In one sense he's got a great plan - "Just Do It! You first, then you, then you, then everybody!" Sure, it doesn't tell you how to do it, but somebody's got to get off their butt and start. However, he's missing a couple of really fundamental problems with IPv6.
Some of them will get eaten up by Moore's Law (you need more router memory, but memory keeps getting bigger and cheaper.) But even so, the IPv6 versions of hardware will cost more than the IPv4 versions, it's just that they'll both get cheaper.
Some of them will only get eaten up by Moore's Law if there's perceived market demand to tell router vendors (or at least Cisco) that it's worth putting in IPv6-capable ASICs or dual-stack ASICs or faster CPUs or whatever they need, and to tell ISPs that they need to beat up their router vendors to get them to do it.
Some of them are software-related and fixable- making sure DNS handlers do the right thing in mixed-version environments when not everybody has IPv6 connectivity, etc.
But some problems are really hard, and critical, and the technical community hasn't solved them yet. The one I deal with most is multi-homing - businesses want to have connectivity from multiple ISPs for reliability, and they want to use ONE IP address block for their servers across all their ISPs, so that clients don't lose sessions even if one of their ISP connections goes down. The traditional solution was for the user to get their own provider-independent address space, which would get advertised to every large router in the Internet, in contrast to a single-homed user whose address gets aggregated together into one block with the rest of their ISP's customers. The demand is growing rapidly, and the main thing that keeps it in check is that businesses often put their servers in colo centers and use addresses provided by the colo. There's an IETF working group doing an ugly hack called shim6 that's supposed to provide an alternative to fix this, but I'm not convinced that approach will work.
Then there are other problems that are hard but less urgent. The Mobile IPv6 folks are trying to do really cool stuff, but they need IPv6 anyway, and there are a bunch of projects like that. And IPSEC was hard, but IPv4 adopted it.
And there are problems that have other solutions these days - one reason IPv6 addresses have 128 bits is so there's enough room to do Netware-like auto-addressing using the 48-bit MAC address, but DHCP is almost as easy, and avoids some of the potential privacy problems, and by now everybody does DHCP.
The extra 3.4% of header bits aren't the main performance problem - nobody worries about them that much.
The big problem is that many kinds of routers have two ways to route packets - either by using ASIC hardware to do easy jobs fast, or by using a general-purpose CPU to do more complex jobs, and if the ASIC doesn't support IPv6, you have to do it in the CPU, which is a limited resource on most routers from some popular vendors. That's a problem that time and design work can fix, but it'll be a while before IPv6-capable routers catch up with IPv4 equivalents in terms of features and performance. The extra address bits _are_ a problem if you're designing ASICs - the address is 4x as big, so you can fit roughly 1/4 as much performance on a given sized chip. Among other things, your forwarding tables are 4x as large.
A separate CPU problem is that the CPU needs to run routing protocols and occasionally need to recalculate the tables, which typically takes N**2 space and therefore a lot of RAM. For some reason, a gigabyte of RAM that costs $100 if you install it in a beige-colored box costs up to $5000 if you install it in a teal-colored box, at least if you're buying a service contract for the teal-colored box:-) But a gig or so seems to be enough RAM for Internet BGP tables, for now, and the people who make teal-colored boxes could perfectly well fix the problem.
The more serious problem is that IPv6 was supposed to do something about hierarchical routing structures that were supposed to reduce the table size growth, but it doesn't realistically support addressing for sites that get connectivity from two different ISPs for redundancy and reliability, which is becoming increasingly common for businesses. There are ugly hacks like Shim6 that are supposed to address this, but don't really cut it.
A few years back there were a couple of guys in New Jersey who got murdered. It was generally believed that they were spammers who were pushing pump&dump stock scams and annoyed the local branch of the Russian Mafiya. Don't know if they'd sold stock to some mafiaya guy, or scammed people that they'd been trying to scam, or if it the murders weren't directly spam-related (e.g. gambling debts or something.)
I didn't get the impression from the article that he was actually trying to hire a hitman, but he did apparently say things that could be interpreted as intimidating a witness, which is potentially a crime (I don't know if he got charged with that or not.)
