She's had several different free email accounts, even Juno dialup back when that was free. She doesn't mind getting a bit of extra spam directed toward someone who's 5 years old and whose hobbies are sleeping and tropical fish.
She did lose her Hotmail account a couple of years ago when they adopted policies to protect childrens' private information, because she was under 13 and her parent didn't want to provide a credit card to verify age for a Passport login.
You can often get legal costs covered if you win, which is one reason the EFF and similar groups can often afford to be helpful. A first-glance look by a non-lawyer says that either their patents are totally bogus for this application or they're covered by prior art. The first one I looked at mentioned being about "terminals" and a central processor, and that certainly shouldn't apply here. There are also some similarities to the BT Hyperlink patent nonsense. It's especially worth talking to the other targets of this abuse.
....then you need to start posting opinions they'd rather not own, such as comments about monopolies being bogusly abusive, questioning and/or declaring the ownership of copyrights on the content of the postings, self-contradictory statements, Top Ten Bad Things about Steve Case, etc....
There's a story, possibly apocryphal, about a service call that Tandem got after the 1989 earthquake. They made highly reliable fault-tolerant servers, with every critical component replicated three or four times, so that if one burned out, it could detect it and keep going until you replaced the part. The customer told Tandem that it had fallen over on its face, and asked them to come turn the (large) box rightside up. It was *working* just fine (except they couldn't use the tape drive which was now on the bottom), but they figured that it would be safer to have Tandem onsite in case anything went wrong when getting it rightside up again.
Unfortunately, while that class of machines from a couple of manufacturers did have very reliable hardware, they tended to have highly proprietary funky operating systems. A friend of mine was trying to do air traffic control with one [machine to remain nameless, wasn't tandem] in ~86-87, and while the hardware was very reliable, the OS would crash weekly.
My computer lab had a Vax, and later a second Vax, back when 1 MIPS of CPU horsepower required 440-volt 3-phase power and gimongous air conditioning machines (we had two of them, for backup, but couldn't run both at once - remember Steven Wright's joke about getting a humidifier and a dehumidifier and letting them fight it out?) The AC units were designed not to come on automatically after the power goes off and on, which is a good way to design things with big honking motors - they'd sit there beeping until a human pushed the Start button. The Vax, however, didn't work that way. Depending on operating system version, it might boot all the way up or might stop in Single User mode, but either way it was using lots of power and generating lots of heat. The first time we had a power failure over the weekend, we came in and found the computer lab at 120 degrees with all the power shut down, but we couldn't turn the AC on - because the power conditioner system knew that 120 degrees was hot enough that there might be a fire, so turning on the AC would be Bad, so it wouldn't let you turn on the AC until the room cooled down. So we opened the back doors to the lab, borrowed a couple of desk fans, and waited an hour until the room was cool enough that it would allow us to turn on the AC, and another half hour to cool down to workable temperature.
We also had a video projector, which was a $20000 ceiling-mounted thing. You were supposed to turn it off when you were done using it, but sometimes people forgot. One Monday morning I came in, and the room was 120 degrees, and there was a puddle of oil on the table under the projector - because somebody had forgotten to turn it off, we'd had a power hit over the weekend, the room had overheated, and the projectors spent the weekend blowing 130-degree ceiling-height air through itself trying to cool down. We had to get the building carpenters to make a crate for it so we could ship it back to Canada to get it repaired...
I worked for a different part of that 3 1/2-letter acronym company, and we were going to set up a bunch of computers for a project that were crunching satellite data to be displayed live at a trade show. The stuff absolutely, positively had to be there overnight (or at least I was going to be cooling my heels in Ft. Collins for an extra day shortly before Christmas waiting for them), but nonetheless, the building shipping bureaucrats don't use Fedex - they used the shipper they usually deal with, which has a local jobber crate the things up with whatever else needs to be shipped to the same area, airfreights it out on Burlington Northern, and uncrates it at the far end. So the next morning, when it was supposed to be there, it didn't surprise me that it wasn't there. What *did* surprise me was that they didn't have this system computerized or at least centrally managed using dead trees and telephones. The only way for them to tell what equipment was in a given crate was to wait for it to arrive, open it up, and see what was in it. There simply wasn't any record saying that my boxes were in Crate#12345, except the records shipped inside the crate itself. Purely decentralized, and works really well unless something gets lost or you want to know where it is in transit. And it did arrive by the next day, and I had a nice spare day out in the mountains.
Looking at it from a decade later, I suppose it does have the advantage that if you're shipping stuff from New Jersey, where the shipping is controlled by Some Guys, you don't want to annoy Those Guys by saying "Hey, you ripped off my packages" when you could just file an insurance claim, assuming you bought their shipping insurance, which you do because "you wouldn' want anyt'ing bad ta happen ta dat package, now, wouldya?"...
