My folks bought a Mercury Villager in 1993 that had ABS as a standard item.
My sister had an old (96?) Chevy Monte Carlo with standard ABS.
I dated a girl with a 1996 Grand Am once, which had standard ABS.
I married a girl who came with a 1996 Pontiac Firebird, which had standard ABS.
And, of course, the '95 BMW has standard ABS. Plus the aforementioned Beretta.
I've driven far more American cars from the 90's that had ABS as standard equipment, than cars that did not.
That all said: Holden is about as American as vegemite is.
I've actually had this conversation:
Me: You should check out the new GTO. It's pretty awesome. Other: What's so cool about it? Me: It's got a big engine, decent handling, it's rear-drive, and the car itself isn't so big. It's based off the Monaro. Other: The what? Me: Monaro. From Holden. Other: What's a Holden?
It is obscure. You can keep saying it's not, but it nonetheless is. (You do the dictionary look-up on that word as homework.)
Combine the tenacity of something like Blaster with the fact that random generally isn't, and such software will land somewhere. Furthermore, I think you genuinely underestimate the number of folks downloading and running such niceties as "FREE Registry Cleaner 9000" and the "OMG PONIES!!!" screensaver, which allows a fair number of seed nodes out of the gate. (I made those names up. You get the point.)
And, of course: Nevermind the fact that such a routeable address will not exactly be secret to begin with: In the absence of NAT, whatever host(s) you communicate with will know this address, and it will no longer be obscure. Given enough datamining on a popular and compromised/ill-intended sites, and producing rather complete maps of an individual's home subnet should be practical.
And speaking practicality: You can have multiple addresses per host. Lots of them. And you can assign them "randomly," and change them periodically. You can multiple firewall rules concocted to mitigate risks individually on these many addresses, and do all manner of other fuckery and confusion (unique addresses for each IPV6 connection pair?) to try to keep things obscure.
But the obscurity will fail. It won't likely fail for you or me (for the same reason I don't plan on winning the lottery), and it won't ever fail for Bruce Schneier (because the force is strong with that one), but it will fail for someone.
So, if we're cannot rely on obscurity by itself, then we'll have to rely on firewalls. This is all well and good, but by the time we firewall the piss out of IPV6 at the border, we've lost most of the advantages of having all of those addresses to begin with.
At that point, one might as well just use NAT and dumb port forwarding anyway, maybe with a couple of extra hooks added to let multicast work and/or some sort of automatia similar to UPnP for those who can't be bothered.
In other news: I've got way more addresses than I can use on my current RFC 1918 subnet. I don't need IPV6 -- the Internet does. I will re-evaluate my position as the environment dictates.
The thing is, the QoS stuff in Tomato actually works in a predictable, useful, and rather flexible fashion. The Linksys QoS stuff, meanwhile, is rather meh.
Further reading has also indicated that TomatoUSB (which I also have installed here, somewhere...) does support IPV6, and although the configuration is all manual at this point. It apparently has the beginning of proper GUI support in the Git repository, though, so it should be ready sometime...
I can break down some of the numbers for you: linky link
Unique wifi networks in DB: 31,592,538 Unique networks w/ location: 30,576,668 Unique wifi locations in DB: 1,211,718,307 Unique cell towers in DB: 25,696
Networks with crypto: 16,194,355 (51.2%) Networks without crypto: 8,295,965 (26.2%) Networks crypto unknown: 7,102,218 (22.4%) Networks with default SSID: 3,181,785 (10.0%) New unique networks today: 35,224 New today with location: 35,221 New yesterday with location: 38,447
I like the WRT54GL. We've got a bunch of them scattered around keeping point-of-sale systems online, and they're absurdly reliable with stock firmware: We never need to fuck with them.
That said, I like them better at home when they're running Tomato, because the QoS features let folks in the house watch a couple of 4-6Mbps Netflix streams while the wife plays WOW and I download torrents and the brother-in-law does whatever he does when his door is shut, all while keeping latency low enough that SSH stays non-annoying.