IMHO, fleeing isn't morally reprehensible, but fleeing and starting another spam business after being ordered to shut down the first one is behaviour that Darwin doesn't take kindly to. And doing so before you've been sentenced is just really really dumb.
Hiding money that the Feds want to confiscate is one thing (at least morally; legally it's obviously a Bad Idea.) But a lot of the money he was trying to hide may have been money he'd been ordered to pay is spamming victims like AOL. (Too bad there's no easy way to forcing him to identify and pay each of the people he'd sent spam to, as opposed to just their ISP - their wasted time is worth a lot more than the extra bits AOL had to handle.)
According to Wikipedia article and this article, Smith sold a lot of different things over the years. Some of them allegedly included
Genuine pharmaceuticals, with high prices and potentially-dodgy prescriptions (e.g. his pet doctor would prescribe you hydrocodone).
Penis pills, including genuine Viagra and also dubious enhancement products.
Cable TV descramblers of dubious legality
Fake college degrees (or real degrees from non-accredited colleges, or something like that.)
He also violated anti-spam laws, and was ordered to pay AOL a lot of money for spamming their customers, and has been accused of using a variety of less-legitimate methods to get his spam delivered.
Also, of course, after being convicted but before being sentenced, he tried to stash some of money where the Feds couldn't get it, which is not the brightest way to get the judge to take it easy on you.
I certainly hope you asked to speak to the manager to get a refund. There are a couple of reasons for doing so - one is to get your ~$10 back, or possibly to get the manager to tell the goon to let you in, but more important is to keep the management aware that what they're doing is stupid and annoying and will lose them customers.
*Everybody* has phones, and almost all phones these days have cameras whether they need them or not, and it's none of the theater's business to mess with you about them, even though you *could* use them to take grainy out-of-focus clips of the movie. Hassling people who bring in professional-quality shoulder-mounted cameras is a different matter (:-), but even professional-quality stuff keeps getting smaller.
Greylist to save CPU before using DNSBLs, SpamAss
on
Choosing a Good DNSBL
·
· Score: 1
SpamAssassin and commercial solutions use a lot of CPU, but there are other low-CPU ways to defend yourself besides DNSBLs. Greylists are a popular one - mail from unknown mailserver addresses gets told to go away and try again in 5 minutes or an hour or whatever. They're currently effective against most zombies, and inherently effective against some of the stolen-address-space attacks, and because they're not permanently blocking mail, false positives aren't a problem. It works at the SMTP-header level, so you don't need to accept the message body or process it. It's not a perfect defense, but I've seen people reporting 80-90% of the spam goes away for minimal CPU cost.
There are also people combining greylists with DNSBLs - senders from blacklisted addresses get told to wait much longer than non-blacklisted addresses, or they get told to wait and non-blacklisted addresses don't.
Even if you're going to also reject on DNSBLs, this'll let you be less aggressive about it, e.g. use SpamHaus's list of known big spammers, then greylist, and use the other DNSBLs only as SpamAssassin weight, or greylist first, then use Spamhaus on the people who called back; you could also do some analysis to see how many of the greylist rejects are covered by people from which RBLs.
Most DNSBLs have problems, and there are few that I'd trust absolutely, though Spamhaus runs a tight enough shop that I'd trust it. But DNSBLs can be used effectively to augment other tools:
SpamAssassin weights - most of the DNSBLs are worth a couple of points of SpamAssassin weight; even rabid ones like SORBS can give you some information, and the country-specific ones are also useful here (e.g. mail from China had better not look spammy at all.)
Greylist Augmentation - The big value of DNSBLs is that you can reject mail from the SMTP headers without needing to receive the message body and grind it through CPU-instensive content filtering. But Greylists also do this, and some people have been using DNSBLs to tune their greylists (e.g. if it's on the DNSBL, then tell the sender to call back in an hour instead of 5 minutes.) Among other things, that gives you a way to use the lists of Dynamic-Address broadband users - the home Linux servers will call you back, the zombies won't, so the list gives you information which you might otherwise have to ignore. And country-code DNSBLs can also get forced to wait an extra hour for spammy places that you don't get much mail from.
TMDA Autoresponders - One of the most annoying and effective anti-spam tools is autoresponders that say "I don't recognize your address - respond to this mail and prove you're a human". You could integrate this with a DNSBL - if the mail's not whitelisted, and it's on some DNSBLs, then maybe it gets a TMDA test instead of bit-bucket. It's lower CPU than SpamAssassin.
