I'm the cynical UNIX bastard, and have been for 15 years now. I started my current job about 10 years ago, on the strength of my UNIX scripting, trouble-shooting, and ad-hoc admin skills. And being able to run the most effective test lab in the previous company with hardly any budget. (A LOT of scrap machines from other departments can be very useful... if you can distribute the workload.)
Anyway. So the new company wants to try out ClearCase. And try it properly: deployment, usage, disaster recovery, everything. So the guy in charge of the IT backup systems says, "We've got the backups working just fine!"
And I ask, "How about the restore?"
And everyone laughs. Except me. And my manager looks... and thinks... and says, "OK, so what about the restore?"
I think the real failing of UAC is that there are no cues you're about to cross a privilege threshold.
That, and since the default is a click-to-continue model, it doesn't really LOOK like it's changing privilege. (Whereas 'sudo'-based privilege changes always ask for the user's password. OS X uses 'sudo' under the hood to do its switching.)
There's nothing about the "Advanced Settings" in the system control panel that would let me think it's going to change privilege when I click on it--because I want to set some environment variables for my account.
(BTW, I've only seen UAC on Server 2008, which is mainly Vista in server drag, but I'm not sure how close.)
So, first of all, all the control panels need to be separated so "Admin" function is separated from "User" function. I should not have to become admin to change _my_ environment, I should only have to do that to change the _System_ environment.
Then there should be visual cues, like closed-or-open-padlock icons, around the stuff that will trigger a privilege elevation.
And there absolutely SHOULD NOT BE "installer detection" resulting in automatic prompting for elevation. I've used--and written--installers that work just fine in a non-Admin account. Especially in Server 2008, teach the admins to "Run as Administrator" if they want. (Modern installers should be able to ask for elevation, so we're talking about "legacy" installers.)
Oh, let's make the default ask-for-password rather than click-to-continue, too?
Oh yeah: Let's fix what happens when you turn UAC off. It doesn't turn off prompting, it actually removes the privilege separation. How about just have it turn off prompting? Sure, that's next to useless on a machine connected to the Big Bad Internet, but it's still better than the "You're in the Administrator group, so everything you do you do as Administrator" behaviour from older versions. So the settings can be, "No UAC (always Administrator); Unprompted Elevation; Prompted Elevation; Ask for password."
I love it when Amazon runs out of data mining steam:
As someone who has purchased books in the past, you might be interested to know that Amazon has several books on sale for this upcoming Mother's Day.
Usually there's some attempt to get things right, like "you bought a movie in Chinese, so anything else in Chinese must be interesting!" Or, "You've only ever bought PlayStation 2, 3 and Nintendo Wii games, so you must be interested in all our new Xbox 360 titles!"
The HL5250-DN has a duplexer, which is nice. But it's got a sharp pointy-bit in just the wrong place so that if you print on 3-hole paper it's just paper jam city. If you take the machine apart, it's easy enough to file smooth so it works fine. But by the time you figure out how to do that, it will be making weird clunky noises for the rest of its life.
Other than that, BR-Script 3 works fine, sits on the wired LAN nicely, and plays well with both the Linux and Mac OS X children.
They claim separate imaging and toner cartridges save money, but I'm not convinced. It just makes it more difficult when the drum needs to be replaced, because most places just have the toner. (But get the extended life carts. Lots of happy DB2 manual printing without worries.)
There's a local catalogue store that has a neat system. They've got paper catalogues and merchandise on display, like most such stores, and you fill out a paper form and bring it to an order taker to get the stuff you want.
But you can also use an in-store computer terminal and use your member ID to get a ready-to-scan completed order form, plus it's already done the inventory verification step.
To protect your personal info, you must stand on a rubber pressure mat like they used to have for opening grocery store doors. When you step off the mat, it wipes the session and brings the machine back to the "Please stand on mat to start a session" logo screen.
You still have to trust that they actually work, but you're handing your membership info over to the store if you want to buy something anyway....
I could see something similar triggering a screen locker instead. Get up off your desk chair for more than 5 seconds and the screen locks.
