A D.J. shouldn't be going near iTMS; "Public Performance" is not one of the rights granted to the purchaser under plain copyright. Let alone the licensed-not-sold stuff from iTMS.
(I believe it is possible, in some jurisdictions, to purchase performance licenses that let you publicly perform works obtained through usual retail channels; in those cases; just over-gain the 6kHz-10kHz bands, and the resulting treble distortion should cover any problems introduced by the 128Kbps AAC coding.)
You've got it. In addition to the RIAA equalization on Phono, magnetic phono inputs are in the millivolt range for signal sensitivity (3 mV p-p?), whie all the "line" level inputs (CD, AUX, Tape, MP3, DVD, TV, and so on) are in the volt range (1V p-p). Line inputs do not have input equalization; just whatever you set the equalizer to. (Even your 2-band rotary eq, a/k/a "bass" and "treble" knobs.)
According to Wikipedia, I'm in the right range, but neither of us seem to know the exact number of millivolts.
There are "releasable" cable-ties out there; you squeeze a little tab and it releases the ratchet so you can pull the tie apart. I use them in exactly the kind of situation where something is likely to fail or need to be changed.
I couldn't tell from the photos if they had used that kind or not.
By a similar "corporate policy", IBM(r) PCs(r) don't have a reset button. They don't need it, the hardware is so good, it never crashes.
Even if you believe that, you still have to run software on that hardware.... Oh yeah, right, OS/2 is so good, it never crashes either.
...which means you're reaching around back to pull the power cable, because the software-controlled power supply won't turn off with the OS frozen. (The hold-power-for-4-seconds trick hadn't been put into the IBM(r) PC(r) back when I last had to deal with OS/2.)
Grandparent should be picking on the "obviousness" criteria. Lots of things are novel--the first time, of course--but that's not enough to be patentable. It has to be non-obvious to someone skilled in the art.
And I don't really see how Netflix is not obvious to someone skilled in the arts of mail-order subscriptions and rentals.
(I know that mail-order and rental has been combined before; the video store I worked at in the late 80s had a mail-order sales and rental business for customers in Northern Ontario. It was very expensive, but there wasn't anything else those people could use at the time. Some corporate libraries do a similar thing, using inter-office mail and the mainframe. So it's been a few years since I used one of those....)
If you need an encrypted connection, use an encrypted protocol; don't rely on the media to be secure. Doesn't everything know how to do TLS SMTP, IMAP, and POP3 these days? And https has been around for almost as long as http.
Or if you really need wire security, here's a thought: buy a service that provides wire security. Don't expect it from city-wide wireless, just as you shouldn't expect that sort of security from any radio communications.
Thing about that being in the fine print... if it is even phrased in a way that Joe Average will realize means "less ink than a brand new cartridge"....
It's in the fine print. People are going to buy the new printer anyway, and throw it out (or throw out their old one), because they didn't read the fine print.
If they really wanted to stop that, they need a BIG RED LABEL that says, "BARELY ANY INK INCLUDED".
Any fixed-matrix screen will give you similar issues; plasma screens are also a fixed matrix. Same applies to DLP and LCD projection. You can only avoid this with a CRT if the gun has adjustable focus; otherwise, you'd get vertical gaps between the scanlines. Of course, focus on a CRT is just an adjustment of a magnetic field, so that's easy.
The scalers (zoomers?) can vary widely. In fact, in comparing similarly-specced LCD computer screens, I found the biggest difference was in the quality of the scaling. Low-end screens just multiply the pixels, so one dot becomes 4, or whatever is needed. High-end screens run an anti-aliasing filter on top of that, to smooth it back out, similar to what you'd get from a CRT. Filter quality varies, too--and spending more money mean you get a better set. On (some?) DVI screens (with an nVidia card anyway) you can pick between several approaches, so you can just use less of the screen at low res--black border the picture rather than scale it. (I only just switched to DVI, so still got some playing around to do. But Myst 4 looks good running in 800x600, scaled to 1280x1024, on my NEC 90GX2.)
Same is true of the TVs. (And bad source material looks bad on anything big--the only way around that is to move farther back so the screen doesn't look as big.) Try the same NTSC feed--and a proper feed, not those co-ax distribution amp things I've seen in most stores--and check out the different sets you are considering.