This, to me, makes it worth messing with.
Granted, Tomato doesn't really do IPV6 yet, either, but it's far more likely to support it on old Linksys product than Linksys's own firmware when the time comes that it really becomes useful and wanted. (We're not there yet, despite the Slashdot hype.)
The 1995 BMW, for instance, has a braking system made by Ate. From the rotors to the pads to the calipers to the hoses to the master cylinder to the electronics, it's all Ate, and is almost (!) independent from the rest of the car. (In this particular example, the ABS system has the ability to also close a secondary throttle to reduce engine output, though it has no ability to open the main throttle to increase it...)
My 2002 GMC van has a system from, IIRC, Bosch. It's the same sort of gig, minus the extra throttle body: All separate and independent.
At any rate, a failure of one system should not impart a failure of the ABS system, at least in these examples. I don't have anything newer to take apart and peer at, however -- I suppose it's possible, or even rather likely, that newer systems are more integrated and might be more susceptible to cascading failure.
I disagree with the notion that ABS is unusual. It's been a pretty standard item on most GM cars for over a decade, and even the not-special-at-all 1995 Chevy Beretta I used to have was equipped with ABS.
You asked. I answer. Pull up a chair, lad, it's story-telling time:
Once upon a time, there was a company called 3dfx. They made video cards, the first of their kind, with all kinds of gee-whiz video acceleration.
Now, along comes nVidia. Their first few 3D chipsets sucked, but later on, they grew up and made real competitors to the 3dfx line.
Eventually, 3dfx was bought by nVidia.
And, it just happened that the weekend after this, I was at a small LAN party with my PC.
3dfx's website was gone. No 404, no response, no nothing. nVidia didn't mirror anything. And I, of course, needed drivers for my 3dfx Voodoo3 2000.
Every. Single. Fucking. Link. From. Every. Single. Fucking. Web page. was a dead link to 3dfx's defunct site. So, there simply weren't any drivers to be found. I didn't even know, yet, that 3dfx had been bought out: I failed miserably at playing games that weekend, despite my keen sense of Internet-fu and the help of the fellow geeks around me.
Some time later, an ad-hoc support structure of hobbyists began providing 3dfx drivers, but that was far too late.
In addition, 3dfx's sudden demise abolished what was left of STB's drivers and documentation. See, sometime before all of the above, 3dfx bought STB, who used to make a variety of different add-on cards for PCs, including video adapters and other stuff.
I have here, somewhere, an STB I/O card which is completely inoperable. Why? While 3dfx did keep STB's old support site alive, nVidia flushed the whole lot. This, despite written, personal assurance from a pre-3dfx STB that they'd keep the old information online indefinitely. The information is simply lost to time, which is plainly fucking retarded.
And don't get me started on the X-Fi card in my desktop. It's an OEM variant that, while quite good and versatile, was never sold separately. Instead, it only came in (IIRC) some Dell, Alienware, and HP machines -- I haven't even seen it for sale in the white-box OEM market.
Creative Labs, in their infinite wisdom, will not let you install software items that have been downloaded from their website unless you already have that software installed (can you spot the glaringly-obvious Catch-22?). Alienware, in their own infinite wisdom, packaged a bad driver disk with the system, wherein the X-Fi drivers and software were both broken and did not match those that were pre-installed with the default load of Windows. And both Dell and HP seem to ignore the fact that the card exists at all when it comes to software support.
Now, Alienware does have 32-bit drivers, with a complete and installable suite of software, available for download on their website (unlike Creative Labs). And, lo, they do work. It's a few hundred megs of shit, though, and it's difficult to find (and Google isn't much help except to find other people looking for the same thing, but failing to find it), and much of it doesn't work on a 64-bit system.
I eventually found and downloaded this, and made it all work on my desktop Alienware box after upgrading to 64-bit Windows. And things are good, the Creative Labs update program works well and automatically downloaded and upgraded stuff, and I'm able to stay current with things.