Any resource allocated is something that needs to be tracked. Those tracking objects take a little bit of management time. What kind of time and when depends on the object: sleeping processes aren't checked by the scheduler, but other process operations like signal need to.
Directory entries add up. Even if you have a hash or tree directory index structure, you may still have to scan the whole thing to find files by pattern (for example). With many directory indexes, you must do an end-to-end scan to find out if a name is truly unique.
And almost ALL of the "cruft" that is added to new systems is stuff that "does things" automatically. In order to do that, they all wake up on certain events (including timer events) and check to see if they need to do something. Like multiple devices sharing an interrupt, it all adds up.
Sure, a new system has enough power that it retains good interactive response with all that. But run an old OS on a new system, and wow is it fast.
DVD region codes are orthogonal to DVD copy protection.
You can have a CSS-protected disk which plays in all regions.
You can have an unprotected disk with region codes.
The only way they become commingled is via contracts with the DVD CCA. To get your CSS decryption key, you must agree to obey region codes and user operation prohibitions. (And as the parent says, getting the patent licenses is tied in there, too.)
The DVDs I bought from Japan and have to watch via a region-free player in Canada honour the copyright holders rights completely. And the retailers: the mail order store in Japan got their part of the transaction, as did the Japanese and Canadian postal services. It's not like the DVD was cheaper to import than it would be had it been available in Canada in the first place.
Us movie viewers are not a party to the ridiculous territory contracts and exclusivity deals the studios have with various regional distribution houses. So we can buy something retail and import it, no problems. We didn't make a copy, so copyright doesn't apply. We--and the selling retailer--aren't a party to the territory contracts, so we're not violating those contracts.
The only laws which might apply are the obscenity ones.
You missed a step in the evolution, that actually I think helps your point.
First manual chokes became integrated with another control. If it was cold, you needed to fully depress AND RELEASE the gas pedal BEFORE turning the ignition key to "START". That would engage the choke, which now had no indication to the driver that such a device existed in the vehicle and no indication of its current status. (There were variations on this, too; like press and release twice. Borrowing a friend's car in winter needed a 10 minute explanation of the gas pedal.)
On some vehicles, release was automatic, based on engine RPM (sensed via vacuum). On others, you had to "tap" the gas pedal to clear the choke function. If you didn't, the car could actually over-rev (to redline), and there wasn't a limiter to protect the engine.
All in all, the worst combination of "hidden controls" and "necessary controls".
Early EFI, on the other hand, came with huge warnings "DO NOT DEPRESS GAS PEDAL WHEN STARTING" to try and break people of habits introduced by carburettors and all the kludges needed to get them working.
You can have my modern EFI engine if you can pry it from my cold, dead hands. Every time I do the choke-and-throttle dance on my motorcycle, I curse everyone involved in inventing it.
They enabled "disconnect on hangup" on your line. If you have a burglar alarm installed that uses your phone line, the alarm company will arrange for the same feature.
Traditionally, POTS lines aren't disconnected until both sides go on-hook. With disconnect on hangup, the line is disconnected when one side goes on-hook, though it may take up to 10 seconds.
I dunno; I've been in two houses where the floor slab has been ripped open and partially re-poured.
My parent's house, when we put a basement bathroom in, needed the slab ripped out around the drain lines for the new drain connections.
My current house has new slab in the front 1/3d of the basement, where the furnace and hot water tank are. I don't know what was wrong with the old concrete, but in the rest of the basement, you've got to be careful with the broom or you sweep up concrete instead of just dust and dirt.
It's only the slab, not the foundation (footings and wall). Breaking the footings open, that's bad. Undermining the footings, that's bad.
Floor slab? This way to the drains....
Besides, if the pipes fail, cut the ends flush with the slab, plug with mortar, and get on with your life. That's what they did with the old natural gas supply pipe; it's cut and plugged on both the inside and outside, where it used to go to the city main.
Come to think of it, they did the same thing with the oil tank fill and breather pipes at my parents when we had the oil furnace removed. (Very important to plug the filler side, especially if the oil company doesn't get the cancel notice to the delivery truck....)
Interestingly, a LOT of people, let's say almost everyone, get Make wrong because they don't understand how the declarative rule structure works.