DNSBL integration with DNS Servers? - One of my pet projects for when I get some copious spare time is to munge a DNS server to check blacklists/whitelists. Trusted or non-blacklisted sites get the MX record for the good mailserver, non-blacklisted sites get the MX record for the heavily-filtered mailserver that occasionally overloads the CPU, blacklisted sites get the MX record for the teergrube or 127.0.0.1. It's certainly not foolproof - many systems are likely to check their ISP's DNS cache before hitting your DNS directly, and if spammers want to do a set of DNS queries from a clean server they could - but at least at the corporate-email level (i.e. where you can afford multiple mail servers) it gives you a way to avoid having your mail server lose mail from legitimate sources because it's overloaded with SpamAssassin CPU load.
I originally thought of this back when Open Relays were the popular spam threat - if you get a DNS MX request from an open relay, tell them that the IP address for spambait.yourdomain.com is some other open relay's address. That would let them spend their time sending mail to each other. But spammers moved on to open proxies and then zombies, so that opportunity went away.
DNSBLs are a really good combination with greylisting - some of the sites you don't want to hear from are running real SMTP servers, but many of them are running zombieware or tuned-for-speed spamware, and setting your greylists to discourage them for a couple of hours instead of just five minutes can help. Also, while greylisting can block legitimate mail from dialup users, it's no problem for DSL/cable users, so you can use those DNSBLs to keep longer greylist times on those, which will also discourage zombies but work fine for home Linux users.
Another class of anti-spam tool that can benefit from greylist info is things like TMDA, those annoying autoresponders that say "I don't know who you are, so click this link/captcha/etc. to prove you're not a spammer". Humans don't like the things, but if you occasionally get mail from spam-heavy places like China, it gives them a way to get through to you that's better than just blocking, and it can be pretty low-CPU, unlike running SpamAssassin.
Ok, perhaps in person Kip Hawley is not an idiot. But he's still running an organization that's offensive, dishonest, unconstitutional, and a bunch of thugs, and the fact that he does have a grip on reality doesn't change that.
I've had flaky behaviour from iPods depending on which of my Windows machines they're plugged into (work laptop, home desktop, etc.) and whether iTunes is running - sometimes they don't seem to charge. I was considering whether to pay Apple's prices for a charger, but decided to buy a powered USB hub instead, since I needed a USB2 hub anyway - AFAICT it charges the iPod ok if it's *not* plugged into the PC.
I don't get the marketing literature written for repressive government ISPs (:-), but I do get the stuff written for corporate computer security and IT departments, and the Deep Packet Inspection folks are definitely marketing to them for their corporate firewalls. Regular firewalls cover a lot of different potential threats, but as more and more applications get written to use Port 80 to get around regular firewalls, the firewall business adapts, and that's what deep packet inspection lets you do.
Part of it's a concern about viruses, and part of it's a concern about cost management, because applications like BitTorrent can really suck down your relatively-expensive corporate bandwidth, and partly it's general fear about having applications like Skype running servers in your network that are providing services to outsiders and aren't under the control of your corporate computer security or even desktop support organizations. Now, it's true that a lot of that fear is FUD generated by people who want to sell you fancier firewalls, but there are some legitimate concerns as well.
In general, what you really want to do is prioritize the VOIP so it gets high priority, but doesn't crowd out your latency-sensitive database applications, and put BitTorrent at lowest priority, because it's good at scavenging anything left over after web, email, and FTP get what they need, assuming of course that it's being used for work-related activity (Linux Distros good, music downloading maybe not.)
In general, what long-haul carriers own is easements letting them run fiber under land owned by various people, or else fibers that they buy or rent from other carriers who own easements and sometimes conduit. It's especially common for them to own easements along railroads, because they're nice long contiguous chunks of land owned by single owners, plus there are customized conduit-diggers built to run on train tracks. (You may remember that SPRint was started by the Southern Pacific Railroad?) There are also lots of easements for cable routes across large farms.