But: Alienware's buried download is the only place the drivers are available from. If/when it vanishes (and it's not so much if as it is when -- see above), it will be gone. I've burned them to CD, of course, but I'm by no means the only person with this card.
So, while currently functional, it's a fragile situation. The next installation of Windows may not be so easy.
Accordingly, having learned from my previous STB/3dfx experiences, I tried to seed the drivers on The Pirate Bay, hoping that TPB's high Pagerank would let others find them easily. I included a good description of how to make them work, what card they were for, and why I was posting it, but they banned my account and nuked the.torrent after a
However, I think drivers should be informed of the function of the switch as intuitively as possible (who honestly R's the entire FM on a car except for my wife?). It also needs to be designed in such a way as to prevent accidental shutdown, so that the car doesn't shut down unexpectedly on a busy highway (which has its own safety problems).
Here's my idea for making it work:
Instead of a stop/start button with push-and-hold functionality, like we have now, plus an e-stop, as you suggest, let's do the following:
First, make it a toggle switch, with obvious visual and tactile indications of "go" and "don't go."
Second, cover it with something spring-loaded. (There's a wealth of military and industrial design knowledge covering this that I needn't go into.)
Third, eliminate the delay: Hit the switch, engine dies. End of story.
Fourth, put it central-ish on the dash, so that it could be reached by a passenger in the event that everything goes wrong and the driver is already busy dealing with that (or goes into insulin shock, or whatever).
Fifth, ensure that in the event that the e-stop is activated on accident, that putting the switch back into the "go" position will automatically restart the engine and resume normal operation.
Thus, starting the vehicle normally will consist of the following: Open cover, activate switch.
Shutting off the vehicle, whether in motion or parked, will also normally be as follows: Open cover, deactivate switch.
In this way, the operator is already familiar with the operation of the switch, because they have to use it every time they drive. Meanwhile, unintended operation of the switch (by bumping it somehow, a kid being an idiot, or whatever) is precluded by the cover, and is safely reversible in an understood fashion.
It's a little less convenient than the unprotected "Start" button in question here, but I don't think that it is unduly so, and it would make it both plain and obvious how to shut down the car if the need were to arise.
Instructions on the switch's exact behavior can be printed on the sun visor right next to the stupid airbag warnings, which should eventually let the complete idea of the concept be absorbed by the populace in general.
(I'd patent this, but I can't afford to. Consider it prior art.)
Even with no assist from the engine, the brake pedal is physically connected to the brake pads via hydraulics. There is no way for software to prevent the brakes being applied if you put your foot on the pedal.
One word: ABS.
Give the pedal a good hard push on ice with bad tires, and you'll quickly see that there most assuredly is a software method to prevent the brakes from being applied.
Anecdotally, I've also experienced reduced braking power on more than one vehicle with malfunctioning ABS due to wiring faults and/or bad wheel sensors: If the system disables itself as it is supposed to, this isn't a big problem. If it has not yet detected the fault (and, mind you, it resets the fault indicator every time you start the vehicle), stopping even on dry pavement can rather more interesting than anticipated.
I'm not suggesting that this is a part of the Toyota problem. I'm merely suggesting that your absolute generalization is generally not absolutely correct.
Even my 1995 BMW had the shift-lever connected electronically, and it wasn't* a particularly fancy transmission: Just a GM 4L30E. The shift lever was just a fancy electrical switch, despite feeling chunky and solid as you pushed it between gear selections.
If a Prius, modern as they're supposed to be, is any different in this regard I'll eat my hat.
*: Past-tense, because the car now has a Getrag 5-speed manual gearbox. This change was the only reason I even had an opportunity to see what the AT shift lever looked like...