I'm not a fan of ANT, actually; I think it exists to solve a problem that Sun should have never created in the first place. (Produce arbitrarily named.class files from a given.java file.) I don't know how declarative it _really_ is under the covers, but everyone I see saying "it's better than make" are doing procedural things with it and using that to justify ANT over make.
(My gripe with ANT--aside from the XML soup--is it affords sloppy build systems: it's a wildcard enabler.)
Same in Canada; PayPal is the best choice if a border is involved.
Postal money orders between Canada and the U.S. are reasonably OK to get, though expensive on the purchaser's side ($3.50-$8.00 depending on your bank or post office). Problem is, you don't know when you receive a money order if it's really going to stay "good"; there's a lot of bogus ones.
I've never been able to find two people who understand what an International Postal Order is. These are supposed to be able to transfer funds in a large number of currencies between a large number of countries. But you have to find a post office on each side with someone working there who knows how to do it.
Wire transfer is OBSCENELY expensive. Like starting at $40 to send $20 to Australia. And we're both Commonwealth Realms!
I can negotiate a U.S. Funds cheque with my Canadian bank account... for $25. "With $20 USD Mail-In Rebate" means "you pay $5 for this rebate ha ha". Most U.S. banks can't take a cheque (er, check) in Canadian funds.
I can endorse my cheque "U.S. FUNDS" and then the U.S. banks... still probably won't take it, but they should. For some fairly extreme fees on my side.
I can open a U.S. Dollar account and write cheques that are processed through a U.S. Bank... for a suitable monthly fee.
(All the cheque stuff, of course, has even more trust and fraud issues than postal money orders, be they international or "named currency".)
Interac e-mail money transfer only works for Interac users, which is everyone with an ATM card in Canada, but that's kind of limited to Canada....
It came down to, "I'll go to the currency exchange and buy AUD$20 in cash and mail it to you" at one point. That or PayPal were the practical choices for sub-multi-thousand cash transfers. And with PayPal, neither of us had to get off our fat lazy butts.
For anyone with enough volume, of course, get a real merchant account. But for otherwise private individuals, especially ones who "know" each other through a separate channel? PayPal is the best choice... as annoying as that as.
There's a good argument for having page zero stuffed with zeros, execute-prohibited and write-prohibited but allow reads. It's got to do with pipelines and test operations.
test cr0,r3 load r4,(r3) bz.L1,cr0 ...do something with r4... .L1: ...and so on
(Any relation to some weird cross combination of MC68K, POWER, PowerPC, and SPARC assembly is entirely likely.)
So, on some systems, it's worth starting the load through a null pointer before the test of the pointer has completed. AIX on POWER and PowerPC, for example, guarantees that Page Zero is mapped read-only and contains only zeros (which is also an illegal instruction). This way, the pipeline stalls are minimized; if you didn't have anything else to do in that subroutine, you'd have an effective nop after the test, like so:
test cr0,r3 nop bz.L1,cr0 load r4,(r3) nop ...do something with r4 .L1: ...and so on
This applies to READ operations only, of course; WRITE operations must verify the pointer, as a write attempt to page zero will fault.
So, Linux: Why is page zero modifiable from userland? A program specifying a load address of zero MUST be rejected.
The thing about UNIVERSAL public health care is...
It's actually not very important if you correctly identify who is getting health care.
Your doctor needs to know who you are, and lab results need to be correctly tied to samples, and so on. But that's not a function of who's paying the bills.
But for determining if the doctors and labs get paid? Not so much.
Basically, all you really need to know is, "is this person a citizen or lawful immigrant?" and "is this procedure covered by the system?". It's not so important to know WHICH citizen or lawful immigrant. It's nice to get it right, but your medical history doesn't need to be part of your public health insurance ID, so it's not critical to treatment.
Different keying problem.
That being said... I'm amazed at how many people think there's some huge government conspiracy out to get them when they can't get simple stuff like this right. Sure, they can listen in on all cell phone calls... but they can't keep a list properly?
Is it not possible to calculate the thermal heat in 1L of gasoline when burned at STP at the stoichiometric ratio? If we can get BTU or BTUh out of it, we've got something to work with.
