Oh good, it's not just me that feels that way about audio blogs or podcasts or whatever they're called this week.
There are some that I do want to listen to, but none enough to actually make time to listen to them. I like reading on the bus, so I can't listen to a podcast. At home, they're boring enough that I want to do something else, but interesting enough that I still have to pay attention.
So they just sit in the download directory and never get played.
And I don't know about other users, but when it rendered in Mozilla 1.7.11, I'd get the last line of one column clipped off and the remainder of that line at the start of the next column.
For example, all the descenders for the first column were at the top of the second column. The bottom line of the second column was cut a few pixels higher, so the letters were split half-and-half between the two columns.
Screw that, if I want to read something, I don't want to visually glue letters back together.
Not only that, but I found it very hard to make the visual transition from one page to the next, my eyes weren't sure where they should focus--did it move one column, two, or three?
Give me a single column of text, let the scroll-wheel work (so don't have JavaScript make any input boxes active on me), and let the layout re-flow to my current browser width--if the ads are so wide the browser needs to horizontally scroll, don't make the artical spill off the right of the viewable area too.
I got fed up with Spotlight pawing through everything, so I changed a line in/etc/hostconfig:
SPOTLIGHT=-NO-
It was particularly unhelpful for Spotlight to start indexing every SyQuest cart I inserted to see what was on it, and then to erase it for sale on eBay.
Us old grouchy UNIX guys prefer find... -name... over Spotlight anyway.
I'm not sure if that change stops Spotlight from accumulating information about newly-created files. It does stop the background indexing service.
I keep hearing that SCSI drives are more reliable. But I think you can get good reliability from IDE, you just have to stay away from the absolute bottom end of the market, the ones that are still 5400 RPM with 2 MB buffers. 10kRPM SCSI drives are hotter, and pull more power, and both those you don't want in a small computer system. (So much for the SCS part of SCSI....)
There are several things that give you good life from your drives:
Cooling--a hot drive will die.
Good power--out-of-spec power will lead to problems, hot spots in the motor controller mainly.
Backups--if you don't have another copy, the drive will die.
I've had SCSI server drives die every 3 months... until I watched the CSR putting the new drive in and pointed out the cooling fan on that particular sled wasn't spinning. (Sadly, he had no idea what to do about that, so I let him use one of my spare sleds... I just wanted the damn machine working again. A better CSR got the defective fan replaced.)
An entire line of IBM RS/6000s was made with the cooling vents around the drives blocked, because the cut-out in the acoustic damping foam hadn't been removed at the factory. Once the CSR knew about this, he and I went through the entire machine room, pulling all those foam strips out. Disk failure rates dropped way down.
I'm running the near-cheapest-per-gigabyte OEM PATA IDE drives with 8mb buffers and 7200 RPM spindles. So not absolute junk, but sure nothing special. Retrospect does incremental backup on all systems every night, to multiple archive drives. I've had one drive die--a laptop HDD in an external USB cage. (Turns out, that one was running too hot--I'm very glad I never put it in the laptop, it would have been much too hot in there.)
But I'm prepared if a drive dies--that makes all the difference. I don't know how they do it, but drives can really tell if you desperately need them to keep working. And that's when they crash.
Further, modern drives really are a lot better than those from 5-10 years ago. There are much fewer cases of thermal cracking in control boards leading to drive failures. With nearly-all-in-one-chip control board designs, there is much less wiring to fail, the circuit boards are smaller and simpler. Power controllers are better and run cooler. (Though my 250 GB Maxtor is handy when I want to fry eggs for breakfast....) Fluid dynamic bearings run quieter. Everything is voice-coil actuated--remember stepper motor HDDs? (Want some? I've still got a couple around.)
I don't have anything sitting on SyQuest cart, QIC-150 or 4mm4gb DAT that isn't still on a "live" HDD volume on some system or another.
But I did still have the backups. So I finally started just erasing the tapes and made an offer to give the drives and media away for the cost of postage. The SyQuest drives and carts went via eBay.
Heck, everything that was on the Amiga barely takes up any space on the Linux fileserver. The Mac Performa's HDD fit on a CD-ROM. The iMac's old drive is in a FireWire cage plugged in to the Mini, mounted read-only. (That's not easy. There's no direct support I can find for having DiskUtility make something read-only, you have to do it with BSD stuff.)