BMW 325I 1995 THROTTLE OPERATES SLOWLY WHEN COLD. THIS CONDITION CAN CAUSE THE ACCELERATOR TO HOLD THE THROTTLE PARTIALLY OPEN, RESULTING IN LOSS OF VEHICLE CONTROL INCREASING THE RISK OF A VEHICLE ACCIDENT.
I made that up, of course -- AFAIK there's no NHTSA verbiage about the problem. But I just wanted to chime in and say throttle cables can have issues, too, and they can exist outside of the realm of NHTSA reports. I experience it every day that it's cold outside, every time I shift gears, until the radiant heat of the engine warms up the cable and/or return spring (and after that it works properly).
In my case, the car is old, has lots of miles on it, and is simply due for a cable replacement. Parts wear out, etc. It's no big deal to me (I know how to use the clutch pedal, the brakes are awesome, and I trust the rev limiter to keep the engine from eating itself if it came to that), so I'll just fix it sometime after it warms up outside.
For the less-initiated, I can easily see how issues like this could become a real and unexpected problem, however.
IIRC, that individual had already roasted the brakes in the course of several half-assed attempts at slowing the vehicle.
Stopping a car stuck at WOT once is easy: Mash the brake pedal hard, and stand on it until stopped while letting ABS sort out whatever difficulties that might entail. Eventually, shift into neutral and/or kill the engine whenever available brain power permits -- either before or after the car comes to a safe rest.
Stopping a car stuck at WOT using more than one half-assed attempt to slow down, as that individual tried, is hard: Mash the pedal, car slows down but doesn't stop as swiftly as normal. Panic. Release pedal, try again. Pump brakes. Act stupid. Dial 911. Try again. Etc.
Each time you do that, the brakes get warmer, until the brake fluid boils and pads turn to muck. In my experience, an average not-so-special passenger car only has a couple of 80-90MPH stops in quick succession before the brakes turn to goo. So, after sufficient fucking around with the brake pedal (instead of actually being stubborn and insistent on it stopping AT ONCE), the pedal gets soft and the brakes don't even appreciably slow the vehicle -- maybe not at all at WOT. After that, panic more. Add this new panic to the existing panic, and shifting to neutral might rationally be the last thing you'll be thinking of. (IIRC from the 911 recording, he started praying instead.)
Also, remember, some of these (most? all?) Toyotas had a funky push-button start configuration which required one to push and hold a button for some number of seconds before the engine would shut down while in gear. This point is obviously an RTFM sort of problem, but that doesn't mean that it's not a problem....
Can you name any files that might be more popular than those infringing the copyrights of the MPAA studios, the major porn studios, the big four record labels, or the major video game publishers?
And as for that use, $599 seems a tad much for a home theater remote control.
If that's the case, then you haven't yet met Crestron.
(Nope, I'm not a shill, though I did ostensibly have a job programming Crestron gear a decade or so ago, mostly for private home theater use. I'm just trying to convey the concept that a $599 for a control system isn't really so far-fetched, if the featureset is right. In fact, $599 might be cheap.)
Does audiogalaxy have clients for multiple platforms, a multi-platform server, an awesome web interface, and a GPL behind it?
I can't tell. All I see on audiogalaxy.com is a thing that says "Sign up! It's FREE!"
With subsonic, I don't have to sign up anywhere; I don't have to use subsonic's "services." I just toss the author a few bucks*, install the software, forward the requisite port on my router, and call it a day. Subsonic has its roots in an open-source web-based media player, dating to well before the day of the smartphone, whereas Audiogalaxy has its heritage rooted in a heavily-litigated P2P clusterfuck.
(*Note: Just because it's free software, doesn't mean that it's afraid of money.)
When I was a kid, we had an XT made by AT&T. It included a hefty set of manuals, on heavy stock, in 3-ring binders, with 5-sided boxes to keep them tidy on the shelf. There was an MS-DOS manual, a GW-BASIC manual, and a system manual -- the latter of which I never tore into much -- and maybe another book, too. (It's been a long time.)