I don't know how ownership of easements along highways works, but that also depends on whether the telcos are buying easements from the highway departments or adjacent landowners. one reason there's so little carrier infrastructure in North Dakota, besides low population, is that the state highway department didn't want to provide access to carriers for a long time on the main east-west route across the state, so instead there was a trickle of fiber coming in from the eastern side.
I don't know how ownership relations work for the local telco portions between the telco office and the customer's home or business - it's a lot more varied, and the scale is different. Where the wiring is aerial, there are lots of different relationships depending on whether the poles are owned by the telco or the power company (and who rents them from the other one under what arrangements.) When I owned a house in a small town, my deed did include a "utility easement" that let the telco and power company run wires and poles along the six feet near the road, and in general underground utilities have similar easements. But it bigger cities, it's a lot messier, and the cities often extort various deals from the utility companies in return for giving them access to the streets - that's most visible with cable TV.
I'm not sure which AT&T service you're talking about "still trying to migrate off ATM". DSL access is ATM underneath, so it's natural to keep the DSLAMs connected by ATM, and it's the cleanest way to handle multiple ISPs on a DSL network, because you can provide pure Layer-2 connections from the user to the ISP. There are DSLAMs that terminate the ATM layer and integrate IP routing into the DSLAM itself, and that's fine if you're only supporting a single ISP on that network, but if you're supporting multiple ISPs (which DSL providers like old-SBC and Covad and New Edge do), then if you do routing at the DSLAM, that forces you to do something ugly like PPPoE to connect the users to their ISPs.
The old-AT&T Internet backbone migrated off ATM back in the 90s, except for access to a few smaller countries. On the other hand, business customers buy a lot of ATM and Frame Relay for private networks, and the frame networks use an ATM backbone. The old SBC network used a lot of ATM and frame to transport everything, at least in California, but I'm not sure if that's what you mean.
In my case, I don't have kids, and I replace my home computers seldom enough that new motherboards probably don't fit and the old computers mostly end up under the desk somewhere because they're not good garage-sale fodder:-)... But for normal people, if they've got kids, that's what happens to the old computers. If their kids are older, sometimes it's the kids who get the new gamer-class computer and the adults get the kids' leftovers, but usually not.
Ok, you've only got 4 USB ports. But hubs are cheap, and get you all the expansion you're likely to need, and even powered hubs are pretty cheap, though they're a bit more annoying. (And a powered hub that's not plugged into a computer looks a lot like an iPod charger:-)
The specs say it can support up to 1GB of RAM. That's enough for most people, but some people will find it limiting. (It comes with 256MB, so you'll need to spend some cash to make it more useful.)
Shouldn't be any reason there's a problem with upgrading disks. I don't know if there's any antique 137-GB limitation in the BIOS - if there is, then you might be limited to 120GB.
1 PCMCIA slot, 3 USB, built-in 802.11g, S/PDIF sound. Because it's 802.11g, you won't need to burn the PCMCIA slot immediately for wireless, though if you're doing.11n you may want to.
Old mainframes did virtualization because it was the simplest way to get the capabilities back then - the guest OS didn't typically have good timesharing schedulers, or good permission mechanisms, or reentrant filesystems, and what they were _really_ good at was batch scheduling and handling large numbers of semi-intelligent terminals and lots of disks. Virtualization let them take their fairly lame OS's and run multiple versions on the same machine so they could share expensive hardware between user groups, and also let them run radically different schedulers in the different partitions (e.g. a batch scheduler in one and an interactuve one in another.)
I remember when we first got VM/CMS back when I was an undergrad - I could now allocation a whole megabyte of virtual address space, which made it possible to crunch bigger matrices for engineering problems.
I'm not really interested in building Windows versions of the browser from source - other people have presumably done that, since they're shipping the source (:-), and the little development I do these days (and lots that I used to do) tends to be on various flavors of Unix systems where there's a decent development environment and working OS. But I spend *lots* of time using Windows, and using Windows browsers, and having a browser that was lighter-weight than Mozilla would be a Good Thing.
The nightlies look like they're just source, and the various home pages and first layer or two of wikiness didn't seem to have any indication that they want to support users as opposed to developers.
I remember when I first ran Opera - there was no reason not to download it, because it fit on half a floppy disk. I didn't really like the tabbed interface (:-) and it wasn't compatible with some web pages, and it wasn't free (even free-as-in-beer), but it *was* certainly a lot smaller and faster than Netscape...
