctly.
North Korea's government is primarily a clear and present danger to its citizens, who are suffering from severe poverty because it's a blatantly incompetent personality cult who keeps them from farming or trading successfully. Sure, its leaders occasionally say "Booga booga booga!" to the world just so someone will take them half-seriously - it helps keep the peasants in line. It's possible that some day they'll decide they have to actually nuke Seoul to keep themselves in power, but they do know that they'll be pounded into little bits of probably-radioactive dust if they ever try it, so it's at most a total desperation move.
Now, they're sufficiently wacko that that that's not an impossibility, but the real threat to them isn't military - it's that the US, South Korea, or Japan might just start air-dropping handheld TV sets on them and let them see the propaganda that modern industrial commercialism puts out for their markets (government propaganda directed at the North Koreans would probably be less effective - random Korean MTV with restaurant and grocery store and clothing commercials tends to be the really devastating stuff for communist regimes.)
Yes, I know that SETI@Home is actually running simpler problems than LINPAK, but it's still frustrating that the fastest computer in the world is being used for evil. Until this computer, the fastest computer in the world was a volunteer effort searching for space aliens using screen savers. Last November's numbers had the Blue Gene at 71 TFLOPS, which might have been faster than SETI (SETI's currently at 56 TFLOPS; I'm not sure if that's a decline or increase from last fall, given the number of people who have switched over to Folding@Home, various cancer-fighting applications, etc., vs. people who've gotten faster computers.) Number 2 was at NASA AMES, so it could have been working for good or evil applications, and Number 3 was the previous champion Earth Simulator, which is on the good side.
Slashdot Mirror
I've got analog cable, and no plans to upgrade. Does this mean that I need to use an analog card like the Hauppage, and that if I had digital cable I'd need a digital card instead?
However, you should still do so openly - build interfaces that people can use, and document them so people can figure out how to use them, and if you're lucky, people will use them for things you've never thought of, so try not to prevent that.
Increasing density is valuable, not only because it lets you put more transistors per chip, but also because the smaller transistors mean shorter travel distances for electons/holes, and therefore faster computation and higher clock rates. But if you could do the mythical true 3D design, instead of just a few layers, then you'd have *lots* of transistors a few microns or tens of microns away from each other instead of hundreds to thousands of microns away on the other side of a chip. Maybe you can't do that easily in silicon, or maybe a coarse-3D approach will help (e.g. 10-20 layers of chip stacked on top of each other, as opposed to hundreds of layers), or maybe carbon nanotubes or buckyballs or nano-unobtainium or whatever will be more flexible. Or maybe holographic memories could be useful. Who knows?
It's hypothetically possible that the Quantum Computing people might make some breakthrough that lets some kinds of problems be solved in small-polynomial time instead of exponential time, with some usable probability of a correct answer, so you'll have to start filling those liquid-cooling systems with liquid helium. (As a cryptographer, I'd find this very annoying, because most or all of the currently useful public-key technology would get trashed, but as a combinatoric mathematician, I'd find it to be really really cool :-) That's definitely not a Moore's Law approach to computers - it's major theoretical breakthroughs as opposed to continual rapid improvement due to technical investment and lots of minor theoretical breakthroughs.
The reason zombies are heavy DNS users is that they're trying to send a million emails a day, so they need to look up probably 10K-100K recipient's domains, depending on how much reuse they can get away with, and even if they're running on a PC with a caching DNS server, they're going to blow out the cache if they don't have them all sorted (and they probably *don't* have them all sorted, because they're trying to evade spam detection on the recipients' ISPs.) Also, the spamware probably doesn't have very intelligent DNS handling in it - if it did, it would probably go to some other DNS server or do something else to evade detection, though using the ISP's server does scale well if the ISP is competent.
An intermediate step they could take would be to put heavy DNS users on different DNS servers than the light users. Most PCs get their DNS server addresses from DHCP configuration, so they can do fancy things at DHCP lease renewal time like load-balance their server assignments, but they can also concentrate the heavy users. That might be a good way to put their email connections through more thorough spam-filtering than the average user gets as well.