My Vax had four RM05 disk drives (washing-machine sized, for you young folks:-), two tape drives (full-height rack), 4MB RAM (meant we needed two cabinets for the machine). We'd ordered the entire thing from DEC to avoid problems - except apparently the shipping wasn't from DEC. So of course the shipper dents up one of the RM05s, rendering that $35000 piece of equipment relatively useless. Took months to straighten out the paperwork and get it replaced. (I used part of that time to go order an operating system, which the supervisor I'd just started working for hadn't thought to do.)
By contrast, a friend of mine (hi, Doris!) was setting up an IBM-based shop a year or two later. One Saturday her machines arrived! The truckers set the ramp from the back of the truck onto the dock and rolled the two big drives onto it. While she was escorting one of the drivers into the building rolling the drive, the other guy backed the truck away, causing the other computer to fall three feet onto concrete. Ooops. She ended up talking to an IBM sales VP at home that day. He told her to accept the shipment and mark it as damaged, and they'd make good on it, so she did. They gave her the form, marked "damaged in shipment", which she crossed out and replaced with "dropped off loading dock"....
If you're installing directly on a machine with a 52x CD-ROM drive, fine, you can put up with playing jukebox with the 7 CDs (and if the installer is bright enough, you only need to use each one once....) But if you have a machine with no CDROM in it (i.e. a machine where you really _want_ Linux instead of MSWindows) this much data can be a big hassle. If you're doing a Netinstall, this either means you need a server machine with a spare 5GB of space to cache the CDs in, or else the installation process needs to be designed intelligently enough that you can swap CDs in and out of the drive in your netboot server machine.
Redhat 7.1, which didn't fit into a single CD, was really annoying about this, but I didn't *want* to go building Debian or Gentoo from scratch on my antique-store lab machines, I wanted a consistent environment. Is SuSe any better? Or should I always pick up another $100 disk drive every time I need to upgrade the OS?
Do you leave mail on your mail server, or download to your PC? If you're leaving it on the server, yes, you'll hit limits, but if you're running POP, and get a typical load of (sigh...) 100 spam/day at 5KB each, that's 500KB, and if your ISP isn't providing enough space to handle it for a couple of days, then you need to get them to provide more buffer space - they're not leaving enough room for people who get Microsoft Office Bloatware attachments mailed to them. The only time I'm aware of losing email because of mailbox overflow was when an account I didn't use much got Sircammed by somebody sending multiple copies of the same 200KB document every day - so I had to clean it out once or twice a week.
This 500KB of spam will take about 2-4 minutes of modem time to download it, which is annoying but won't break anybody's bank, and you'll be spending far more time on your modem reading slashdot - and this Slashdot page has a 9KB banner ad on it that you're downloading, which is probably typical for most web pages your read. That's about 200MB of spam a year, which would cost $1 for disk drive if you actually stored it (20GB disk at $100) (and might mean that your ISP is paying $1/year extra for storing your share of their spam - they're not keeping it for long, unless they're keeping full backups on disk for long times, but they may have multiple copies in several tiers of backup and higher-performance disks that you have.)
The real cost to the average internet user isn't the resources consumed - it's how much of their attention span gets consumed shoveling the stuff out the door and into the bit bucket, and if you're a Responsible Netizen and occasionally hunt down spammers or at elast spamcop them, that takes time also.
The High Priced hardware was always rock-solid - the HP35 calculators weren't the only things you could accidentally run through a snowblower without damaging them. However, their equipment was often very proprietary and quirky in the way it did things, with as much Not Invented Here So We'll Build Our Own Deliberately Different Spec as anybody in the industry. Remember the HP-IL Interface Loop for talking to peripherals (competing with Appletalk)? Remember using HP-IB (the IEEE-488 Interface Bus stuff) as a way to connect all sorts of things together? (It was pretty cool for what it did, and actually did make sense in the test-equipment world, but as a computer interface it meant you had to buy all the peripherals from HP.)
HP3000s were Really Funky Mainframe-like Things. The Unix-based machines ran HPUX, which was almost exactly like Unix, but had really bizarre ideas about how RS-232 should be dealt with, and I spent far too much of my career for a couple years haggling with it and with drivers for HP printers.
On the other hand, remember when HP printers came with manuals that actually told you what the escape sequences were so you could do anything you wanted, not just 'how to tell Microsoft Products X/Y/Z that you have an HP printer?? That's because they were written by engineers for engineers, so you could actually understand what the equipment was doing and how to use it. Nobody writes manuals like that any more, unfortunately.