I don't know how good Hotmail is about reporting message rejections to the sender - though I also don't know how good his other ISPs are about it. Unfortunately some mail senders don't seem to be that good at telling the user, and many spam-blockers have the policy of accepting messages that pass the initial filters and not sending bouncegrams if they later decide something's spam.
I don't mind if the anti-spam system lies to the sender (e.g. "username unknown" instead of "die, spammer, die"), though I generally prefer true responses ("553 your system is RBL'd at http://etc/"). There's an argument that says that most of the mail you're rejecting is spam, so it's better to leave the spammers ignorant about how you recognized them as opposed to telling legitimate senders what kind of false positive they got hit by. IMHO that's mostly wrong-headed - most of the spammers aren't going to respond to your error message except by giving up on that delivery, and often they won't even bother removing the "bad" addresses from their target lists (though some of them make money selling that kind of detail to other spammers.) But there is some small fraction of spammers and spamware sellers who will use that information to push the arms race.
Friends of mine have an international human-rights organization that really *does* get legitimate email from some people in Nigeria who are probably using cybercafes....
No, he doesn't say if he checked the spam folders, but his outgoing mail from Hotmail to them got lost too. If he checked and didn't find them, that's interesting. If 81% ended up in spam buckets, then of course that's just probably-overactive spam filters.
It's possible that it ended up in the spam folders on his other ISPs - certainly *I'd* expect email from Hotmail containing a random attachment to be spam:-)
If you RTFM, you'd see that this was paid Hotmail service, not just the free service. So they ought to be providing professional quality service, and apparently they're not.
And as far as other ISPs charging you lots of money per month, that's not normally the case for *email* service. My DSL service does cost me about $50/month (but I've got static IP addresses), but my mail-forwarder is $15/year, my ISP where I've got a shell account and run procmail is $7/month, and my wife uses Fastmail as an email provider for $19/year (they've also got free mail and $15-onetime options.)
Their price-competitiveness depends on what you're using the phone for. If you just need an emergency phone, it's really cheap, and if you don't make a lot of calls it's not bad - you can find cheaper competition, but it's within the not-too-annoying range. My mom makes 1-2 calls a day, generally not very long, and their plans work reasonably well for her.
If you _do_ make a lot of calls, you can probably find a better deal. On the other hand, if you do a lot of calling, and your vision's good enough to read the screens on a cheap cellphone, it's probably worthwhile to shop around, find a phone you can tolerate, invest half an hour in reading the manual and just ignore the features you don't use.
Slashdot Mirror
Sure, the parent article was flamebait, but it made some useful points and attracted some useful flames.
The big problem is that many kinds of routers have two ways to route packets - either by using ASIC hardware to do easy jobs fast, or by using a general-purpose CPU to do more complex jobs, and if the ASIC doesn't support IPv6, you have to do it in the CPU, which is a limited resource on most routers from some popular vendors. That's a problem that time and design work can fix, but it'll be a while before IPv6-capable routers catch up with IPv4 equivalents in terms of features and performance. The extra address bits _are_ a problem if you're designing ASICs - the address is 4x as big, so you can fit roughly 1/4 as much performance on a given sized chip. Among other things, your forwarding tables are 4x as large.
A separate CPU problem is that the CPU needs to run routing protocols and occasionally need to recalculate the tables, which typically takes N**2 space and therefore a lot of RAM. For some reason, a gigabyte of RAM that costs $100 if you install it in a beige-colored box costs up to $5000 if you install it in a teal-colored box, at least if you're buying a service contract for the teal-colored box
The more serious problem is that IPv6 was supposed to do something about hierarchical routing structures that were supposed to reduce the table size growth, but it doesn't realistically support addressing for sites that get connectivity from two different ISPs for redundancy and reliability, which is becoming increasingly common for businesses. There are ugly hacks like Shim6 that are supposed to address this, but don't really cut it.
A few years back there were a couple of guys in New Jersey who got murdered. It was generally believed that they were spammers who were pushing pump&dump stock scams and annoyed the local branch of the Russian Mafiya. Don't know if they'd sold stock to some mafiaya guy, or scammed people that they'd been trying to scam, or if it the murders weren't directly spam-related (e.g. gambling debts or something.)