But there are several other protocols for sending email that don't look like Port 25 to the ISP. There are a couple of SMTP-submission protocols which let you set up a connection to a mail server where you have an account and do various kinds of authentication, including some that use SSL encryption. Alternatively, you can do SSH or IPSEC or other VPN tunnels to your email provider. And then for us old folks, there's always "login to a shell account" :-) (Kids can use webmail instead.)
As far as email-over-telepathy goes, Dan Kaminsky recently demonstrated IP-over-DNS tunnelling at Codecon. It's really really evil - he was even able to do video-over-IP-over DNS by coopting about 25000 DNS servers. I'm pretty sure he was the guy who did a lot of the IP-over-HTTP tunnelling a couple of years back, and he;s done lots of other creative work with detailed protocol analysis.
Also, "broken" printers *can* be a problem for desktop supporters. Sometimes the problem is the printer hardware, or the ink running out, or the coffee-cup holder on their machine breaking, but often the problem is something with Windows setup. I'm using a work-managed Windows 2000 laptop, and there's some sort of permissions problem that keeps me from using my USB printer at home (it can support it on a parallel port, but if I plug in non-storage USB devices it says I'm not allowed to do that.) My home PC supported the same printer ok until XP SP2 came along, and now it complains about drivers every time I reboot because I'm using the vendor's driver and not Microsoft's (it still works fine - *because* I'm using the vendor's driver and not Microsoft's :-)
Also, it's a lot harder to tell what's really spyware - most of the spyware products I've used complain if they see cookies from Usual Suspects, but some of them have complained about the adware in Sponsored-Mode Eudora (yes, it's there, and it's the cost of using the free version of the software.) Some of it has even complained about things in found in my Windows Recycle Bin
But a spammer *could* adapt to this by using DNS servers that aren't from the local ISP, or using spamware that downloads the victim's IP address along with their domain name. It's an arms race. Blocking Port 25 is more effective - Blocking it for everybody is a Bad Thing, but blocking it by default and letting users enable it themselves is fine. But until the spammers start working around DNS servers, they're a potentially useful hook for identifying problem customers.
Real privatization usually has some market distortions, because the former monopoly is usually in a strong position, and it takes a while for competition to build up. But it's better than not doing it. On the other hand, bogus privatization, e.g. spinning off the company into a profit-making corporation but letting them keep monopoly power through regulatory mechanisms and owned by friends&relatives of the politicians in power is usualy worse. (Not always - even a quasi-monopoly can occasionally see that it makes more money if its customers like it and if it has higher-value services to offer.) A typical problem in places like Carribbean islands was that the monopoly PTT was Cable&Wireless with a few highly-paid positions for high politicians, and they'd insist on maintaining an expensive and antiquated wired telephone system because it "creates jobs" for lots of people, when it would be cheaper to replace the whole system with a couple of cell towers stuck on different sides of the mountain.
It's a bit hard to moderate "overrated" well, because you'd really want to know what the rating was when the moderator moderated it.
I use BitTorrent to download and upload lots of music on my DSL line - all of it legal concert tapes from etree etc. or free downloads. And I also use it for Linux and other software distributions.
If Jumpdomain's just a reseller for eNom, then eNom is definitely the right place to go, because they're the real registrar.
Even if you are or are not maintaining the website for her, the domain name ought to be hers, and if you're being stubborn about the $6-$35 "ownership" you might have in it, you're being a jerk about the wrong things and ought to put the energy into something else. And if she feels like pointing the domain name somewhere else and writing a blog or a rant about her ex-boyfriend, well, it goes with the territory :-)
So Jump Domain's "About Us" page calls itself a legal disclaimer, not that it's particularly disclaiming anything, though it does indicate an unwillingness to provide accurate contact information, like the name of the human who operates it.
Sure, spelling/grammar flames are normally tacky, but not only do compilers bitch at you for getting your programming-language spelling and syntax incorrect (thus geeks ought to be good at them), but your ability to get spelling and grammar correct does indicate something about your level of effort, and about your willingness to use automated tools (like spell-checkers) if you're not natively good at spelling, grammar, or (in my case) typing.