Taco's probably using a Linux system, with some popular Mail Transfer Agent like Sendmail or Postfix or [3 or 4 others] that are smart enough to be able to deliver mail without needing help, like just about any other Unix machine since we started using domain names in the mid-1980s. If you're on the DUL, your main net connection is probably dialup or DSL or cable modem, so you're a MERE LUSER instead of owning a T1 all your own, but so what? You've got a computer with a Real Operating System, and there's no need to pretend you're a Windows-using couch potato that's running a mail client too dumb to deliver its own email - even if you are running a dumb client, sendmail on 127.0.0.1 fixes that problem.
In other words, Taco does have an smtp server of his own to access. On his own machine. Like he should. If you're a dialup user, it's beneficial to have an inbound mailbox server somewhere that's always connected, whether it delivers the mail to you by SMTP or POP/IMAP. But no need to do that for outbound.
The reason the DUL is helpful for blocking spammers is not because there's no legitimate reason for a dialup user to run SMTP - it's just that many of the popular clients use a relay so they don't have to handle error messages or hang out trying to deliver to slow servers or delay delivery on temporarily unavailable servers, and that many spammers abuse cheap disposable dialup accounts, but they get booted off of their ISPs' mail servers too fast to make them practical, or rate-limited, so they deliver their own email so they can reach more suckers before being squashed.
Some ISPs block outgoing Port 25 that doesn't go through their servers - really annoying if you've got more than one ISP account, and don't like having to reconfigure your machine just because you're dialing in from work or on the DSL at home instead of the other dialup.
Does anybody know what spamming technologies they used? I don't mean the actual content of what they're selling, but whether they were abusing open relays, or were they sending out spam that was easily traced back to their IP addresses, and were they sending it all out as "Monsterhut", or as dozens of different domain names?
If they're sending it out directly, without abusing relays, it's easy for ISPs to block their IP address space to avoid receiving spam from them.
Also, while Paetec was enjoined from cutting them off, Paetec's upstream providers, who also have AUPs that ban spamming, could still have done so - either by filtering the packets at the routers where they connect to it, or by advertising blackhole routes (or both - BGP is your friend...) That would cut off abuse of relays as well as direct-delivery spam.
If you're a typical American ISP customer, you're *not* paying any more to receive spam than to not receive spam - you're paying your ISP the same ~$20/month to get email that's really for you and to surf the web, and the spam is a free "bonus", like the advertising banners. Fax spam burns real paper and toner - email spam just puts useless pixels on your screen.
The real problem is that it's a bunch of obvious crap demanding your attention, and it's the attention that you and most of the other spam recipients care about (plus the bad taste of many of the advertised products, and the people who are offended by porn, especially porn spam sent to their kids.) ISPs, of course, are in a different situation, since they're dealing with it in volume.
The big advantage of snail mail spam is that, because there's a non-trivial cost associated with it, they don't send as much to people who don't want it, so you don't receive as much, and you don't waste much of your time trashing it. At some point I should tell the supermarkets to stop spamming my mailbox (it fills up the space, as well as wasting paper), but the only spam that really demands my attention is all the "pre-approved credit card offers" that might let somebody else get a credit card in my name if I simply throw them out instead of shredding them.
Todd P. Spammer claims that his targets have opted in and said "Yes I would accept offers that interest me" and that therefore his email is not spam, claimed it in court, and attempted to use it for financial gain, by continuing to use his PaeTec connection when in fact he knows he's violating the AUP, and by using it to spam people. But there are lots of affidavits out there from recipients of his spam, and presumably complaints about it have gone to him, so it's extremely difficult to claim ignorance even if he could.
This is different from the typical spammer who just lies to the recipient about "you must have opted in so we're sending you this junk offer" or "we'll remove you from the list we used today if you email us", because it's about specific facts, and it's also in court. It's still lying, of course, but sometimes lying becomes fraud and perjury as opposed to merely an attempt to gain attention or deflect complaints.
In most of the world, May 1 is Labor Day already, so there will already be world-wide protests. (For you Yankees, Labor Day outside the US is a day for Trade Unionists and Marxists and such to go out and march in parades and hold one-day general strikes. For you non-Yankees, Labor Day in the US is the day for end-of-summer picnics, cooking food on the barbeque, going to the beach, or alternatively deciding whether to go to Burning Man or Worldcon.)
So if the product's late, obviously the team is just on strike....
While I do occasionally trade frequent shopper cards with friends, whenever I apply for one, I'm John Doe, address General Delivery, my town, my zip code. I don't mind them collecting demographics that say that people living on one side of my town are more likely to buy tortillas and both sides to buy rice, or to decide that when they're promoting chicken whether to also promote charcoal and barbecue sauce or white wine and shallots. They don't need my name, street address, SSN, height/weight/eyecolor, iris prints, or finger prints to do that - and they're perfectly happy to give me frequent shopper cards.