Then compare that to the kWh at your power meter to charge the vehicle enough to travel until the battery system shuts down. (Disable the gasoline system on Volt-like cars.) Count miles. Repeat, average, etc.
I was able to find the right figures for natural gas, so I was able to figure out if it was better to use a single-room space heater or turn up the furnace. (Unsurprising result: A little bit of electricity for one room is better than a lot of gas for the whole house.)
They're also throwing around figures like "40 cents to charge, for 10 kWh, at Michigan off-peak rates." OK, sure, but in Ontario, that'll set you back over a Canadian dollar, as almost nobody has time-of-day metering. Massachusetts will be closer to $2 USD (20 cents/kWh)--which is still half the price of gasoline at last summer's prices in Ontario ($1.25/L).
Here we go. 125,000 BTU in 1 gallon of gasoline, which is about 37 kWh.
So, at 40 miles/10 kWh we've got 4 miles/kWh, which I didn't need Google for but so you can see what I'm doing (show your work). That's the easy one.
I've seen 50 miles/gallon cited for the Volt, so we want miles/kWh... 1.36 miles/kWh.
Both of those are "at the pump/plug" numbers: What you use in electricity net of any generation and transmission losses, compared to volume of gasoline from the pump at your filling station net of energy used to process it from the Alberta tar sands.
(My physics teacher would freak out at that SI and US Imperial unit soup, too.)
What I want to know about all of these electric things though... especially if they're quoting Michigan off-peak power prices... what happens in winter? Those of us in northerly climates don't just throw away all of the thermal inefficiency in the internal combustion motor. We vent some of it in to the passenger cabin as "heat". I'm not giving that up; and resistive electric head for the passenger cabin will kill your battery range real fast--everyone's left just the headlights on and needed a boost at one point, right? That's only about 180 Watts (two headlights + assorted markers.)
The problem with patenting a word processor file is, it's not their words.
There's a patent on my kitchen faucet, but the water that comes out of it isn't affected by that.
The way the USPTO reads software patents, also, makes things weird. With a physical object, if you achieve the same result by a different method, it is non-infringing. But with a software process, if you achieve the same result by a different method, it IS infringing. The fact that that is possible should invalidate the original patent, rather than result in an infringement.
(In the post-In Re Bilsky days, I'm not sure how much of that has gone away, and if Microsoft's patent now isn't useless because of it.)
Microsoft knew what they needed to do back in the NT 3.1 days.
Remember, Microsoft used to make a UNIX product (Xenix); Microsoft was very involved in the OS/2 development.
Part of their selling strategy for NT was it was more convenient and easier than those cumbersome systems. And that was because you didn't have to have admins setting up security contexts; there weren't any. Even though the system (NT/2K/XP/2003/Vista/..., not 95/98/ME) could do them.
They created the mess that is Windows today so they could be number one in the market. And it's been catching up to them for the last few years.
Heck, if we're going by ability to do routing, rather than "enabled by default", the operating systems I've used where you can configure a box this way are: AIX, Mac OS (with OpenTransport), Windows NT family, Linux, BSD, Mac OS X, and so on.
Solaris comes in "pwn me" mode by default. This is not a feature. (Patches will even re-enable things like telnet and rsh if you've previously disabled them... at least when I patched a Solaris 10 Update 4 box it did.)
I've run into one of those annoying runtime licensing systems that not only uses out-of-the-box static build of OpenSSL in its code, it's an older version, too.
Unobfuscated. With all the original OpenSSL symbol names. But they don't provide _all_ of OpenSSL, so you can't just use their old & busted one.
Yes, this causes a serious multiple-definition problem if you want to use that library in an SSL application.
Their "fix"? "Remove these filenames from the.a file we sent you. And these ones. And these, too."
Slashdot Mirror
Been there....
I'm the cynical UNIX bastard, and have been for 15 years now. I started my current job about 10 years ago, on the strength of my UNIX scripting, trouble-shooting, and ad-hoc admin skills. And being able to run the most effective test lab in the previous company with hardly any budget. (A LOT of scrap machines from other departments can be very useful... if you can distribute the workload.)