And the "About Us
It turns out that for a whole range of applications, including VOIP, you can detect most of the packets that need QoS by simply picking small UDP packets. (In fact, almost ANY UDP packet, except for BitTorrent and file-sharing protocols, tends to be good prioritization material - and file-sharing shouldn't be going out over your WAN except through secure tunnels anyway.)
Some blocklist implementations tell the sender's MTA they've been blocked, so the MTA can give the sender some useful information about the problem. Some blocklist implementations trash the sender's email silently, because if the sender _is_ a spammer, that feedback would let them listwash and verify which addresses were valid and can be sold for more money.
So what do you do? They're all useful as SpamAssassin weights, or for filters that decide which messages get the full-blast SpamAssassin treatment and which ones don't, because most of them do have some information about the likelihood of a given source being a spammer, even if you don't want to trust some of them not to get lots of false positives. MAPS is, IMHO, in this category.
They're also useful for greylists, at least until spamware authors figure out how to work around greylists. After all, a false positive isn't a big problem for a greylist, because real mailers will keep trying. They can also be useful for teergrubes, if you're running the kind that does eventually accept messages eventually as opposed to junking them entirely.
As far as 4) goes, I've been on Usenet since 1981, and mainly using one email address for some noisy mailing lists for almost a decade. Waaaayyyy too late for that one :-)
After all, the term "opt-in" once meant that the recipient had actually opted in to the list. Then it meant "we're lying about the user having opted in to the list."
But the Nine Billion Names Of God wasn't a learning system - the monks had figured out that just reciting all the names of God would do the trick, and printing them out on paper using an appropriate Tibetan font would do the trick. It turns out that Tibetan typography is actually rather more complicated than in the story, so simply making an X-Windows counter that runs in the background isn't going to do the job very well :-)
- If you're talking about pr0n on the Internet, they'll tell you that the First Amendment is designed to protect political speech, and that it doesn't apply to indecent or obscene speech.
- If you're talking about tobacco advertising, they'll tell you that the First Amendment doesn't apply to commercial speech, only non-commercial speech, especially political speech, and that we have to ban tobacco advertising to Protect the Children.
- But if you're talking about campaign financing, well, no, no, Elections are WAYYYY too important to allow unrestricted political speech, especially speech that people pay to print or broadcast.
- And anonymous political speech? We certainly can't have that at all! Why, people could say Bad Things about our Politicians, and the public wouldn't know not to believe them!
- And anonymous non-political speech? Unnamed Administration Sources say that that technology is too dangerous, because somebody might use it to talk about politics without filling out campaign finance papers, or to express terrorist opinions. Oh, or pr0n.
It's bad enough to hear this kind of demand for censorship coming from sources like "Right-Wingers For A Politically Correct Tomorrow!" But this is San Francisco, a city that's famous for hippies, rock&roll, beatniks reading subversive poetry, underground newspapers, anarchist bookstores, and proximity to Berkeley, where the Free Speech Movement advocated the right to stand on top of a car with a bullhorn and call the [expletive-deleted] University Administration and the [expletive-deleted] Governor Ronald Reagan a bunch of [Expletive deleted][expletive-deleted]s.And if the Supes can't figure out that the First Amendment applies to political speech in all forms, well then [expletive deleted], it's time to start a blog just to advocate throwing the [expletive deleted] bums out.
And you don't get your stinkin' papers, either - only journalists working for established journalism businesses get press passes (except for the White House, where the Department of Toady Affairs can arrange them.) And by the way, if you're planning to enter the US and you're hoping that calling yourself a Journalist will get you better treatment, it turns out that foreign journalists have technically needed special journalist visas since the McCarthy witch-hunt days...
Personally I'd rather have a couple of extra USB2 ports and an extra network port (though you can add the network port via USB, or you can add a wireless connection via USB since there's no PCMCIA.)
Now, they're sufficiently wacko that that that's not an impossibility, but the real threat to them isn't military - it's that the US, South Korea, or Japan might just start air-dropping handheld TV sets on them and let them see the propaganda that modern industrial commercialism puts out for their markets (government propaganda directed at the North Koreans would probably be less effective - random Korean MTV with restaurant and grocery store and clothing commercials tends to be the really devastating stuff for communist regimes.)
I'm currently running Folding@Home...