If they insist on my fingerprints, I'm outta there.
I worked for NCR for a couple of years in the early 90s, and yes, a cash register these days is basically just a PC with different I/O devices - no point in making a specialized processor system when the general-purpose ones are cheaper and more powerful. The main alternatives were from IBM and ran funky IBM POS operating systems, but they were also becoming PCs.
We tended to run OS/2 on ours, since back in those days it was a major step up from DOS, better at networking, and could get a way with fewer resources than most Unix systems.
Might have been a good meeting room, though, or you could set up your Beowulf Cluster on a slightly larger Viking Dragon Boat than Mr. Beowulf himself used.:-)
The science parts of Ronald Reagan's Star Wars program weren't very good, but it was exceptionally successful fiction. There were some nice papers in the mid-80s about the targeting problem being NP-complete (ignored...), and a bunch of rigged demos (See the nice movie of the missile hitting the other missile with the transponder in it!) and the Post-Soviet attempts to retread SDI as a project to stop Rogue Nations (while it's more likely to stop a single ICBM than 5-10000 MIRVed warheads, it's really useless against truck bombs, and probably useless against cruise missiles or repainted commercial airliners. UA35 Heavy requests clearance for landing at SFO...)
It's too heavy - the things weigh 5 pounds or so. And the screen doesn't have enough pixels, though it's otherwise gorgeous. The battery life rocks - I don't know if I believe 5.5 hours in the real world, but it should scale pretty well with the 2-3 hour promises for many other laptops. So it's about long enough for an across-North-America airplane flight. Being able to connect the Firewire to the back and use it as a disk drive for your desktop Mac also rocks, and in traditional Apple fashion, there are a lot of things that just fit together nicely and make it a friendly environment.
I work in a corporate environment where people back in New Jersey who've never actually been on the road decide what laptops we're going to have out here in the field. So they pick machines that are powerful and have lots of features like built-in CDROMs and floppies and big screens and fast processors, which are all nice things to have for a machine that sits on your desk, but if you actually take them on the road, I want a machine that's lightweight and underpowered, and the only thing I'll spend more weight on is extra batteries. A few years ago, they were getting machines with the fanciest screens they could find, which meant fewer pixels and stunning 32-bit true color. Sorry - that's nice for Photoshop, but if you've gotta spend extra on screens, give me more pixels so I can read more text, and 8-bit color is fine. Dead-Tree Magazine Reviewers have similar tastes to the REMF\\\\ headquarters bureaucrats - their idea of a non-poweruser notebook is 5 pounds, just as their idea of a "budget home pc" is $999.
I don't need speed - the only things I burn CPU on are Microsoft Office and (in the past) The Great Internet Mersenne Prime Search. I need something I can carry on the train without breaking laptop-bag straps all the time, and without breaking my shoulder. The one good thing about overspec'd overweight devices is that since the folks back at headquarters are cheapskates and don't give us new toys every year, at least they don't get obsolete as fast:-) But I need something I can use on the train every day and haul on airplanes as long as that stays legal.
Over the years, we've used a variety of vendors - Toshiba, Dell, IBM, and back when we sold computers, OEM NEC machines with AT&T Death Star logos on the front. Unfortunately, we've always seemed to have been doing Toshibas when my machine was up for refresh - they've mostly been heavy unreliable pieces of junk that aren't made for the physical abuse that laptops get on the road, and they've got quirky power management that tends to have real trouble restoring from power-save mode without having to reboot half the time. Fortunately, my current-generation machine died, and the backup they've found me is one of the Portege7020s - the battery's ancient, so battery life is too short, but the machine's lighter and thinner than its faster predecessor:-)
OS-9 was a cute little real-time operating system for Motorola chips. Apparently they're still around, though Microware has been bought out by Radisys. They've ported it to half a dozen chips (mainly ARM/MIPS/etc but also x86) and it's got a lot more development environment with it than it had 15 years ago.
Avogadros' number lets you calculate, if you know the molecular weight of a substance, how many molecules of it are in a given weight of stuff. Standard homeopathic practice defines a specific amount of stuff that you disolve in a specific quantity of solvent (usually water or ethanol for liquids, or some neutral solid for pills), and each 1X dilution dilutes the solution by a factor of 10 - so a 2X dilution has 10**-2 times as many molecules as the reference quantity, and a 10X solution has 10**-10 times as many.