Anyway. So the new company wants to try out ClearCase. And try it properly: deployment, usage, disaster recovery, everything. So the guy in charge of the IT backup systems says, "We've got the backups working just fine!"
And I ask, "How about the restore?"
And everyone laughs. Except me. And my manager looks... and thinks... and says, "OK, so what about the restore?"
"We don't need to try that, it's working."
My manager insists.
Next status meeting: "So how did the restore go?"
"Well, we're having some trouble with it...."
I tried not to be smug.
That's fine with stuff they develop.
But when they buy a working system from someone else, that people depend on, they shouldn't mess with it.
I think the real failing of UAC is that there are no cues you're about to cross a privilege threshold.
That, and since the default is a click-to-continue model, it doesn't really LOOK like it's changing privilege. (Whereas 'sudo'-based privilege changes always ask for the user's password. OS X uses 'sudo' under the hood to do its switching.)
There's nothing about the "Advanced Settings" in the system control panel that would let me think it's going to change privilege when I click on it--because I want to set some environment variables for my account.
(BTW, I've only seen UAC on Server 2008, which is mainly Vista in server drag, but I'm not sure how close.)
So, first of all, all the control panels need to be separated so "Admin" function is separated from "User" function. I should not have to become admin to change _my_ environment, I should only have to do that to change the _System_ environment.
Then there should be visual cues, like closed-or-open-padlock icons, around the stuff that will trigger a privilege elevation.
And there absolutely SHOULD NOT BE "installer detection" resulting in automatic prompting for elevation. I've used--and written--installers that work just fine in a non-Admin account. Especially in Server 2008, teach the admins to "Run as Administrator" if they want. (Modern installers should be able to ask for elevation, so we're talking about "legacy" installers.)
Oh, let's make the default ask-for-password rather than click-to-continue, too?
Oh yeah: Let's fix what happens when you turn UAC off. It doesn't turn off prompting, it actually removes the privilege separation. How about just have it turn off prompting? Sure, that's next to useless on a machine connected to the Big Bad Internet, but it's still better than the "You're in the Administrator group, so everything you do you do as Administrator" behaviour from older versions. So the settings can be, "No UAC (always Administrator); Unprompted Elevation; Prompted Elevation; Ask for password."
I'm <postmaster@localhost.localdomain>.
Send all your advertising fliers there.
I love it when Amazon runs out of data mining steam:
Usually there's some attempt to get things right, like "you bought a movie in Chinese, so anything else in Chinese must be interesting!" Or, "You've only ever bought PlayStation 2, 3 and Nintendo Wii games, so you must be interested in all our new Xbox 360 titles!"
Not without its flaws, though.
The HL5250-DN has a duplexer, which is nice. But it's got a sharp pointy-bit in just the wrong place so that if you print on 3-hole paper it's just paper jam city. If you take the machine apart, it's easy enough to file smooth so it works fine. But by the time you figure out how to do that, it will be making weird clunky noises for the rest of its life.
Other than that, BR-Script 3 works fine, sits on the wired LAN nicely, and plays well with both the Linux and Mac OS X children.
They claim separate imaging and toner cartridges save money, but I'm not convinced. It just makes it more difficult when the drum needs to be replaced, because most places just have the toner. (But get the extended life carts. Lots of happy DB2 manual printing without worries.)
Yeah.
Just grep the spool like kibo intended.
There's a local catalogue store that has a neat system. They've got paper catalogues and merchandise on display, like most such stores, and you fill out a paper form and bring it to an order taker to get the stuff you want.
But you can also use an in-store computer terminal and use your member ID to get a ready-to-scan completed order form, plus it's already done the inventory verification step.
To protect your personal info, you must stand on a rubber pressure mat like they used to have for opening grocery store doors. When you step off the mat, it wipes the session and brings the machine back to the "Please stand on mat to start a session" logo screen.
You still have to trust that they actually work, but you're handing your membership info over to the store if you want to buy something anyway....
I could see something similar triggering a screen locker instead. Get up off your desk chair for more than 5 seconds and the screen locks.
Yes.
Any resource allocated is something that needs to be tracked. Those tracking objects take a little bit of management time. What kind of time and when depends on the object: sleeping processes aren't checked by the scheduler, but other process operations like signal need to.