I've only just gotten around to destroying bills and stuff from 1990; it's not like any it matters, not even if the tax people audit me. (And audits only go back 7 years, tops.)
It is also comparitively painless to build Perl natively Windows, given the now-free Visual C++ commandline compiler environment. (We don't do much C++ on Windows where I work, so Visual Studio 6 is still state-of-the-art.)
The native Win32 version does not suffer from any of the issues the Cygwin port has; it's a real Windows program, so you get real Windows paths and so on. It can even look for things like perl5lib in the Registry, if you're into that sort of thing.
Yes, the article completely missed out on the Very Most Important Font Size Issue on the Web.
That is, of course, how Windows treats points as equivelent to pixels, whereas Macintosh and UNIX system treat points as 72-per-inch like they're supposed to be. (X11 has lots of problems with font rendering if you use the older APIs, but it does know how to read the DDC codes from your monitor to calculate the correct resolution: check xdpyinfo | grep dimensions.)
(I don't use Windows enough to know--do Mozilla and Opera have the same pixels vs. points issue, or is it Internet Explorer only?)
But the absolute best advice on font size is the one you offer: relative sizes from the user's chosen default.
The Minimum Font Size setting in the new Mozilla and Safari browsers is crucial to us non-Windows users. But, that just sets a floor--without relative sizes, we see no font size changes for anything that is specified "too small".
Spam to my domain does not go to any of the e-mail addresses visible in whois. OK, maybe it does and pobox.com takes care of it like any other spam.
But stuff that does make it--by direct e-mail to the domain--they just make up. accounting@, help@, support@, admin@, webmaster@, root@, postmaster@, and so on. And usually they mail to my backup MX, instead of the primary MX, so I can't use a DNSBL to block them, either. (My backup MX is provided by dyndns.org, so I don't have control over it.)
It was fun to allow any address through, so I could make up addresses for whatever I wanted and Postfix would drop them all in one inbox. But 10 "Click to Advertise Your Site" spams a week made that not fun anymore.
You should see the stuff that came with my step-father's Olympus.
It looks like NOTHING ELSE I've ever seen on a computer. Not under Windows, X11, MacOS, Mac OS X, Amiga, Atari ST, whatever. It looks a bit like a bad flash-driven videogame. Even my mom was saying, "Why did they do that? Why didn't they just make it like a normal program? How am I supposed to remember to do that, that doesn't make any sense at all! Why did you click that? I don't want to do that... oh, now you can print? But why did you have to do that first thing? Where did it put the pictures? Why didn't it just put them in the picture directory?"
The extra-pathetic part is, that's the software that the people who don't understand computers and cameras and how they go together are going to use. The rest of us just pop the flash card out, into a USB or FireWire reader, and drag stuff into whatever app we want to use or whichever folder we want.
I'd really appreciate iPhoto for Windows... "Here folks, plug the camera in, click the 'Import' button. When it's done, click on the pictures you want, and everything you need is in the menu called 'Share'."
All they want to do is print and e-mail family photos... that stupid Olympus thing didn't even have a way to scale a photo for e-mail... and the hoops you had to go through to get one on paper, wow. That was amazing.
Properly-crinkled tinfoil will dampen the echos instead of reinforcing them.
But, for Best Results(tm), get some christmas tree flocking and spray the inside of your tinfoil hat with that. Then you'll have an anechoic tinfoil hat, and have the best of both worlds: no echos from the voices AND govenment mind-control protection.
I think we really want to be talking about "reasonable doubt" and not "plausible deniablility".
Except, in civil suits, it's "balance of probability".
So, having an intentionally-open access point to allow others access to the network increases doubt that the accused, or the accused's computer, performed the allegedly infringing download. (The access point being separate from the computer.)
Re:Better than mobile phone addiction
on
Hooked On The Web
·
· Score: 1
You just need to start a steady diet of greasy pizza and baked beans.
Then you can provide appropriate accompanyment to your co-workers who use the phone in the can.
OK, but for case (1) ("Content-disposition: attachment"), you've still asked FF to save it to disk automatically. In particular, it SHOULD NOT ask "Do you want to save this?". Ever.
Even if there isn't "; filename=" on the Content-disposition header, you can guess at one by removing the last path element of the request URI. FireFox already asks for filenames much less often than Mozilla, so I don't want to see a filename request, either.