So if you know the reference quantity, and know the molecular weight of the medical substance, you can figure out how many molecules are in a drop of liquid medicine or a tablet. If the dilution starts to get pretty high, it's no longer simple division, because the number of molecules you have is an integer, so you need to apply probability and statistics to determine how many of those molecules were in the pill *you* got, and whether it's likely to be Zero or not. I forget the reference quantities, and obviously the divisibility is much different for a simple molecule like sodium chloride than for a complex plant mixture where each type of molecule might be a few thousand atoms, but that's just a scale factor - if something is 3X, your pill has a lot of stuff in it, if it's 20X, you might or might not have the interesting molecule in it, because that's about the limit, and if the preparation is 30X you're more likely to win the Powerball Lottery than to have a molecule of Foobaricus Magicus in *your* pill or dropperfull unless there are product contamination problems. "Clumping" doesn't really help - it increases the chances that some pills/drops will have a higher dose and others will have none at all.
Now, the effect of the concentration is still a discussable item, though the "more dilution makes the medicine stronger" is one of the quasi-religious tenets of homeopathy that make it a quack theory rather than science. Obviously if there are no molecules at all of your materia medica, it's not going to have any effect beyond the placebo effects or the effects of the dilutant. (Unless, of course, you actually believe, as one gushing enthusiast wrote, that the energy the molecules left behind after diluting them all away is what really causes the powerful effects. And she was talking about a 10X product, where there actually was still a trivial but non-zero concentration of material in it.) But within the more common ranges I've seen on products I've used, usually 2X-6X, there really is stuff there, at quantities you could do scientific testing on, or at least lots of trial&error if you're not the scientific sort, to determine what works best.
Homeopathy is a quack theory backed up by 200 years of trial and error experiments. There are things it's just not good at - being uninformed about the not-then-developed Germ Theory of Disease, it's not useful for treating bacterial diseases or most other diseases. And of course you shouldn't be reading most homeopathic literature with the assumption that it's scientific, any more than you should read Dianetics or Microsoft manuals or other bad science fiction that way. But it's not bad for treating allergies and similar conditions where you're trying to fix the symptoms, not the causes, or things that modern scientific medicine doesn't have much useful help for.
I don't know about you, but I get allergies, and moving from Eastern North America to Western North America merely changed which annoying plant byproducts I was exposed to. They're finally starting to make antihistamines that have more beneficial effects than annoying side effects (yay, Zyrtec!), but if your main choices are between taking drowsiness-causing antihistamines every 4-12 hours or wimpy not-very-useful ones (did Seldane ever actually work?) or mildly-digestive-upsetting homeopathic remedies every 1-2 hours, sometimes I'll go for the homeopathic and sometimes I'll go for the chemicals. Or you can go with Allergy shots, which are somewhere in between the homeopathic and allopathic flavors of medicine - for some people they really help.
For "flu", it's often helpful to get flu vaccinations, but sometimes they don't cover the flu you actually get, and Modern Scientific Medicine's advice usually runs to "Bummer, man, it's viral - stay in bed and drink chicken soup like your mom told you./i>" I've found that "Alpha CF" from Boericke & Tafel can often take me from feeling really really lousy to merely not feeling very good, which is enough to be worth driving to The Quackery Shop for ; other people like a preparation that probably isn't spelled Oscillococcinium. If it doesn't work for you, don't take it.
Meanwhile, the FDA's fundamental principle is that a bunch of guys with guns paid for by your tax dollars know more about medicine than you do, and that that therefore they should throw you in the clink for using or selling medicines that they haven't approved, just like their buddies at the DEA or Prohibition Agency can. Sorry, wrong answer - it's my body, and if I feel like buying Dr. Feelgood's Super Snake Oil, that's my business. Often, of course, they're right - there's a lot less bad medicine on the market because of them, and they've usually done a really excellent job of enforcing quality control on the big drug companies. But they ought to be acting like Consumer Reports or Underwriters' Laboratories, not like omniscient cops. And they've also done a huge amount of harm
years of delay on some of the good drugs, like calcium-channel beta blockers, have led to hundreds of thousands of unnecessary deaths
raising the cost of drug approval to the level that only really big companies can afford it has prevented many potential small companies from getting into the business, and prevented many big companies from developing drugs for small-market diseases
raising the cost has led to increased use of long-term global patent protection, reducing the use of generic drugs in the US and new drugs in poorer countries, and strongly encouraging the US medical profession to churn drug selection toward new expensive patent-protected products where older patent-expired products can work just fine.
FDA practices also limit the real scientific work done by the herbal medicine people, since scientifically acceptance-testing those products is expensive and patenting plant-based medicines is difficult. This pushes the whole herbal-medicine and alternative-medicine world more toward quackery, because the herbal players can't compete in the scientific-medicine market and aren't allowed to print real scientific testing results on their products even if they bother testing them.