It wasn't anything as voluminous as the set of VMS manuals that a girl I dated later had on her bookshelf, but it was plenty to keep my 8- or 9-year-old brain occupied as bedtime reading.
IIRC, the programming manual for my Soundblaster 1.5 was available separately, for a fee. The manual that was included with the card was pretty sparse by the standards of the day, but compared to now, was a volume of information.
Modems, at least, included real books. Nowadays, winmodems generally work just fine, but trying to find a list of supported AT commands is like picking ticks from the ass of a rhino.
Not all Dells are crapfested. Most of those that we buy at work have pretty lean installs, and we don't do anything particularly special when we buy them except order from Small Business instead of Home.
At home, my wife and I each have an Alienware box. Neither of those came with any extra crap -- there's a couple of little support-related widgets, and it is otherwise a clean install of Windows (plus or minus some drivers). (Alienware, of course, is owned by Dell.)
Meanwhile, Dell generally does at least ship an unfucked copy of Windows with their PCs, making a clean reinstall at least straight-forward and cheap, so your options 1 and 2 are a little blurry...
I have limited experience with the other big OEMs, simply because most of the Dell hardware I've had my hands on seems to last more-or-less forever, and I've therefore not had good a reason to explore other options.
My local municipality collects income tax. It's a simple tax: 1%. It usually fits onto a simple, one-page form. But there's still some data entry and calculations for exemptions and crap and so, like anything else more complicated than taking a leak, it could be improved.
For the 1999 tax year, they issued a PDF tax form that automagically did the simple math for me, just by filling out the values in Adobe Reader/Acrobat/X/whatever it was then.
It worked well. My brain already hurt from filing Federal and State taxes that evening, and it was refreshing and very surprising to have something easy to work with for the city by the time I got down to them. I filled in a couple of fields, printed the resultant form, dropped it off at the local tax office, and it was fast and simple and done.
For 2010, they use the same form, with a different tax rate (1.25%), but they dropped the code that does the math (they must have downsized the one clever person who figured it out for 2009). All it is for 2010 is a simple PDF that one can print and then fill out by hand.
I'll reiterate: By hand. It is dumb.
Yes, it's a corner-case. Yes, it would perhaps be better-solved with an online form on a web site (eliminating both paper and physical delivery). But yes, it was useful.
It's improper to say things like "nobody uses this," when some people do.
Now, if you ask me if such things belong in a Portable Document Format file to begin with, I guess I'd have to say "no." And if you ask me if it's a terrible burden to write this stuff out by hand in this instance, I'd also say "no." But it was handy, and it did work well.
On the other hand, I'm not aware of any other cross-platform semi-open system that can handle document layout, user entry on forms, and math, all from a single concise file.
We might be running short on IPV4 addresses, but I think we'll have plenty to support such a system as you describe for as long as it is necessary, once companies begin return their existing IPV4 subscriber blocks.
There's no compelling reason, long-term, for a home user to have an ISP connection that includes a routable IPV4 address, so there should be giant swaths of address space being returned to the pool once companies like Comcast are finished with them after transition.
I dare say that enough addresses will become available as IPV6 rolls out properly, that there will be sufficient quantities of IPV4 addresses for every public-facing server in the world for a long, long time.
Meanwhile, a poster in another thread has written a bit about NAT-PT, which would (in theory) solve the problem at the end-user's router, including the case of my NTP example.
It seems to me that it is an eventuality that we just won't need IPV4 on the public network at all.
Slashdot Mirror
Perhaps the market is just different.
My folks bought a Mercury Villager in 1993 that had ABS as a standard item.
My sister had an old (96?) Chevy Monte Carlo with standard ABS.
I dated a girl with a 1996 Grand Am once, which had standard ABS.
I married a girl who came with a 1996 Pontiac Firebird, which had standard ABS.
And, of course, the '95 BMW has standard ABS. Plus the aforementioned Beretta.
I've driven far more American cars from the 90's that had ABS as standard equipment, than cars that did not.