I also tend to buy stuff I'm not sure about at shops with a "satisfaction gaurantee" policy... it's still annoying to have to go through the return, but it beats being out the money for a set you're not happy with.
For example, on a 64-bit machine, a 64-bit integer works quite nicely -- but of course, most people aren't (yet) using 64-bit machines.
You sure about that?
There are a lot of HP PA, Sun SPARC, and IBM POWER systems in banking and finance. All of those architectures have had 64-bit support for a long while now, even though the bulk of programs running on them still use 32-bit mode. The IBM big iron has got all sorts of toys for arbitrary precision arithmetic--they've been in banking and insurance for a while now, and it's all a solved problem on whatever the mainframes are called this month. (Still requires your programmers to use the tools available, which is always the hardest problem.)
On the desktops, it doesn't matter so much if you don't have native 64-bit support. As long as you can communicate sufficiently large numbers to the back-end system, and the safest way to do that is send the undecoded ASCII string. But even if you do convert to a 64-bit int on a 32-bit CPU, you're not doing heavy math with it on the desktop, so performance isn't a big deal.
I thought the AMD64 instruction set did offer extended 32-bit instructions. This allows 32-bit applications to access the improved register file and so on.
But it isn't compelling: You need a new operating system supervisor to handle it, because your context switch now has to save and restore the larger register file. To really get good use out of it, you would want to change your ABI calling convetion. This means introducing new magic numbers into the loader, so that IA32 programs can still run.
Providing support for 64-bit isn't all that much more, because not _everything_ becomes 64-bit like in RISC. You've still got IA32-style extension words making up instructions, so IA32-code is still the same. You've just got some new extension words and some more registers to play with, which you can do with their 32-bit (...l) or 64-bit (...q) instructions.
While true, perhaps a related problem that actually is a vulnerability is the fact that Firefox (apparently) only checks for a valid signature on the plugin at download/install time. Maybe the Firefox configuration file, or at the very least the binaries for each extension, should be cryptographically verified at runtime.
Once someone's system is compromised, they can replace or alter the FireFox binary which verifies the signatures, replace libnssckbi.so, libsoftokn3.so, whatever.
You can't win at that point. If you're storing your operating system and executables on writable media, it can never be trusted to that level. The hardware would have to cryptographically verify the boot loader on disk, which would verify the kernel, which would then be able to verify everything it executes--FireFox alone can't do it.
(Say, what was that hardware-based Trusted Computing stuff supposed to do? In addition to ramming DRM down everyone's PCI bus, wasn't there system verification too?)
I had a problem with Bell Canada's Sympatico service and a "cancellation fee". They had this promotion, where you got a discount on your cell phone if you signed up for one of these "bundles". But they didn't tell you there was a 2-year contract and a cancellation fee; and at the time, you couldn't see the details on their web site without Internet Explorer or Windows or Flash or something, I wasn't prepared to figure out what I needed--the page just came up blank.
So, when I got the paperwork in the mail (at least they sent paperwork in the mail), I read the terms, and saw the 2-year agreement. Since I wanted to change ISPs to one that didn't suck (static IP available, inbound port 25 available, no broken proxy cache on 80, &c.) in the near future, I didn't sign.
But they still tacked on the cancellation fee to my phone bill when I cancelled Sympatico. And calls to customer service did get them to admit they had no contract, and it was wrong... and it was still there the next month. And the next month.
So I bundled up all the paperwork, wrote a letter describing the situation, and a demand something be done. I sent the whole mess to all the non-billing addresses I could find for Bell (two, one in Ontario, one in Quebec), and the CRTC in Ottawa.
Very next month everything was fixed.
If you go this wrong, make sure you indicate on the letter to Telus that you are copying the CRTC, and indicate to the CRTC that they are receiving a copy of a letter you sent to Telus. The CRTC will not help if you have not taken steps to contact the telecommunications company first.
If you want your page to work the way The Fine Manual says those tags work, then you really do want the correct DTD at the top of the page. Otherwise, the browsers will drop into "compatibility" mode, and you'll be left wondering why there's a gap between two things that have "padding: 0; margin: 0; border: none", and various other tags work somewhat... oddly.
Of course, you can still use an HTML DTD, you don't have to use XHTML. Actually, for hand-coding, I don't want to use XHTML--I'm too used to the HTML rules to get XHTML right by hand without a lot of thinking about it....