Directory entries add up. Even if you have a hash or tree directory index structure, you may still have to scan the whole thing to find files by pattern (for example). With many directory indexes, you must do an end-to-end scan to find out if a name is truly unique.
And almost ALL of the "cruft" that is added to new systems is stuff that "does things" automatically. In order to do that, they all wake up on certain events (including timer events) and check to see if they need to do something. Like multiple devices sharing an interrupt, it all adds up.
Sure, a new system has enough power that it retains good interactive response with all that. But run an old OS on a new system, and wow is it fast.
DVD region codes are orthogonal to DVD copy protection.
You can have a CSS-protected disk which plays in all regions.
You can have an unprotected disk with region codes.
The only way they become commingled is via contracts with the DVD CCA. To get your CSS decryption key, you must agree to obey region codes and user operation prohibitions. (And as the parent says, getting the patent licenses is tied in there, too.)
The DVDs I bought from Japan and have to watch via a region-free player in Canada honour the copyright holders rights completely. And the retailers: the mail order store in Japan got their part of the transaction, as did the Japanese and Canadian postal services. It's not like the DVD was cheaper to import than it would be had it been available in Canada in the first place.
Us movie viewers are not a party to the ridiculous territory contracts and exclusivity deals the studios have with various regional distribution houses. So we can buy something retail and import it, no problems. We didn't make a copy, so copyright doesn't apply. We--and the selling retailer--aren't a party to the territory contracts, so we're not violating those contracts.
The only laws which might apply are the obscenity ones.
You missed a step in the evolution, that actually I think helps your point.
First manual chokes became integrated with another control. If it was cold, you needed to fully depress AND RELEASE the gas pedal BEFORE turning the ignition key to "START". That would engage the choke, which now had no indication to the driver that such a device existed in the vehicle and no indication of its current status. (There were variations on this, too; like press and release twice. Borrowing a friend's car in winter needed a 10 minute explanation of the gas pedal.)
On some vehicles, release was automatic, based on engine RPM (sensed via vacuum). On others, you had to "tap" the gas pedal to clear the choke function. If you didn't, the car could actually over-rev (to redline), and there wasn't a limiter to protect the engine.
All in all, the worst combination of "hidden controls" and "necessary controls".
Early EFI, on the other hand, came with huge warnings "DO NOT DEPRESS GAS PEDAL WHEN STARTING" to try and break people of habits introduced by carburettors and all the kludges needed to get them working.
You can have my modern EFI engine if you can pry it from my cold, dead hands. Every time I do the choke-and-throttle dance on my motorcycle, I curse everyone involved in inventing it.
They enabled "disconnect on hangup" on your line. If you have a burglar alarm installed that uses your phone line, the alarm company will arrange for the same feature.
Traditionally, POTS lines aren't disconnected until both sides go on-hook. With disconnect on hangup, the line is disconnected when one side goes on-hook, though it may take up to 10 seconds.
I dunno; I've been in two houses where the floor slab has been ripped open and partially re-poured.
My parent's house, when we put a basement bathroom in, needed the slab ripped out around the drain lines for the new drain connections.
My current house has new slab in the front 1/3d of the basement, where the furnace and hot water tank are. I don't know what was wrong with the old concrete, but in the rest of the basement, you've got to be careful with the broom or you sweep up concrete instead of just dust and dirt.
It's only the slab, not the foundation (footings and wall). Breaking the footings open, that's bad. Undermining the footings, that's bad.
Floor slab? This way to the drains....
Besides, if the pipes fail, cut the ends flush with the slab, plug with mortar, and get on with your life. That's what they did with the old natural gas supply pipe; it's cut and plugged on both the inside and outside, where it used to go to the city main.
Come to think of it, they did the same thing with the oil tank fill and breather pipes at my parents when we had the oil furnace removed. (Very important to plug the filler side, especially if the oil company doesn't get the cancel notice to the delivery truck....)
Interestingly, a LOT of people, let's say almost everyone, get Make wrong because they don't understand how the declarative rule structure works.