IMHO, fleeing isn't morally reprehensible, but fleeing and starting another spam business after being ordered to shut down the first one is behaviour that Darwin doesn't take kindly to. And doing so before you've been sentenced is just really really dumb.
Hiding money that the Feds want to confiscate is one thing (at least morally; legally it's obviously a Bad Idea.) But a lot of the money he was trying to hide may have been money he'd been ordered to pay is spamming victims like AOL. (Too bad there's no easy way to forcing him to identify and pay each of the people he'd sent spam to, as opposed to just their ISP - their wasted time is worth a lot more than the extra bits AOL had to handle.)
Also, of course, after being convicted but before being sentenced, he tried to stash some of money where the Feds couldn't get it, which is not the brightest way to get the judge to take it easy on you.
There are a couple of reasons for doing so - one is to get your ~$10 back, or possibly to get the manager to tell the goon to let you in, but more important is to keep the management aware that what they're doing is stupid and annoying and will lose them customers.
*Everybody* has phones, and almost all phones these days have cameras whether they need them or not, and it's none of the theater's business to mess with you about them, even though you *could* use them to take grainy out-of-focus clips of the movie. Hassling people who bring in professional-quality shoulder-mounted cameras is a different matter (:-), but even professional-quality stuff keeps getting smaller.
There are also people combining greylists with DNSBLs - senders from blacklisted addresses get told to wait much longer than non-blacklisted addresses, or they get told to wait and non-blacklisted addresses don't.
Even if you're going to also reject on DNSBLs, this'll let you be less aggressive about it, e.g. use SpamHaus's list of known big spammers, then greylist, and use the other DNSBLs only as SpamAssassin weight, or greylist first, then use Spamhaus on the people who called back; you could also do some analysis to see how many of the greylist rejects are covered by people from which RBLs.
TMDA Autoresponders - One of the most annoying and effective anti-spam tools is autoresponders that say "I don't recognize your address - respond to this mail and prove you're a human". You could integrate this with a DNSBL - if the mail's not whitelisted, and it's on some DNSBLs, then maybe it gets a TMDA test instead of bit-bucket. It's lower CPU than SpamAssassin.
I originally thought of this back when Open Relays were the popular spam threat - if you get a DNS MX request from an open relay, tell them that the IP address for spambait.yourdomain.com is some other open relay's address. That would let them spend their time sending mail to each other. But spammers moved on to open proxies and then zombies, so that opportunity went away.
Another class of anti-spam tool that can benefit from greylist info is things like TMDA, those annoying autoresponders that say "I don't know who you are, so click this link/captcha/etc. to prove you're not a spammer". Humans don't like the things, but if you occasionally get mail from spam-heavy places like China, it gives them a way to get through to you that's better than just blocking, and it can be pretty low-CPU, unlike running SpamAssassin.
Ok, perhaps in person Kip Hawley is not an idiot. But he's still running an organization that's offensive, dishonest, unconstitutional, and a bunch of thugs, and the fact that he does have a grip on reality doesn't change that.
I've had flaky behaviour from iPods depending on which of my Windows machines they're plugged into (work laptop, home desktop, etc.) and whether iTunes is running - sometimes they don't seem to charge. I was considering whether to pay Apple's prices for a charger, but decided to buy a powered USB hub instead, since I needed a USB2 hub anyway - AFAICT it charges the iPod ok if it's *not* plugged into the PC.
Part of it's a concern about viruses, and part of it's a concern about cost management, because applications like BitTorrent can really suck down your relatively-expensive corporate bandwidth, and partly it's general fear about having applications like Skype running servers in your network that are providing services to outsiders and aren't under the control of your corporate computer security or even desktop support organizations. Now, it's true that a lot of that fear is FUD generated by people who want to sell you fancier firewalls, but there are some legitimate concerns as well.
In general, what you really want to do is prioritize the VOIP so it gets high priority, but doesn't crowd out your latency-sensitive database applications, and put BitTorrent at lowest priority, because it's good at scavenging anything left over after web, email, and FTP get what they need, assuming of course that it's being used for work-related activity (Linux Distros good, music downloading maybe not.)