But the DTD thing is easy to remember. You just copy the first few lines out of an existing document, or let Emacs HTML mode set it up for you. Failing that, it's in the HTML 4.01 reference.
Yeah, my TV has that feature, I used it all the time. Only thing is, when I got digital TV ("wireless cable" and then satellite), none of the tuner functions on the actual TV set matter any more, so it doesn't matter how clever they were. *sigh* The set has been left on VIDEO1 for like 8 years solid. Don't even know if I can find the remote anywhere....
Don't forget IBM's VM/SP (Virtual Machine/System Product) that started on the S/360 series of mainframes; most commonly found running systems like VM/CMS, VM/EMS, VM/XA (basically 3 generations of the same system), or MVS. And, of course, Linux nowadays.
It would (well, still does) virtualize the system hardware into logical partitions (LPARs), allowing allocation of whole units of hardware or factions of it to various virtual machines. Like you'd expect, these LPARs could then run whichever systems they wanted, reboot independently, and so on.
When I left IBM in 1999, all their Toronto Lab "mainframes" (6 or 8) were really just different LPARs on a single piece of hardware... at headquarters. They weren't even physically present at the Lab. (And air-cooled too, so no more going home early just because TOROLAB4 developed a leak in the main cooling pump.)
They've since ported the LPAR concept to the pSeries (former RS/6000), so you can run multiple configurations of AIX and/or Linux on a single machine, and move hardware resources from LPAR to LPAR as your load on the different VMs change.
They do something similar internally for their "Virtual Loaner Program". It looks like you're getting a whole machine to yourself--with root access and everything--but, of course, it's just an LPAR on the appropriate hardware, with the system you ordered (AIX or Linux) pre-loaded on it. (Which is really an AMAZING way to get more pSeries hardware if you're in a crunch because Critical Customer #1 just opened a deal-breaking ticket and you need to get development time on a slightly older version of AIX RIGHT AWAY... don't ask me how I know.)
I have added "rare" when it was an apt part of the description of the item, as in: "Rare WORKING SyQuest SyJet 1.5GB removable cartridge drive". (The "working" part is what made it rare, there were massive QC problems with SyQuest's last two drive models.)
OK, I really put "rare" on just to be silly.
I think the bidding went almost to $20--I just use eBay as that glorified garage sale a grandparent post mentioned.
Turn off SSID broadcast for "private but unencrypted". Tell the people you want using it the SSID, like by putting up a sign in your shop saying, "For customers of our coffee shop, you may use our Wi-Fi Hotspot while in our store, the SSID is 'blahblahblah'." Putting in some tips for entering an SSID on Windows XP and Mac OS X would be nice, too.
SSID broadcast on an unencrypted network is like the network saying, "Hey, I'm here, join me, I'm unencrypted!"
It sure doesn't happen much any more. But, in the good ol' days, back when there was still a significant bunch of electro-mechanical switches still running the phone system, it was reasonably likely that you'd pick up the handset and get that "phone's plugged in but no dialtone" sound. Or it would take several seconds to get dialtone. The best day to try this was Mother's Day; New Year's Day and Christmas were also fairly busy. I haven't had that happen since the mid-80s; even though I was connected to the Last Remaining Electro-Mechanical Switch in Ottawa in 1991. (Which;}} s gre}t for$the}}odem.}
The digital switches have made installing sufficient capacity very economical, so if you don't get a dialtone these days, you probably want to find out if someone's just crashed their pick-up truck into a telephone pole. (Happened to a friend, who called the phone company from his cell to find out what was wrong. They didn't know. We saw the line crews and the smashed pole on the way to the store, and figured it out for ourselves.) Or maybe someone's playing "Hungry Hungry Backhoes" with the fibre.
Trivia: The DMS-100 from Northern Telecom would switch dialtone-issue from FIFO (queue) to LIFO (stack) when things got busy. The reason? People get frustrated and hang up--and if you stick with FIFO, you wind up granting dialtone to someone who's in the process of putting the handset down in the cradle... and then you do it again, and again.... So, instead, they give dialtone to the person least likely to hang up and try again--someone who just picked up the phone.
In that particular case (dual operating switches, one taped down), you can. You can require that BOTH switches go OFF before the machine cycles again. So if a switch is taped down, it will cycle once and stop.