I'm not a fan of ANT, actually; I think it exists to solve a problem that Sun should have never created in the first place. (Produce arbitrarily named .class files from a given .java file.) I don't know how declarative it _really_ is under the covers, but everyone I see saying "it's better than make" are doing procedural things with it and using that to justify ANT over make.
(My gripe with ANT--aside from the XML soup--is it affords sloppy build systems: it's a wildcard enabler.)
Same in Canada; PayPal is the best choice if a border is involved.
Postal money orders between Canada and the U.S. are reasonably OK to get, though expensive on the purchaser's side ($3.50-$8.00 depending on your bank or post office). Problem is, you don't know when you receive a money order if it's really going to stay "good"; there's a lot of bogus ones.
I've never been able to find two people who understand what an International Postal Order is. These are supposed to be able to transfer funds in a large number of currencies between a large number of countries. But you have to find a post office on each side with someone working there who knows how to do it.
Wire transfer is OBSCENELY expensive. Like starting at $40 to send $20 to Australia. And we're both Commonwealth Realms!
I can negotiate a U.S. Funds cheque with my Canadian bank account... for $25. "With $20 USD Mail-In Rebate" means "you pay $5 for this rebate ha ha". Most U.S. banks can't take a cheque (er, check) in Canadian funds.
I can endorse my cheque "U.S. FUNDS" and then the U.S. banks... still probably won't take it, but they should. For some fairly extreme fees on my side.
I can open a U.S. Dollar account and write cheques that are processed through a U.S. Bank... for a suitable monthly fee.
(All the cheque stuff, of course, has even more trust and fraud issues than postal money orders, be they international or "named currency".)
Interac e-mail money transfer only works for Interac users, which is everyone with an ATM card in Canada, but that's kind of limited to Canada....
It came down to, "I'll go to the currency exchange and buy AUD$20 in cash and mail it to you" at one point. That or PayPal were the practical choices for sub-multi-thousand cash transfers. And with PayPal, neither of us had to get off our fat lazy butts.
For anyone with enough volume, of course, get a real merchant account. But for otherwise private individuals, especially ones who "know" each other through a separate channel? PayPal is the best choice... as annoying as that as.
For free?
Damn, I'd pay a dollar a drive to watch the shredder go!
There's a good argument for having page zero stuffed with zeros, execute-prohibited and write-prohibited but allow reads. It's got to do with pipelines and test operations.
(Any relation to some weird cross combination of MC68K, POWER, PowerPC, and SPARC assembly is entirely likely.)
So, on some systems, it's worth starting the load through a null pointer before the test of the pointer has completed. AIX on POWER and PowerPC, for example, guarantees that Page Zero is mapped read-only and contains only zeros (which is also an illegal instruction). This way, the pipeline stalls are minimized; if you didn't have anything else to do in that subroutine, you'd have an effective nop after the test, like so:
This applies to READ operations only, of course; WRITE operations must verify the pointer, as a write attempt to page zero will fault.
So, Linux: Why is page zero modifiable from userland? A program specifying a load address of zero MUST be rejected.
The thing about UNIVERSAL public health care is...
It's actually not very important if you correctly identify who is getting health care.
Your doctor needs to know who you are, and lab results need to be correctly tied to samples, and so on. But that's not a function of who's paying the bills.
But for determining if the doctors and labs get paid? Not so much.
Basically, all you really need to know is, "is this person a citizen or lawful immigrant?" and "is this procedure covered by the system?". It's not so important to know WHICH citizen or lawful immigrant. It's nice to get it right, but your medical history doesn't need to be part of your public health insurance ID, so it's not critical to treatment.
Different keying problem.
That being said... I'm amazed at how many people think there's some huge government conspiracy out to get them when they can't get simple stuff like this right. Sure, they can listen in on all cell phone calls... but they can't keep a list properly?
Is it not possible to calculate the thermal heat in 1L of gasoline when burned at STP at the stoichiometric ratio? If we can get BTU or BTUh out of it, we've got something to work with.
Then compare that to the kWh at your power meter to charge the vehicle enough to travel until the battery system shuts down. (Disable the gasoline system on Volt-like cars.) Count miles. Repeat, average, etc.