I have heard that manually adding an "application/binary" entry in Helper Applications will prevent that; apparently, FireFox and Mozilla don't actually save the choice you just made for that MIME type.
I think I did it on at least one of my machines, and have since forgotten if I did and/or if it worked. Which isn't very helpful... but Safari saves without prompting just fine.
Basically, if you don't have the entire stack, and a completely independent way to compile it, you have no idea what is happening in a completed stack. Especially if the code running at high privilege; you could have your I/O drivers replacing code blocks on load so that the application suite audits correctly.
Look at how much spyware for Windows works by intercepting basic system calls. Unless you have a trustable, independent way of re-creating the software stack, and then verifying that exact stack is actually running on the machine, you've got no reason to trust the box.
So, for any environment where trust is important, almost any operating system is too complicated.
Maybe not "COMMODORE BASIC V2", even though it's from Microsoft.
Yeah. I had a Toshiba 80 gig laptop drive fail on me after a couple of years use.
It wasn't even on as much as the laptop; I had never bothered opening up the laptop to swap the drive in, so it just lived in the external aluminum enclosure that I originally planned to put the laptop's old drive in.
If the drive is stone-cold, I get about an hour or two runtime out of it. Any longer and it stops working. And the bearings sound really, really bad.
I guess it would take a lot longer for them to test longevity, though. Hate to wait 3 years to find out what would have been a good choice....
I've got two Macs talking wirelessly to my Linksys router/access point, which used to be a BEFW11S4 and is now a WRT54GS. I've got a D-Link AirPlus 3-port print server talking to that, and a Roku soundbridge playing very happy with iTunes on one of the Macs and streaming radio from Digitally Imported.
My very first-ever Apple base station arrived this week, an AirPort Express Apple sent me to make up for the fact that my stupid G3 iBook is now on its SIXTH Logic Board... at Apple's expense, but it's still annoying to have to take the dumb thing in all the time for repairs. And all that is doing is AirTunes to the hi-fi. (Eliminating a long USB cable that was running an Edirol USB ACPA for digital out to the hi-fi.)
The only thing you have to keep in mind is, the password that you put in the Linksys "Generate" box is useless for anything else. Everything else has to use the hex key for WEP--the D-Link print server (which never forgets its IP), the Roku Soundbridge, the two Macs, and the Airport Express. (Still got 802.11b devices, so no WPA for my 'net.)
I did have trouble with the WRT54GS going into some strange "instant security" mode where it wouldn't take new wireless clients. But that was because I saved and loaded a config file across a firmware upgrade, and the old config file did not have "don't use that instant security button thing" set properly. (The instant security mode was added in the firmware upgrade. It's when that button on the front flashes white. It's useless unless you have to gadgets from Linksys with that button.) If I'd left the firmware alone, or had done the upgrade BEFORE setting up the router, it would have been no problem (unless I accidentally pressed the button).
But getting a Windows machine to work on a secured wireless network? That's something I've never succeeded at. Maybe it's time to try again, get a cheap PCI WiFi card for the sacrificial XP box.
And I have a fast (1.8 GHz processor running Konqueror) setup and broadband. I can just imaging the difference if I was on an old sub-GHz machine[....]
I can. I'm putting together the scraps of old Macs and PCs in the basement and finding which go to landfill (2 Macs), which go to eBay (1 Mac), and which go to Freecycle (2 PCs).
Part of testing was checking out various NuBus Ethernet cards, so I loaded up iCab, normally a lightweight quick and snappy browser. I haven't watched a jpeg render line-by-line in a l-o-n-g time.... Granted, the NuBus machine is a 40 MHz 68040, so one shouldn't expect... well, anything, really.
I know there was a time when I thought 40 MHz was pretty impressive, but for the life of me I cannot remember it....
> greet english speaker What do you want to greet the Native English Speaker with? > axe You wave the axe at the Native English Speaker and a friendly way. Unfortunately, your grip slips, and the axe strikes the Native English Speaker. You have killed the Native English Speaker. > restore
There's nothing wrong with amortizing the cost of the vehicle onto monthly basis.
When I bought my car, I got a good loan rate, so I put a couple K down, and financed the rest.
When it was paid off, expenses didn't stop. I had picked the loan length so that it would be paid off about a year before the first major scheduled service point--new tires, new timing belts, everything liquid, you name it, they change it, measure it, or adjust it.