That would also allow the interlock to have a safer failure mode--if a switch jams ON, then the machine will not continuously cycle.
I'm a fan of safe failure modes. But I'm also a fan of letting lusers make the Darwin Awards each year.
First, you need to get every company to make sure their mail servers that relay to the outside world to:
Make sure their HELO hostname is resolvable by users outside the company.
Make sure their IP address has a PTR resolvable by users outside the company.
Until you get that fixed, you can't do a whole lot about header address verification.
Further, From: address verification has to be handled differently. From: addresses for my domain will ALWAYS appear to come DIRECTLY from a machine in the pobox.com domain. This is because I have pobox.com handling MX for my domain, and my outbound mail goes through their SASL AUTH gateway. (Otherwise, it gets (correctly) tagged as being "consumer broadband").
If, however, you look up the SPF records for my domain, it says "include=pobox.com -all", which means the SPF records for pobox.com apply (include=), and that this is authoritative (-all)--fail hard if any other machine claims to be mailing on behalf of my domain.
What I would like to see is a flag raised if envelope-from and header-from do not match. You'd want a way to "train" the mechanism for mailing lists you belong to. (Mailing lists should set envelope-from to the list-owner, and header-from to the author of the message (except digests).) That'd be a big way to catch phishers.
But all the good ways of catching phishers and spammers, as others have pointed out, also catch incompetent and ignorant corporate mailings.
What we need is people with more backbone to tell these companies that they're WRONG, customers cannot expect to get mail from them until they fix their server configuration (be it missing HELO host records, missing PTRs, made-up from addresses, and so on).
Big Companies (and I'm including Bell Canada and the Canadian telecommunications regulator in this one): Fix your servers, or your mail will be treated as spam and/or phishing attempts too.
Similarly, paypal.com's SPF records are non-authoritative--they say "don't error out on failure" (~all). What's the point of that? Get it right, make it authoritative--then at least people with SPF-checked mail will stop getting paypal phishes (or the envelope-from will not match header-from and the check for that will work well).
The torque curve of an electric motor varies greatly depending on its design. A series-excited DC machine, for example, produces maximum torque at 0 RPM, and is very handy for allowing your household blender to be able to crush ice and grind coffee, and your vacuum cleaner to make that awful shrieking noise. (Try it--take the fan off a Shop Vac(tm). It's still just as loud. Err, if you never get it back together and working again, it's not my fault.) But parallel-excited DC machines have different torque curves--they're not so impressive from a stand-still. And you can combine the two so that you get load-compensating motors, like used to be done in mixers, before solid-state speed controls worked.
Transit rail traction motors were often series-excited, for their great low-speed torque. And your streetcar really isn't going to be running any races.
But that's just DC machines. If you go to AC induction and synchronous machines, and add solid-state motor controls, you've got a world of possibility. Modern subways use synchronous AC motors and a combination of frequency- and pole-changing control circuitry so that the motors can be driven up from a very low speed at high torque, and then the number of poles is reduced to allow the motor to run up to higher speeds. The noise the motors make sounds like "electronic gears". Top speed is limited by the lowest number of poles and the highest frequency AC you can produce--3600 RPM from 2 poles at 60 Hz, 7200 RPM if you can get to 120 Hz, and so on. (Hmmm, what's the circumference of a car tire... say 16" tire, that's.4 m in diameter, so about 1.25 meters around. 100 km/h gives you 1666 m/minute, and divide by 1.25 m/revolution gives 1333 revolutions/miniute. So a two-pole AC motor driven at up to 45 Hz can get you to 200 km/h, without gearing. Heck, stick with a four-pole configuration and run up to 90 or 100 Hz.)
And AC synchronous motors produce BUCKETS of torque at high speed--unless you hit one of those suckers with a load heavy enough to cause it to stall. And you'll know if you get a load that high--the vibration from the motor reversing direction twice a cycle will be rather loud and hard on the drive train. Even induction motors have great torque at speed, though they slow down a bit from the "coasting" speed to produce it. And stalling an induction motor isn't as damaging, if you cut the power before it overheats.
In other words, "these ain't your grampa's electric motors."
Go ahead and ask 100 people on the street whether they use Windows and whether they know what an ACL is and how to change it.
It doesn't help that one of the features left out in XP Home Edition is the ACL Editor. Sure, it's obtuse and hard to figure out--but it's a damn sight simpler than trying to get anywhere with CACLS.EXE.