You might be using RFC1918 space because you're using NAT, but there are other reasons and other ways to configure firewalls. The important reason is that you aren't getting your IP address space from your ISP, so you're doing the right thing rather than picking random numbers that belong to somebody else. You might be using a proxy firewall in a DMZ to fetch web pages and handle email instead of using NAT, and you can implement it relatively simply even without the proper router filters:-)
Of course, ISPs should be filtering out packets in RFC1918 space, and their DNSs should be managing the requests rather than bugging the root servers with them.
Slashdot Mirror
She did lose her Hotmail account a couple of years ago when they adopted policies to protect childrens' private information, because she was under 13 and her parent didn't want to provide a credit card to verify age for a Passport login.
You can often get legal costs covered if you win, which is one reason the EFF and similar groups can often afford to be helpful. A first-glance look by a non-lawyer says that either their patents are totally bogus for this application or they're covered by prior art. The first one I looked at mentioned being about "terminals" and a central processor, and that certainly shouldn't apply here. There are also some similarities to the BT Hyperlink patent nonsense. It's especially worth talking to the other targets of this abuse.
....then you need to start posting opinions they'd rather not own, such as comments about monopolies being bogusly abusive, questioning and/or declaring the ownership of copyrights on the content of the postings, self-contradictory statements, Top Ten Bad Things about Steve Case, etc....
Unfortunately, while that class of machines from a couple of manufacturers did have very reliable hardware, they tended to have highly proprietary funky operating systems. A friend of mine was trying to do air traffic control with one [machine to remain nameless, wasn't tandem] in ~86-87, and while the hardware was very reliable, the OS would crash weekly.
We also had a video projector, which was a $20000 ceiling-mounted thing. You were supposed to turn it off when you were done using it, but sometimes people forgot. One Monday morning I came in, and the room was 120 degrees, and there was a puddle of oil on the table under the projector - because somebody had forgotten to turn it off, we'd had a power hit over the weekend, the room had overheated, and the projectors spent the weekend blowing 130-degree ceiling-height air through itself trying to cool down. We had to get the building carpenters to make a crate for it so we could ship it back to Canada to get it repaired...
Looking at it from a decade later, I suppose it does have the advantage that if you're shipping stuff from New Jersey, where the shipping is controlled by Some Guys, you don't want to annoy Those Guys by saying "Hey, you ripped off my packages" when you could just file an insurance claim, assuming you bought their shipping insurance, which you do because "you wouldn' want anyt'ing bad ta happen ta dat package, now, wouldya?"...
By contrast, a friend of mine (hi, Doris!) was setting up an IBM-based shop a year or two later. One Saturday her machines arrived! The truckers set the ramp from the back of the truck onto the dock and rolled the two big drives onto it. While she was escorting one of the drivers into the building rolling the drive, the other guy backed the truck away, causing the other computer to fall three feet onto concrete. Ooops. She ended up talking to an IBM sales VP at home that day. He told her to accept the shipment and mark it as damaged, and they'd make good on it, so she did. They gave her the form, marked "damaged in shipment", which she crossed out and replaced with "dropped off loading dock"....
Redhat 7.1, which didn't fit into a single CD, was really annoying about this, but I didn't *want* to go building Debian or Gentoo from scratch on my antique-store lab machines, I wanted a consistent environment. Is SuSe any better? Or should I always pick up another $100 disk drive every time I need to upgrade the OS?
This 500KB of spam will take about 2-4 minutes of modem time to download it, which is annoying but won't break anybody's bank, and you'll be spending far more time on your modem reading slashdot - and this Slashdot page has a 9KB banner ad on it that you're downloading, which is probably typical for most web pages your read.
That's about 200MB of spam a year, which would cost $1 for disk drive if you actually stored it (20GB disk at $100) (and might mean that your ISP is paying $1/year extra for storing your share of their spam - they're not keeping it for long, unless they're keeping full backups on disk for long times, but they may have multiple copies in several tiers of backup and higher-performance disks that you have.)
The real cost to the average internet user isn't the resources consumed - it's how much of their attention span gets consumed shoveling the stuff out the door and into the bit bucket, and if you're a Responsible Netizen and occasionally hunt down spammers or at elast spamcop them, that takes time also.
HP3000s were Really Funky Mainframe-like Things. The Unix-based machines ran HPUX, which was almost exactly like Unix, but had really bizarre ideas about how RS-232 should be dealt with, and I spent far too much of my career for a couple years haggling with it and with drivers for HP printers.