Granted, the service cost, spread out monthly, is less than half of what the loan cost. But it doesn't make sense to NOT get a 2.9% car loan (this was in 1998, rates are lower today) and put more money towards your 5.9% mortgage, or invest in a 6% GIC, or add more to those mutual funds that are returning 8% to 12%.
You DO want to get the "no depreciation for 2 years" version of your insurance, so if the car is written off in the first two years, you get full purchase value for another try, not the depreciated value.
And don't ever cancel comprehensive unless you're prepared to pay cash, right now, for repairs or a new vehicle if something happens.
Filesystem performace is slower than ext3 with small files? Er....
System hangs with Adaptec SCSI adapters? No, that's not good either....
I know! Even Worse Video Card Support! Yay Solaris!
I tried "Solaris Admins Already Know How To Admin It," an argument recommended by Sun, and our Solaris admins said, "We're Not Touching Solaris x86. You want x86, you get Linux."
Normally, on a gasoline motor, power steering is powered by a continuously-driven pump connected to the "serpentine belt" (formerly "fan belt"), and power braking is done by manifold-vacuum assist. As it turns out, when you've got the throttle closed (foot off the gas), and the transmission engaged (so the motor RPM is kept up), you get a large amount of manifold vacuum, and so that works out really well. (On a manual transmission, don't jump on the clutch until the last moment, or you may need more brake pedal pressure than you would otherwise....)
Many (most? all?) diesel motors regulate power by adjusting the fuel injection, not by throttling airflow. Consequently, they don't produce much in the way of manifold vacuum... and so driving power brakes from the power steering pump is pretty much the only way to get power brakes.
You might want to Google up "hydroboost", a common name for this technique.
IBM dropped the classic meaning of the POWER term with the Power3 CPUs; they are actually from the 64-bit PowerPC family.
Comments about pinout and AltiVec are relevant; TTBOMK, the CPUs IBM is using do not have the same SIMD unit that the Mac G[45] chips use.
And I would sure hope there's a different pinout on the embedded chip....
What really makes the two lines mostly distinct is the string ops in POWER which are not present in the PowerPC family (except the 601), and the presence of single-precision floating point in PowerPC which POWER did not have. (Resulting in a SPECfp "optimal option" flag of "-Dfloat=double" for performance reasons....)
There was enough similarity in the instruction set and CPU architecture that other distinctions are more of a nit-pick; like changing the assembler mnemonics for some operations. The PowerPC is bi-endian; the POWER is big-endian; some of the "lost" instructions on PowerPC are only lost in little-endian mode....
That'd be the Patrician of Ankh-Morpork; a more cynical and realistic leader you would be hard-pressed to find. It's part of the "One Man, One Vote" system: The Patrician is the Man, and he has the Vote.
Slashdot Mirror
They must have been spending their entire earnings on insurance; 2-3 crashes a year on a 30-40 minute commute is insane.
Oh good, it's not just me that feels that way about audio blogs or podcasts or whatever they're called this week.
There are some that I do want to listen to, but none enough to actually make time to listen to them. I like reading on the bus, so I can't listen to a podcast. At home, they're boring enough that I want to do something else, but interesting enough that I still have to pay attention.
So they just sit in the download directory and never get played.
Screw that, if I want to read something, I don't want to visually glue letters back together.
Not only that, but I found it very hard to make the visual transition from one page to the next, my eyes weren't sure where they should focus--did it move one column, two, or three?
Give me a single column of text, let the scroll-wheel work (so don't have JavaScript make any input boxes active on me), and let the layout re-flow to my current browser width--if the ads are so wide the browser needs to horizontally scroll, don't make the artical spill off the right of the viewable area too.
It was particularly unhelpful for Spotlight to start indexing every SyQuest cart I inserted to see what was on it, and then to erase it for sale on eBay.
Us old grouchy UNIX guys prefer find ... -name ... over Spotlight anyway.
I'm not sure if that change stops Spotlight from accumulating information about newly-created files. It does stop the background indexing service.
There are several things that give you good life from your drives:
I've had SCSI server drives die every 3 months... until I watched the CSR putting the new drive in and pointed out the cooling fan on that particular sled wasn't spinning. (Sadly, he had no idea what to do about that, so I let him use one of my spare sleds... I just wanted the damn machine working again. A better CSR got the defective fan replaced.)