It would be one thing to leave out the ACL Editor (and the advanced features of the user editor, you know, like more than "Limited" and "Administrator" choices) from XP Home Edition if the underlying operating system didn't have those concepts, either. But, all the system features all there, you just really, really, really have to know how to drive the command line and/or hack the registry to do the work.
Frankly, I think it's fairly repellent that Microsoft wants a premium price for those two features--for an "advanced" home user, those are pretty much the only useful things in XP Professional. (Nearly everything else is relevant only for those working in Windows Server domains.)
Slashdot Mirror
A D.J. shouldn't be going near iTMS; "Public Performance" is not one of the rights granted to the purchaser under plain copyright. Let alone the licensed-not-sold stuff from iTMS.
(I believe it is possible, in some jurisdictions, to purchase performance licenses that let you publicly perform works obtained through usual retail channels; in those cases; just over-gain the 6kHz-10kHz bands, and the resulting treble distortion should cover any problems introduced by the 128Kbps AAC coding.)
You've got it. In addition to the RIAA equalization on Phono, magnetic phono inputs are in the millivolt range for signal sensitivity (3 mV p-p?), whie all the "line" level inputs (CD, AUX, Tape, MP3, DVD, TV, and so on) are in the volt range (1V p-p). Line inputs do not have input equalization; just whatever you set the equalizer to. (Even your 2-band rotary eq, a/k/a "bass" and "treble" knobs.)
According to Wikipedia, I'm in the right range, but neither of us seem to know the exact number of millivolts.
There are "releasable" cable-ties out there; you squeeze a little tab and it releases the ratchet so you can pull the tie apart. I use them in exactly the kind of situation where something is likely to fail or need to be changed.
I couldn't tell from the photos if they had used that kind or not.
By a similar "corporate policy", IBM(r) PCs(r) don't have a reset button. They don't need it, the hardware is so good, it never crashes.
Even if you believe that, you still have to run software on that hardware.... Oh yeah, right, OS/2 is so good, it never crashes either.
Grandparent should be picking on the "obviousness" criteria. Lots of things are novel--the first time, of course--but that's not enough to be patentable. It has to be non-obvious to someone skilled in the art.
And I don't really see how Netflix is not obvious to someone skilled in the arts of mail-order subscriptions and rentals.
(I know that mail-order and rental has been combined before; the video store I worked at in the late 80s had a mail-order sales and rental business for customers in Northern Ontario. It was very expensive, but there wasn't anything else those people could use at the time. Some corporate libraries do a similar thing, using inter-office mail and the mainframe. So it's been a few years since I used one of those....)
Ummmm.
If you need an encrypted connection, use an encrypted protocol; don't rely on the media to be secure. Doesn't everything know how to do TLS SMTP, IMAP, and POP3 these days? And https has been around for almost as long as http.
Or if you really need wire security, here's a thought: buy a service that provides wire security. Don't expect it from city-wide wireless, just as you shouldn't expect that sort of security from any radio communications.
Thing about that being in the fine print... if it is even phrased in a way that Joe Average will realize means "less ink than a brand new cartridge"....
It's in the fine print. People are going to buy the new printer anyway, and throw it out (or throw out their old one), because they didn't read the fine print.
If they really wanted to stop that, they need a BIG RED LABEL that says, "BARELY ANY INK INCLUDED".
Any fixed-matrix screen will give you similar issues; plasma screens are also a fixed matrix. Same applies to DLP and LCD projection. You can only avoid this with a CRT if the gun has adjustable focus; otherwise, you'd get vertical gaps between the scanlines. Of course, focus on a CRT is just an adjustment of a magnetic field, so that's easy.
The scalers (zoomers?) can vary widely. In fact, in comparing similarly-specced LCD computer screens, I found the biggest difference was in the quality of the scaling. Low-end screens just multiply the pixels, so one dot becomes 4, or whatever is needed. High-end screens run an anti-aliasing filter on top of that, to smooth it back out, similar to what you'd get from a CRT. Filter quality varies, too--and spending more money mean you get a better set. On (some?) DVI screens (with an nVidia card anyway) you can pick between several approaches, so you can just use less of the screen at low res--black border the picture rather than scale it. (I only just switched to DVI, so still got some playing around to do. But Myst 4 looks good running in 800x600, scaled to 1280x1024, on my NEC 90GX2.)