That all said: Holden is about as American as vegemite is.
I've actually had this conversation:
Me: You should check out the new GTO. It's pretty awesome.
Other: What's so cool about it?
Me: It's got a big engine, decent handling, it's rear-drive, and the car itself isn't so big. It's based off the Monaro.
Other: The what?
Me: Monaro. From Holden.
Other: What's a Holden?
It is obscure. You can keep saying it's not, but it nonetheless is. (You do the dictionary look-up on that word as homework.)
Combine the tenacity of something like Blaster with the fact that random generally isn't, and such software will land somewhere. Furthermore, I think you genuinely underestimate the number of folks downloading and running such niceties as "FREE Registry Cleaner 9000" and the "OMG PONIES!!!" screensaver, which allows a fair number of seed nodes out of the gate. (I made those names up. You get the point.)
And, of course: Nevermind the fact that such a routeable address will not exactly be secret to begin with: In the absence of NAT, whatever host(s) you communicate with will know this address, and it will no longer be obscure. Given enough datamining on a popular and compromised/ill-intended sites, and producing rather complete maps of an individual's home subnet should be practical.
And speaking practicality: You can have multiple addresses per host. Lots of them. And you can assign them "randomly," and change them periodically. You can multiple firewall rules concocted to mitigate risks individually on these many addresses, and do all manner of other fuckery and confusion (unique addresses for each IPV6 connection pair?) to try to keep things obscure.
But the obscurity will fail. It won't likely fail for you or me (for the same reason I don't plan on winning the lottery), and it won't ever fail for Bruce Schneier (because the force is strong with that one), but it will fail for someone.
So, if we're cannot rely on obscurity by itself, then we'll have to rely on firewalls. This is all well and good, but by the time we firewall the piss out of IPV6 at the border, we've lost most of the advantages of having all of those addresses to begin with.
At that point, one might as well just use NAT and dumb port forwarding anyway, maybe with a couple of extra hooks added to let multicast work and/or some sort of automatia similar to UPnP for those who can't be bothered.
In other news: I've got way more addresses than I can use on my current RFC 1918 subnet. I don't need IPV6 -- the Internet does. I will re-evaluate my position as the environment dictates.
Right. I know.
The thing is, the QoS stuff in Tomato actually works in a predictable, useful, and rather flexible fashion. The Linksys QoS stuff, meanwhile, is rather meh.
Further reading has also indicated that TomatoUSB (which I also have installed here, somewhere...) does support IPV6, and although the configuration is all manual at this point. It apparently has the beginning of proper GUI support in the Git repository, though, so it should be ready sometime...
And nothing of value was lost.
I think it sounds more like V-mail than Sneakernet.
Yes. And 40-bit SSL should be enough for anybody.
Er. Uh. I mean to say: "It's really, really obscure! So it must be safe!"
I can break down some of the numbers for you: linky link
Unique wifi networks in DB: 31,592,538
Unique networks w/ location: 30,576,668
Unique wifi locations in DB: 1,211,718,307
Unique cell towers in DB: 25,696
Networks with crypto: 16,194,355 (51.2%)
Networks without crypto: 8,295,965 (26.2%)
Networks crypto unknown: 7,102,218 (22.4%)
Networks with default SSID: 3,181,785 (10.0%)
New unique networks today: 35,224
New today with location: 35,221
New yesterday with location: 38,447
By manufacturer:
Linksys 2846742 9.010%
D-Link 1365841 4.323%
Cisco 1225600 3.879%
Dell 909165 2.877%
Netgear 849644 2.689%
Belkin 486015 1.538%
2wire 458674 1.451%
Symbol 322504 1.020%
Apple Computer 243631 0.771%
I like the WRT54GL. We've got a bunch of them scattered around keeping point-of-sale systems online, and they're absurdly reliable with stock firmware: We never need to fuck with them.