I don't know how ownership of easements along highways works, but that also depends on whether the telcos are buying easements from the highway departments or adjacent landowners. one reason there's so little carrier infrastructure in North Dakota, besides low population, is that the state highway department didn't want to provide access to carriers for a long time on the main east-west route across the state, so instead there was a trickle of fiber coming in from the eastern side.
I don't know how ownership relations work for the local telco portions between the telco office and the customer's home or business - it's a lot more varied, and the scale is different. Where the wiring is aerial, there are lots of different relationships depending on whether the poles are owned by the telco or the power company (and who rents them from the other one under what arrangements.) When I owned a house in a small town, my deed did include a "utility easement" that let the telco and power company run wires and poles along the six feet near the road, and in general underground utilities have similar easements. But it bigger cities, it's a lot messier, and the cities often extort various deals from the utility companies in return for giving them access to the streets - that's most visible with cable TV.
The old-AT&T Internet backbone migrated off ATM back in the 90s, except for access to a few smaller countries. On the other hand, business customers buy a lot of ATM and Frame Relay for private networks, and the frame networks use an ATM backbone. The old SBC network used a lot of ATM and frame to transport everything, at least in California, but I'm not sure if that's what you mean.
In my case, I don't have kids, and I replace my home computers seldom enough that new motherboards probably don't fit and the old computers mostly end up under the desk somewhere because they're not good garage-sale fodder :-)... But for normal people, if they've got kids, that's what happens to the old computers. If their kids are older, sometimes it's the kids who get the new gamer-class computer and the adults get the kids' leftovers, but usually not.
Ok, you've only got 4 USB ports. But hubs are cheap, and get you all the expansion you're likely to need, and even powered hubs are pretty cheap, though they're a bit more annoying. :-)
(And a powered hub that's not plugged into a computer looks a lot like an iPod charger
Shouldn't be any reason there's a problem with upgrading disks. I don't know if there's any antique 137-GB limitation in the BIOS - if there is, then you might be limited to 120GB.
1 PCMCIA slot, 3 USB, built-in 802.11g, S/PDIF sound. Because it's 802.11g, you won't need to burn the PCMCIA slot immediately for wireless, though if you're doing
I remember when we first got VM/CMS back when I was an undergrad - I could now allocation a whole megabyte of virtual address space, which made it possible to crunch bigger matrices for engineering problems.
The nightlies look like they're just source, and the various home pages and first layer or two of wikiness didn't seem to have any indication that they want to support users as opposed to developers.
I remember when I first ran Opera - there was no reason not to download it, because it fit on half a floppy disk. I didn't really like the tabbed interface (:-) and it wasn't compatible with some web pages, and it wasn't free (even free-as-in-beer), but it *was* certainly a lot smaller and faster than Netscape...
I don't mind if the anti-spam system lies to the sender (e.g. "username unknown" instead of "die, spammer, die"), though I generally prefer true responses ("553 your system is RBL'd at http://etc/"). There's an argument that says that most of the mail you're rejecting is spam, so it's better to leave the spammers ignorant about how you recognized them as opposed to telling legitimate senders what kind of false positive they got hit by. IMHO that's mostly wrong-headed - most of the spammers aren't going to respond to your error message except by giving up on that delivery, and often they won't even bother removing the "bad" addresses from their target lists (though some of them make money selling that kind of detail to other spammers.) But there is some small fraction of spammers and spamware sellers who will use that information to push the arms race.
Friends of mine have an international human-rights organization that really *does* get legitimate email from some people in Nigeria who are probably using cybercafes....
It's possible that it ended up in the spam folders on his other ISPs - certainly *I'd* expect email from Hotmail containing a random attachment to be spam
And as far as other ISPs charging you lots of money per month, that's not normally the case for *email* service. My DSL service does cost me about $50/month (but I've got static IP addresses), but my mail-forwarder is $15/year, my ISP where I've got a shell account and run procmail is $7/month, and my wife uses Fastmail as an email provider for $19/year (they've also got free mail and $15-onetime options.)
If you _do_ make a lot of calls, you can probably find a better deal. On the other hand, if you do a lot of calling, and your vision's good enough to read the screens on a cheap cellphone, it's probably worthwhile to shop around, find a phone you can tolerate, invest half an hour in reading the manual and just ignore the features you don't use.