Same is true of the TVs. (And bad source material looks bad on anything big--the only way around that is to move farther back so the screen doesn't look as big.) Try the same NTSC feed--and a proper feed, not those co-ax distribution amp things I've seen in most stores--and check out the different sets you are considering.
I also tend to buy stuff I'm not sure about at shops with a "satisfaction gaurantee" policy... it's still annoying to have to go through the return, but it beats being out the money for a set you're not happy with.
You sure about that?
There are a lot of HP PA, Sun SPARC, and IBM POWER systems in banking and finance. All of those architectures have had 64-bit support for a long while now, even though the bulk of programs running on them still use 32-bit mode. The IBM big iron has got all sorts of toys for arbitrary precision arithmetic--they've been in banking and insurance for a while now, and it's all a solved problem on whatever the mainframes are called this month. (Still requires your programmers to use the tools available, which is always the hardest problem.)
On the desktops, it doesn't matter so much if you don't have native 64-bit support. As long as you can communicate sufficiently large numbers to the back-end system, and the safest way to do that is send the undecoded ASCII string. But even if you do convert to a 64-bit int on a 32-bit CPU, you're not doing heavy math with it on the desktop, so performance isn't a big deal.
I thought the AMD64 instruction set did offer extended 32-bit instructions. This allows 32-bit applications to access the improved register file and so on.
But it isn't compelling: You need a new operating system supervisor to handle it, because your context switch now has to save and restore the larger register file. To really get good use out of it, you would want to change your ABI calling convetion. This means introducing new magic numbers into the loader, so that IA32 programs can still run.
Providing support for 64-bit isn't all that much more, because not _everything_ becomes 64-bit like in RISC. You've still got IA32-style extension words making up instructions, so IA32-code is still the same. You've just got some new extension words and some more registers to play with, which you can do with their 32-bit (...l) or 64-bit (...q) instructions.
If we're going to do that, the study carrels should have a big mirror up top too, so that people still using dead trees are given the same exposure.
Once someone's system is compromised, they can replace or alter the FireFox binary which verifies the signatures, replace libnssckbi.so, libsoftokn3.so, whatever.
You can't win at that point. If you're storing your operating system and executables on writable media, it can never be trusted to that level. The hardware would have to cryptographically verify the boot loader on disk, which would verify the kernel, which would then be able to verify everything it executes--FireFox alone can't do it.
(Say, what was that hardware-based Trusted Computing stuff supposed to do? In addition to ramming DRM down everyone's PCI bus, wasn't there system verification too?)
I had a problem with Bell Canada's Sympatico service and a "cancellation fee". They had this promotion, where you got a discount on your cell phone if you signed up for one of these "bundles". But they didn't tell you there was a 2-year contract and a cancellation fee; and at the time, you couldn't see the details on their web site without Internet Explorer or Windows or Flash or something, I wasn't prepared to figure out what I needed--the page just came up blank.
So, when I got the paperwork in the mail (at least they sent paperwork in the mail), I read the terms, and saw the 2-year agreement. Since I wanted to change ISPs to one that didn't suck (static IP available, inbound port 25 available, no broken proxy cache on 80, &c.) in the near future, I didn't sign.
But they still tacked on the cancellation fee to my phone bill when I cancelled Sympatico. And calls to customer service did get them to admit they had no contract, and it was wrong... and it was still there the next month. And the next month.
So I bundled up all the paperwork, wrote a letter describing the situation, and a demand something be done. I sent the whole mess to all the non-billing addresses I could find for Bell (two, one in Ontario, one in Quebec), and the CRTC in Ottawa.
Very next month everything was fixed.
If you go this wrong, make sure you indicate on the letter to Telus that you are copying the CRTC, and indicate to the CRTC that they are receiving a copy of a letter you sent to Telus. The CRTC will not help if you have not taken steps to contact the telecommunications company first.
If you want your page to work the way The Fine Manual says those tags work, then you really do want the correct DTD at the top of the page. Otherwise, the browsers will drop into "compatibility" mode, and you'll be left wondering why there's a gap between two things that have "padding: 0; margin: 0; border: none", and various other tags work somewhat... oddly.