That said, I like them better at home when they're running Tomato, because the QoS features let folks in the house watch a couple of 4-6Mbps Netflix streams while the wife plays WOW and I download torrents and the brother-in-law does whatever he does when his door is shut, all while keeping latency low enough that SSH stays non-annoying.
This, to me, makes it worth messing with.
Granted, Tomato doesn't really do IPV6 yet, either, but it's far more likely to support it on old Linksys product than Linksys's own firmware when the time comes that it really becomes useful and wanted. (We're not there yet, despite the Slashdot hype.)
On my car(s), it's a separate system.
The 1995 BMW, for instance, has a braking system made by Ate. From the rotors to the pads to the calipers to the hoses to the master cylinder to the electronics, it's all Ate, and is almost (!) independent from the rest of the car. (In this particular example, the ABS system has the ability to also close a secondary throttle to reduce engine output, though it has no ability to open the main throttle to increase it...)
My 2002 GMC van has a system from, IIRC, Bosch. It's the same sort of gig, minus the extra throttle body: All separate and independent.
At any rate, a failure of one system should not impart a failure of the ABS system, at least in these examples. I don't have anything newer to take apart and peer at, however -- I suppose it's possible, or even rather likely, that newer systems are more integrated and might be more susceptible to cascading failure.
I disagree with the notion that ABS is unusual. It's been a pretty standard item on most GM cars for over a decade, and even the not-special-at-all 1995 Chevy Beretta I used to have was equipped with ABS.
Oh, right. You guys get more efficient engine tech, and we get cleaner air.
I always try to forget that, since it seems like we've got the short end of the stick.
You asked. I answer. Pull up a chair, lad, it's story-telling time:
Once upon a time, there was a company called 3dfx. They made video cards, the first of their kind, with all kinds of gee-whiz video acceleration.
Now, along comes nVidia. Their first few 3D chipsets sucked, but later on, they grew up and made real competitors to the 3dfx line.
Eventually, 3dfx was bought by nVidia.
And, it just happened that the weekend after this, I was at a small LAN party with my PC.
3dfx's website was gone. No 404, no response, no nothing. nVidia didn't mirror anything. And I, of course, needed drivers for my 3dfx Voodoo3 2000.
Every. Single. Fucking. Link. From. Every. Single. Fucking. Web page. was a dead link to 3dfx's defunct site. So, there simply weren't any drivers to be found. I didn't even know, yet, that 3dfx had been bought out: I failed miserably at playing games that weekend, despite my keen sense of Internet-fu and the help of the fellow geeks around me.
Some time later, an ad-hoc support structure of hobbyists began providing 3dfx drivers, but that was far too late.
In addition, 3dfx's sudden demise abolished what was left of STB's drivers and documentation. See, sometime before all of the above, 3dfx bought STB, who used to make a variety of different add-on cards for PCs, including video adapters and other stuff.
I have here, somewhere, an STB I/O card which is completely inoperable. Why? While 3dfx did keep STB's old support site alive, nVidia flushed the whole lot. This, despite written, personal assurance from a pre-3dfx STB that they'd keep the old information online indefinitely. The information is simply lost to time, which is plainly fucking retarded.
And don't get me started on the X-Fi card in my desktop. It's an OEM variant that, while quite good and versatile, was never sold separately. Instead, it only came in (IIRC) some Dell, Alienware, and HP machines -- I haven't even seen it for sale in the white-box OEM market.
Creative Labs, in their infinite wisdom, will not let you install software items that have been downloaded from their website unless you already have that software installed (can you spot the glaringly-obvious Catch-22?). Alienware, in their own infinite wisdom, packaged a bad driver disk with the system, wherein the X-Fi drivers and software were both broken and did not match those that were pre-installed with the default load of Windows. And both Dell and HP seem to ignore the fact that the card exists at all when it comes to software support.
Now, Alienware does have 32-bit drivers, with a complete and installable suite of software, available for download on their website (unlike Creative Labs). And, lo, they do work. It's a few hundred megs of shit, though, and it's difficult to find (and Google isn't much help except to find other people looking for the same thing, but failing to find it), and much of it doesn't work on a 64-bit system.