On the other hand, remember when HP printers came with manuals that actually told you what the escape sequences were so you could do anything you wanted, not just 'how to tell Microsoft Products X/Y/Z that you have an HP printer?? That's because they were written by engineers for engineers, so you could actually understand what the equipment was doing and how to use it. Nobody writes manuals like that any more, unfortunately.
In other words, Taco does have an smtp server of his own to access. On his own machine. Like he should. If you're a dialup user, it's beneficial to have an inbound mailbox server somewhere that's always connected, whether it delivers the mail to you by SMTP or POP/IMAP. But no need to do that for outbound.
The reason the DUL is helpful for blocking spammers is not because there's no legitimate reason for a dialup user to run SMTP - it's just that many of the popular clients use a relay so they don't have to handle error messages or hang out trying to deliver to slow servers or delay delivery on temporarily unavailable servers, and that many spammers abuse cheap disposable dialup accounts, but they get booted off of their ISPs' mail servers too fast to make them practical, or rate-limited, so they deliver their own email so they can reach more suckers before being squashed.
Some ISPs block outgoing Port 25 that doesn't go through their servers - really annoying if you've got more than one ISP account, and don't like having to reconfigure your machine just because you're dialing in from work or on the DSL at home instead of the other dialup.
If they're sending it out directly, without abusing relays, it's easy for ISPs to block their IP address space to avoid receiving spam from them.
Also, while Paetec was enjoined from cutting them off, Paetec's upstream providers, who also have AUPs that ban spamming, could still have done so - either by filtering the packets at the routers where they connect to it, or by advertising blackhole routes (or both - BGP is your friend...) That would cut off abuse of relays as well as direct-delivery spam.
The real problem is that it's a bunch of obvious crap demanding your attention, and it's the attention that you and most of the other spam recipients care about (plus the bad taste of many of the advertised products, and the people who are offended by porn, especially porn spam sent to their kids.) ISPs, of course, are in a different situation, since they're dealing with it in volume.
The big advantage of snail mail spam is that, because there's a non-trivial cost associated with it, they don't send as much to people who don't want it, so you don't receive as much, and you don't waste much of your time trashing it. At some point I should tell the supermarkets to stop spamming my mailbox (it fills up the space, as well as wasting paper), but the only spam that really demands my attention is all the "pre-approved credit card offers" that might let somebody else get a credit card in my name if I simply throw them out instead of shredding them.
This is different from the typical spammer who just lies to the recipient about "you must have opted in so we're sending you this junk offer" or "we'll remove you from the list we used today if you email us", because it's about specific facts, and it's also in court. It's still lying, of course, but sometimes lying becomes fraud and perjury as opposed to merely an attempt to gain attention or deflect complaints.
So if the product's late, obviously the team is just on strike....
If they insist on my fingerprints, I'm outta there.
We tended to run OS/2 on ours, since back in those days it was a major step up from DOS, better at networking, and could get a way with fewer resources than most Unix systems.
Might have been a good meeting room, though, or you could set up your Beowulf Cluster on a slightly larger Viking Dragon Boat than Mr. Beowulf himself used.
The science parts of Ronald Reagan's Star Wars program weren't very good, but it was exceptionally successful fiction. There were some nice papers in the mid-80s about the targeting problem being NP-complete (ignored...), and a bunch of rigged demos (See the nice movie of the missile hitting the other missile with the transponder in it!) and the Post-Soviet attempts to retread SDI as a project to stop Rogue Nations (while it's more likely to stop a single ICBM than 5-10000 MIRVed warheads, it's really useless against truck bombs, and probably useless against cruise missiles or repainted commercial airliners. UA35 Heavy requests clearance for landing at SFO...)
It's too heavy - the things weigh 5 pounds or so. And the screen doesn't have enough pixels, though it's otherwise gorgeous. The battery life rocks - I don't know if I believe 5.5 hours in the real world, but it should scale pretty well with the 2-3 hour promises for many other laptops. So it's about long enough for an across-North-America airplane flight. Being able to connect the Firewire to the back and use it as a disk drive for your desktop Mac also rocks, and in traditional Apple fashion, there are a lot of things that just fit together nicely and make it a friendly environment.