Of course, you can still use an HTML DTD, you don't have to use XHTML. Actually, for hand-coding, I don't want to use XHTML--I'm too used to the HTML rules to get XHTML right by hand without a lot of thinking about it....
But the DTD thing is easy to remember. You just copy the first few lines out of an existing document, or let Emacs HTML mode set it up for you. Failing that, it's in the HTML 4.01 reference.
That's a waste of time. Burn it with iTunes and rip it back with iTunes--then all the ID3 info is copied over, too. (Not the artwork, though.)
Yeah, my TV has that feature, I used it all the time. Only thing is, when I got digital TV ("wireless cable" and then satellite), none of the tuner functions on the actual TV set matter any more, so it doesn't matter how clever they were. *sigh* The set has been left on VIDEO1 for like 8 years solid. Don't even know if I can find the remote anywhere....
Don't forget IBM's VM/SP (Virtual Machine/System Product) that started on the S/360 series of mainframes; most commonly found running systems like VM/CMS, VM/EMS, VM/XA (basically 3 generations of the same system), or MVS. And, of course, Linux nowadays.
It would (well, still does) virtualize the system hardware into logical partitions (LPARs), allowing allocation of whole units of hardware or factions of it to various virtual machines. Like you'd expect, these LPARs could then run whichever systems they wanted, reboot independently, and so on.
When I left IBM in 1999, all their Toronto Lab "mainframes" (6 or 8) were really just different LPARs on a single piece of hardware... at headquarters. They weren't even physically present at the Lab. (And air-cooled too, so no more going home early just because TOROLAB4 developed a leak in the main cooling pump.)
They've since ported the LPAR concept to the pSeries (former RS/6000), so you can run multiple configurations of AIX and/or Linux on a single machine, and move hardware resources from LPAR to LPAR as your load on the different VMs change.
They do something similar internally for their "Virtual Loaner Program". It looks like you're getting a whole machine to yourself--with root access and everything--but, of course, it's just an LPAR on the appropriate hardware, with the system you ordered (AIX or Linux) pre-loaded on it. (Which is really an AMAZING way to get more pSeries hardware if you're in a crunch because Critical Customer #1 just opened a deal-breaking ticket and you need to get development time on a slightly older version of AIX RIGHT AWAY... don't ask me how I know.)
I have added "rare" when it was an apt part of the description of the item, as in: "Rare WORKING SyQuest SyJet 1.5GB removable cartridge drive". (The "working" part is what made it rare, there were massive QC problems with SyQuest's last two drive models.)
OK, I really put "rare" on just to be silly.
I think the bidding went almost to $20--I just use eBay as that glorified garage sale a grandparent post mentioned.
Turn off SSID broadcast for "private but unencrypted". Tell the people you want using it the SSID, like by putting up a sign in your shop saying, "For customers of our coffee shop, you may use our Wi-Fi Hotspot while in our store, the SSID is 'blahblahblah'." Putting in some tips for entering an SSID on Windows XP and Mac OS X would be nice, too.
SSID broadcast on an unencrypted network is like the network saying, "Hey, I'm here, join me, I'm unencrypted!"
It sure doesn't happen much any more. But, in the good ol' days, back when there was still a significant bunch of electro-mechanical switches still running the phone system, it was reasonably likely that you'd pick up the handset and get that "phone's plugged in but no dialtone" sound. Or it would take several seconds to get dialtone. The best day to try this was Mother's Day; New Year's Day and Christmas were also fairly busy. I haven't had that happen since the mid-80s; even though I was connected to the Last Remaining Electro-Mechanical Switch in Ottawa in 1991. (Which;}} s gre}t for$the}}odem.}
The digital switches have made installing sufficient capacity very economical, so if you don't get a dialtone these days, you probably want to find out if someone's just crashed their pick-up truck into a telephone pole. (Happened to a friend, who called the phone company from his cell to find out what was wrong. They didn't know. We saw the line crews and the smashed pole on the way to the store, and figured it out for ourselves.) Or maybe someone's playing "Hungry Hungry Backhoes" with the fibre.
Trivia: The DMS-100 from Northern Telecom would switch dialtone-issue from FIFO (queue) to LIFO (stack) when things got busy. The reason? People get frustrated and hang up--and if you stick with FIFO, you wind up granting dialtone to someone who's in the process of putting the handset down in the cradle... and then you do it again, and again.... So, instead, they give dialtone to the person least likely to hang up and try again--someone who just picked up the phone.