An entire line of IBM RS/6000s was made with the cooling vents around the drives blocked, because the cut-out in the acoustic damping foam hadn't been removed at the factory. Once the CSR knew about this, he and I went through the entire machine room, pulling all those foam strips out. Disk failure rates dropped way down.
I'm running the near-cheapest-per-gigabyte OEM PATA IDE drives with 8mb buffers and 7200 RPM spindles. So not absolute junk, but sure nothing special. Retrospect does incremental backup on all systems every night, to multiple archive drives. I've had one drive die--a laptop HDD in an external USB cage. (Turns out, that one was running too hot--I'm very glad I never put it in the laptop, it would have been much too hot in there.)
But I'm prepared if a drive dies--that makes all the difference. I don't know how they do it, but drives can really tell if you desperately need them to keep working. And that's when they crash.
Further, modern drives really are a lot better than those from 5-10 years ago. There are much fewer cases of thermal cracking in control boards leading to drive failures. With nearly-all-in-one-chip control board designs, there is much less wiring to fail, the circuit boards are smaller and simpler. Power controllers are better and run cooler. (Though my 250 GB Maxtor is handy when I want to fry eggs for breakfast....) Fluid dynamic bearings run quieter. Everything is voice-coil actuated--remember stepper motor HDDs? (Want some? I've still got a couple around.)
But I did still have the backups. So I finally started just erasing the tapes and made an offer to give the drives and media away for the cost of postage. The SyQuest drives and carts went via eBay.
Heck, everything that was on the Amiga barely takes up any space on the Linux fileserver. The Mac Performa's HDD fit on a CD-ROM. The iMac's old drive is in a FireWire cage plugged in to the Mini, mounted read-only. (That's not easy. There's no direct support I can find for having DiskUtility make something read-only, you have to do it with BSD stuff.)
I've only just gotten around to destroying bills and stuff from 1990; it's not like any it matters, not even if the tax people audit me. (And audits only go back 7 years, tops.)
I think I might be a packrat....
It is also comparitively painless to build Perl natively Windows, given the now-free Visual C++ commandline compiler environment. (We don't do much C++ on Windows where I work, so Visual Studio 6 is still state-of-the-art.)
The native Win32 version does not suffer from any of the issues the Cygwin port has; it's a real Windows program, so you get real Windows paths and so on. It can even look for things like perl5lib in the Registry, if you're into that sort of thing.
Yes, the article completely missed out on the Very Most Important Font Size Issue on the Web.
That is, of course, how Windows treats points as equivelent to pixels, whereas Macintosh and UNIX system treat points as 72-per-inch like they're supposed to be. (X11 has lots of problems with font rendering if you use the older APIs, but it does know how to read the DDC codes from your monitor to calculate the correct resolution: check xdpyinfo | grep dimensions.)
(I don't use Windows enough to know--do Mozilla and Opera have the same pixels vs. points issue, or is it Internet Explorer only?)
But the absolute best advice on font size is the one you offer: relative sizes from the user's chosen default.
The Minimum Font Size setting in the new Mozilla and Safari browsers is crucial to us non-Windows users. But, that just sets a floor--without relative sizes, we see no font size changes for anything that is specified "too small".
But stuff that does make it--by direct e-mail to the domain--they just make up. accounting@, help@, support@, admin@, webmaster@, root@, postmaster@, and so on. And usually they mail to my backup MX, instead of the primary MX, so I can't use a DNSBL to block them, either. (My backup MX is provided by dyndns.org, so I don't have control over it.)
It was fun to allow any address through, so I could make up addresses for whatever I wanted and Postfix would drop them all in one inbox. But 10 "Click to Advertise Your Site" spams a week made that not fun anymore.
Though appropriate body_checks help.
It looks like NOTHING ELSE I've ever seen on a computer. Not under Windows, X11, MacOS, Mac OS X, Amiga, Atari ST, whatever. It looks a bit like a bad flash-driven videogame. Even my mom was saying, "Why did they do that? Why didn't they just make it like a normal program? How am I supposed to remember to do that, that doesn't make any sense at all! Why did you click that? I don't want to do that... oh, now you can print? But why did you have to do that first thing? Where did it put the pictures? Why didn't it just put them in the picture directory?"