I was able to find the right figures for natural gas, so I was able to figure out if it was better to use a single-room space heater or turn up the furnace. (Unsurprising result: A little bit of electricity for one room is better than a lot of gas for the whole house.)
They're also throwing around figures like "40 cents to charge, for 10 kWh, at Michigan off-peak rates." OK, sure, but in Ontario, that'll set you back over a Canadian dollar, as almost nobody has time-of-day metering. Massachusetts will be closer to $2 USD (20 cents/kWh)--which is still half the price of gasoline at last summer's prices in Ontario ($1.25/L).
Here we go. 125,000 BTU in 1 gallon of gasoline, which is about 37 kWh.
So, at 40 miles/10 kWh we've got 4 miles/kWh, which I didn't need Google for but so you can see what I'm doing (show your work). That's the easy one.
I've seen 50 miles/gallon cited for the Volt, so we want miles/kWh... 1.36 miles/kWh.
Both of those are "at the pump/plug" numbers: What you use in electricity net of any generation and transmission losses, compared to volume of gasoline from the pump at your filling station net of energy used to process it from the Alberta tar sands.
(My physics teacher would freak out at that SI and US Imperial unit soup, too.)
What I want to know about all of these electric things though... especially if they're quoting Michigan off-peak power prices... what happens in winter? Those of us in northerly climates don't just throw away all of the thermal inefficiency in the internal combustion motor. We vent some of it in to the passenger cabin as "heat". I'm not giving that up; and resistive electric head for the passenger cabin will kill your battery range real fast--everyone's left just the headlights on and needed a boost at one point, right? That's only about 180 Watts (two headlights + assorted markers.)
And System Rescue Linux and a couple of moments with the l33t haxxor tool "vi" can bypass that check in Dell System Restore.
This message brought to you by "Partition 3", "autoexec" and "bat".
Plugging a big monitor into an old Mac was quite something. The menubar would be all in the top left (Apple File Edit ...).
I thought my mouse clicks should echo, there was so much screen space.
That "huge screen" is, of course, the 21" one I'm working on now, that I'm finding cramped and full.
Don't ask me how I survived with 640x200 on the Amiga, or 40 column text (320x200) on a C=64. But I got my stuff done....
The problem with patenting a word processor file is, it's not their words.
There's a patent on my kitchen faucet, but the water that comes out of it isn't affected by that.
The way the USPTO reads software patents, also, makes things weird. With a physical object, if you achieve the same result by a different method, it is non-infringing. But with a software process, if you achieve the same result by a different method, it IS infringing. The fact that that is possible should invalidate the original patent, rather than result in an infringement.
(In the post-In Re Bilsky days, I'm not sure how much of that has gone away, and if Microsoft's patent now isn't useless because of it.)
Microsoft knew what they needed to do back in the NT 3.1 days.
Remember, Microsoft used to make a UNIX product (Xenix); Microsoft was very involved in the OS/2 development.
Part of their selling strategy for NT was it was more convenient and easier than those cumbersome systems. And that was because you didn't have to have admins setting up security contexts; there weren't any. Even though the system (NT/2K/XP/2003/Vista/..., not 95/98/ME) could do them.
They created the mess that is Windows today so they could be number one in the market. And it's been catching up to them for the last few years.
Heck, if we're going by ability to do routing, rather than "enabled by default", the operating systems I've used where you can configure a box this way are: AIX, Mac OS (with OpenTransport), Windows NT family, Linux, BSD, Mac OS X, and so on.
Solaris comes in "pwn me" mode by default. This is not a feature. (Patches will even re-enable things like telnet and rsh if you've previously disabled them... at least when I patched a Solaris 10 Update 4 box it did.)
Mua-ha-ha-ha.
I've run into one of those annoying runtime licensing systems that not only uses out-of-the-box static build of OpenSSL in its code, it's an older version, too.
Unobfuscated. With all the original OpenSSL symbol names. But they don't provide _all_ of OpenSSL, so you can't just use their old & busted one.
Yes, this causes a serious multiple-definition problem if you want to use that library in an SSL application.
Their "fix"? "Remove these filenames from the .a file we sent you. And these ones. And these, too."