That would also allow the interlock to have a safer failure mode--if a switch jams ON, then the machine will not continuously cycle.
I'm a fan of safe failure modes. But I'm also a fan of letting lusers make the Darwin Awards each year.
Trailers, and other promotional videos, have been distributed like that for a while. I've seen downloads of commercials that prohibited copying....
I think some of these people are so keen on control that they forget what advertising is for.
Until you get that fixed, you can't do a whole lot about header address verification.
Further, From: address verification has to be handled differently. From: addresses for my domain will ALWAYS appear to come DIRECTLY from a machine in the pobox.com domain. This is because I have pobox.com handling MX for my domain, and my outbound mail goes through their SASL AUTH gateway. (Otherwise, it gets (correctly) tagged as being "consumer broadband").
If, however, you look up the SPF records for my domain, it says "include=pobox.com -all", which means the SPF records for pobox.com apply (include=), and that this is authoritative (-all)--fail hard if any other machine claims to be mailing on behalf of my domain.
What I would like to see is a flag raised if envelope-from and header-from do not match. You'd want a way to "train" the mechanism for mailing lists you belong to. (Mailing lists should set envelope-from to the list-owner, and header-from to the author of the message (except digests).) That'd be a big way to catch phishers.
But all the good ways of catching phishers and spammers, as others have pointed out, also catch incompetent and ignorant corporate mailings.
What we need is people with more backbone to tell these companies that they're WRONG, customers cannot expect to get mail from them until they fix their server configuration (be it missing HELO host records, missing PTRs, made-up from addresses, and so on).
Big Companies (and I'm including Bell Canada and the Canadian telecommunications regulator in this one): Fix your servers, or your mail will be treated as spam and/or phishing attempts too.
Similarly, paypal.com's SPF records are non-authoritative--they say "don't error out on failure" (~all). What's the point of that? Get it right, make it authoritative--then at least people with SPF-checked mail will stop getting paypal phishes (or the envelope-from will not match header-from and the check for that will work well).
Transit rail traction motors were often series-excited, for their great low-speed torque. And your streetcar really isn't going to be running any races.
But that's just DC machines. If you go to AC induction and synchronous machines, and add solid-state motor controls, you've got a world of possibility. Modern subways use synchronous AC motors and a combination of frequency- and pole-changing control circuitry so that the motors can be driven up from a very low speed at high torque, and then the number of poles is reduced to allow the motor to run up to higher speeds. The noise the motors make sounds like "electronic gears". Top speed is limited by the lowest number of poles and the highest frequency AC you can produce--3600 RPM from 2 poles at 60 Hz, 7200 RPM if you can get to 120 Hz, and so on. (Hmmm, what's the circumference of a car tire... say 16" tire, that's .4 m in diameter, so about 1.25 meters around. 100 km/h gives you 1666 m/minute, and divide by 1.25 m/revolution gives 1333 revolutions/miniute. So a two-pole AC motor driven at up to 45 Hz can get you to 200 km/h, without gearing. Heck, stick with a four-pole configuration and run up to 90 or 100 Hz.)
And AC synchronous motors produce BUCKETS of torque at high speed--unless you hit one of those suckers with a load heavy enough to cause it to stall. And you'll know if you get a load that high--the vibration from the motor reversing direction twice a cycle will be rather loud and hard on the drive train. Even induction motors have great torque at speed, though they slow down a bit from the "coasting" speed to produce it. And stalling an induction motor isn't as damaging, if you cut the power before it overheats.
In other words, "these ain't your grampa's electric motors."
It doesn't help that one of the features left out in XP Home Edition is the ACL Editor. Sure, it's obtuse and hard to figure out--but it's a damn sight simpler than trying to get anywhere with CACLS.EXE.
It would be one thing to leave out the ACL Editor (and the advanced features of the user editor, you know, like more than "Limited" and "Administrator" choices) from XP Home Edition if the underlying operating system didn't have those concepts, either. But, all the system features all there, you just really, really, really have to know how to drive the command line and/or hack the registry to do the work.
Frankly, I think it's fairly repellent that Microsoft wants a premium price for those two features--for an "advanced" home user, those are pretty much the only useful things in XP Professional. (Nearly everything else is relevant only for those working in Windows Server domains.)