The extra-pathetic part is, that's the software that the people who don't understand computers and cameras and how they go together are going to use. The rest of us just pop the flash card out, into a USB or FireWire reader, and drag stuff into whatever app we want to use or whichever folder we want.
I'd really appreciate iPhoto for Windows... "Here folks, plug the camera in, click the 'Import' button. When it's done, click on the pictures you want, and everything you need is in the menu called 'Share'."
All they want to do is print and e-mail family photos... that stupid Olympus thing didn't even have a way to scale a photo for e-mail... and the hoops you had to go through to get one on paper, wow. That was amazing.
Nice enough point-n-shoot camera, though.
But, for Best Results(tm), get some christmas tree flocking and spray the inside of your tinfoil hat with that. Then you'll have an anechoic tinfoil hat, and have the best of both worlds: no echos from the voices AND govenment mind-control protection.
Except, in civil suits, it's "balance of probability".
So, having an intentionally-open access point to allow others access to the network increases doubt that the accused, or the accused's computer, performed the allegedly infringing download. (The access point being separate from the computer.)
You just need to start a steady diet of greasy pizza and baked beans.
Then you can provide appropriate accompanyment to your co-workers who use the phone in the can.
OK, but for case (1) ("Content-disposition: attachment"), you've still asked FF to save it to disk automatically. In particular, it SHOULD NOT ask "Do you want to save this?". Ever.
Even if there isn't "; filename=" on the Content-disposition header, you can guess at one by removing the last path element of the request URI. FireFox already asks for filenames much less often than Mozilla, so I don't want to see a filename request, either.
I have heard that manually adding an "application/binary" entry in Helper Applications will prevent that; apparently, FireFox and Mozilla don't actually save the choice you just made for that MIME type.
I think I did it on at least one of my machines, and have since forgotten if I did and/or if it worked. Which isn't very helpful... but Safari saves without prompting just fine.
You haven't read Ken Thompson's famous bit on how to trojan the compiler and a particular application so that you can't find any trace of the trojan in the source code for either one, then? (Was the first hit on a Google for "compiler trojan trust".)
Basically, if you don't have the entire stack, and a completely independent way to compile it, you have no idea what is happening in a completed stack. Especially if the code running at high privilege; you could have your I/O drivers replacing code blocks on load so that the application suite audits correctly.
Look at how much spyware for Windows works by intercepting basic system calls. Unless you have a trustable, independent way of re-creating the software stack, and then verifying that exact stack is actually running on the machine, you've got no reason to trust the box.
So, for any environment where trust is important, almost any operating system is too complicated.
Maybe not "COMMODORE BASIC V2", even though it's from Microsoft.
It wasn't even on as much as the laptop; I had never bothered opening up the laptop to swap the drive in, so it just lived in the external aluminum enclosure that I originally planned to put the laptop's old drive in.
If the drive is stone-cold, I get about an hour or two runtime out of it. Any longer and it stops working. And the bearings sound really, really bad.
I guess it would take a lot longer for them to test longevity, though. Hate to wait 3 years to find out what would have been a good choice....
I've got two Macs talking wirelessly to my Linksys router/access point, which used to be a BEFW11S4 and is now a WRT54GS. I've got a D-Link AirPlus 3-port print server talking to that, and a Roku soundbridge playing very happy with iTunes on one of the Macs and streaming radio from Digitally Imported.
My very first-ever Apple base station arrived this week, an AirPort Express Apple sent me to make up for the fact that my stupid G3 iBook is now on its SIXTH Logic Board... at Apple's expense, but it's still annoying to have to take the dumb thing in all the time for repairs. And all that is doing is AirTunes to the hi-fi. (Eliminating a long USB cable that was running an Edirol USB ACPA for digital out to the hi-fi.)
The only thing you have to keep in mind is, the password that you put in the Linksys "Generate" box is useless for anything else. Everything else has to use the hex key for WEP--the D-Link print server (which never forgets its IP), the Roku Soundbridge, the two Macs, and the Airport Express. (Still got 802.11b devices, so no WPA for my 'net.)
I did have trouble with the WRT54GS going into some strange "instant security" mode where it wouldn't take new wireless clients. But that was because I saved and loaded a config file across a firmware upgrade, and the old config file did not have "don't use that instant security button thing" set properly. (The instant security mode was added in the firmware upgrade. It's when that button on the front flashes white. It's useless unless you have to gadgets from Linksys with that button.) If I'd left the firmware alone, or had done the upgrade BEFORE setting up the router, it would have been no problem (unless I accidentally pressed the button).