I don't need speed - the only things I burn CPU on are Microsoft Office and (in the past) The Great Internet Mersenne Prime Search. I need something I can carry on the train without breaking laptop-bag straps all the time, and without breaking my shoulder. The one good thing about overspec'd overweight devices is that since the folks back at headquarters are cheapskates and don't give us new toys every year, at least they don't get obsolete as fast
Over the years, we've used a variety of vendors - Toshiba, Dell, IBM, and back when we sold computers, OEM NEC machines with AT&T Death Star logos on the front. Unfortunately, we've always seemed to have been doing Toshibas when my machine was up for refresh - they've mostly been heavy unreliable pieces of junk that aren't made for the physical abuse that laptops get on the road, and they've got quirky power management that tends to have real trouble restoring from power-save mode without having to reboot half the time. Fortunately, my current-generation machine died, and the backup they've found me is one of the Portege7020s - the battery's ancient, so battery life is too short, but the machine's lighter and thinner than its faster predecessor :-)
OS-9 was a cute little real-time operating system for Motorola chips. Apparently they're still around, though Microware has been bought out by Radisys. They've ported it to half a dozen chips (mainly ARM/MIPS/etc but also x86) and it's got a lot more development environment with it than it had 15 years ago.
So if you know the reference quantity, and know the molecular weight of the medical substance, you can figure out how many molecules are in a drop of liquid medicine or a tablet. If the dilution starts to get pretty high, it's no longer simple division, because the number of molecules you have is an integer, so you need to apply probability and statistics to determine how many of those molecules were in the pill *you* got, and whether it's likely to be Zero or not. I forget the reference quantities, and obviously the divisibility is much different for a simple molecule like sodium chloride than for a complex plant mixture where each type of molecule might be a few thousand atoms, but that's just a scale factor - if something is 3X, your pill has a lot of stuff in it, if it's 20X, you might or might not have the interesting molecule in it, because that's about the limit, and if the preparation is 30X you're more likely to win the Powerball Lottery than to have a molecule of Foobaricus Magicus in *your* pill or dropperfull unless there are product contamination problems. "Clumping" doesn't really help - it increases the chances that some pills/drops will have a higher dose and others will have none at all.
Now, the effect of the concentration is still a discussable item, though the "more dilution makes the medicine stronger" is one of the quasi-religious tenets of homeopathy that make it a quack theory rather than science. Obviously if there are no molecules at all of your materia medica, it's not going to have any effect beyond the placebo effects or the effects of the dilutant. (Unless, of course, you actually believe, as one gushing enthusiast wrote, that the energy the molecules left behind after diluting them all away is what really causes the powerful effects. And she was talking about a 10X product, where there actually was still a trivial but non-zero concentration of material in it.) But within the more common ranges I've seen on products I've used, usually 2X-6X, there really is stuff there, at quantities you could do scientific testing on, or at least lots of trial&error if you're not the scientific sort, to determine what works best.
I don't know about you, but I get allergies, and moving from Eastern North America to Western North America merely changed which annoying plant byproducts I was exposed to. They're finally starting to make antihistamines that have more beneficial effects than annoying side effects (yay, Zyrtec!), but if your main choices are between taking drowsiness-causing antihistamines every 4-12 hours or wimpy not-very-useful ones (did Seldane ever actually work?) or mildly-digestive-upsetting homeopathic remedies every 1-2 hours, sometimes I'll go for the homeopathic and sometimes I'll go for the chemicals. Or you can go with Allergy shots, which are somewhere in between the homeopathic and allopathic flavors of medicine - for some people they really help.
For "flu", it's often helpful to get flu vaccinations, but sometimes they don't cover the flu you actually get, and Modern Scientific Medicine's advice usually runs to "Bummer, man, it's viral - stay in bed and drink chicken soup like your mom told you./i>" I've found that "Alpha CF" from Boericke & Tafel can often take me from feeling really really lousy to merely not feeling very good, which is enough to be worth driving to The Quackery Shop for ; other people like a preparation that probably isn't spelled Oscillococcinium. If it doesn't work for you, don't take it.
Meanwhile, the FDA's fundamental principle is that a bunch of guys with guns paid for by your tax dollars know more about medicine than you do, and that that therefore they should throw you in the clink for using or selling medicines that they haven't approved, just like their buddies at the DEA or Prohibition Agency can. Sorry, wrong answer - it's my body, and if I feel like buying Dr. Feelgood's Super Snake Oil, that's my business. Often, of course, they're right - there's a lot less bad medicine on the market because of them, and they've usually done a really excellent job of enforcing quality control on the big drug companies. But they ought to be acting like Consumer Reports or Underwriters' Laboratories, not like omniscient cops. And they've also done a huge amount of harm
FDA practices also limit the real scientific work done by the herbal medicine people, since scientifically acceptance-testing those products is expensive and patenting plant-based medicines is difficult. This pushes the whole herbal-medicine and alternative-medicine world more toward quackery, because the herbal players can't compete in the scientific-medicine market and aren't allowed to print real scientific testing results on their products even if they bother testing them.
Of course, ISPs should be filtering out packets in RFC1918 space, and their DNSs should be managing the requests rather than bugging the root servers with them.