But getting a Windows machine to work on a secured wireless network? That's something I've never succeeded at. Maybe it's time to try again, get a cheap PCI WiFi card for the sacrificial XP box.
I can. I'm putting together the scraps of old Macs and PCs in the basement and finding which go to landfill (2 Macs), which go to eBay (1 Mac), and which go to Freecycle (2 PCs).
Part of testing was checking out various NuBus Ethernet cards, so I loaded up iCab, normally a lightweight quick and snappy browser. I haven't watched a jpeg render line-by-line in a l-o-n-g time.... Granted, the NuBus machine is a 40 MHz 68040, so one shouldn't expect... well, anything, really.
I know there was a time when I thought 40 MHz was pretty impressive, but for the life of me I cannot remember it....
> greet english speaker
What do you want to greet the Native English Speaker with?
> axe
You wave the axe at the Native English Speaker and a friendly
way. Unfortunately, your grip slips, and the axe strikes
the Native English Speaker.
You have killed the Native English Speaker.
> restore
When I bought my car, I got a good loan rate, so I put a couple K down, and financed the rest.
When it was paid off, expenses didn't stop. I had picked the loan length so that it would be paid off about a year before the first major scheduled service point--new tires, new timing belts, everything liquid, you name it, they change it, measure it, or adjust it.
Granted, the service cost, spread out monthly, is less than half of what the loan cost. But it doesn't make sense to NOT get a 2.9% car loan (this was in 1998, rates are lower today) and put more money towards your 5.9% mortgage, or invest in a 6% GIC, or add more to those mutual funds that are returning 8% to 12%.
You DO want to get the "no depreciation for 2 years" version of your insurance, so if the car is written off in the first two years, you get full purchase value for another try, not the depreciated value.
And don't ever cancel comprehensive unless you're prepared to pay cash, right now, for repairs or a new vehicle if something happens.
...and if you're in Canada, Staples has a $25 in-store rebate and $15 on-line rebate for the WRT54GS model right now, ending Nov 15 I believe.
Filesystem performace is slower than ext3 with small files? Er....
System hangs with Adaptec SCSI adapters? No, that's not good either....
I know! Even Worse Video Card Support! Yay Solaris!
I tried "Solaris Admins Already Know How To Admin It," an argument recommended by Sun, and our Solaris admins said, "We're Not Touching Solaris x86. You want x86, you get Linux."
Normally, on a gasoline motor, power steering is powered by a continuously-driven pump connected to the "serpentine belt" (formerly "fan belt"), and power braking is done by manifold-vacuum assist. As it turns out, when you've got the throttle closed (foot off the gas), and the transmission engaged (so the motor RPM is kept up), you get a large amount of manifold vacuum, and so that works out really well. (On a manual transmission, don't jump on the clutch until the last moment, or you may need more brake pedal pressure than you would otherwise....)
Many (most? all?) diesel motors regulate power by adjusting the fuel injection, not by throttling airflow. Consequently, they don't produce much in the way of manifold vacuum... and so driving power brakes from the power steering pump is pretty much the only way to get power brakes.
You might want to Google up "hydroboost", a common name for this technique.
Comments about pinout and AltiVec are relevant; TTBOMK, the CPUs IBM is using do not have the same SIMD unit that the Mac G[45] chips use.
And I would sure hope there's a different pinout on the embedded chip....
What really makes the two lines mostly distinct is the string ops in POWER which are not present in the PowerPC family (except the 601), and the presence of single-precision floating point in PowerPC which POWER did not have. (Resulting in a SPECfp "optimal option" flag of "-Dfloat=double" for performance reasons....)
There was enough similarity in the instruction set and CPU architecture that other distinctions are more of a nit-pick; like changing the assembler mnemonics for some operations. The PowerPC is bi-endian; the POWER is big-endian; some of the "lost" instructions on PowerPC are only lost in little-endian mode....
OK, some of those are big nits....
That'd be the Patrician of Ankh-Morpork; a more cynical and realistic leader you would be hard-pressed to find. It's part of the "One Man, One Vote" system: The Patrician is the Man, and he has the Vote.