I eventually found and downloaded this, and made it all work on my desktop Alienware box after upgrading to 64-bit Windows. And things are good, the Creative Labs update program works well and automatically downloaded and upgraded stuff, and I'm able to stay current with things.
But: Alienware's buried download is the only place the drivers are available from. If/when it vanishes (and it's not so much if as it is when -- see above), it will be gone. I've burned them to CD, of course, but I'm by no means the only person with this card.
So, while currently functional, it's a fragile situation. The next installation of Windows may not be so easy.
Accordingly, having learned from my previous STB/3dfx experiences, I tried to seed the drivers on The Pirate Bay, hoping that TPB's high Pagerank would let others find them easily. I included a good description of how to make them work, what card they were for, and why I was posting it, but they banned my account and nuked the .torrent after a
You folks in the UK have a manual gearbox for your version of the Prius?
Do tell. :)
I, for one, agree almost completely.
However, I think drivers should be informed of the function of the switch as intuitively as possible (who honestly R's the entire FM on a car except for my wife?). It also needs to be designed in such a way as to prevent accidental shutdown, so that the car doesn't shut down unexpectedly on a busy highway (which has its own safety problems).
Here's my idea for making it work:
Instead of a stop/start button with push-and-hold functionality, like we have now, plus an e-stop, as you suggest, let's do the following:
First, make it a toggle switch, with obvious visual and tactile indications of "go" and "don't go."
Second, cover it with something spring-loaded. (There's a wealth of military and industrial design knowledge covering this that I needn't go into.)
Third, eliminate the delay: Hit the switch, engine dies. End of story.
Fourth, put it central-ish on the dash, so that it could be reached by a passenger in the event that everything goes wrong and the driver is already busy dealing with that (or goes into insulin shock, or whatever).
Fifth, ensure that in the event that the e-stop is activated on accident, that putting the switch back into the "go" position will automatically restart the engine and resume normal operation.
Thus, starting the vehicle normally will consist of the following: Open cover, activate switch.
Shutting off the vehicle, whether in motion or parked, will also normally be as follows: Open cover, deactivate switch.
In this way, the operator is already familiar with the operation of the switch, because they have to use it every time they drive. Meanwhile, unintended operation of the switch (by bumping it somehow, a kid being an idiot, or whatever) is precluded by the cover, and is safely reversible in an understood fashion.
It's a little less convenient than the unprotected "Start" button in question here, but I don't think that it is unduly so, and it would make it both plain and obvious how to shut down the car if the need were to arise.
Instructions on the switch's exact behavior can be printed on the sun visor right next to the stupid airbag warnings, which should eventually let the complete idea of the concept be absorbed by the populace in general.
(I'd patent this, but I can't afford to. Consider it prior art.)
One word: ABS.
Give the pedal a good hard push on ice with bad tires, and you'll quickly see that there most assuredly is a software method to prevent the brakes from being applied.
Anecdotally, I've also experienced reduced braking power on more than one vehicle with malfunctioning ABS due to wiring faults and/or bad wheel sensors: If the system disables itself as it is supposed to, this isn't a big problem. If it has not yet detected the fault (and, mind you, it resets the fault indicator every time you start the vehicle), stopping even on dry pavement can rather more interesting than anticipated.
I'm not suggesting that this is a part of the Toyota problem. I'm merely suggesting that your absolute generalization is generally not absolutely correct.
Aye.
Even my 1995 BMW had the shift-lever connected electronically, and it wasn't* a particularly fancy transmission: Just a GM 4L30E. The shift lever was just a fancy electrical switch, despite feeling chunky and solid as you pushed it between gear selections.
If a Prius, modern as they're supposed to be, is any different in this regard I'll eat my hat.
*: Past-tense, because the car now has a Getrag 5-speed manual gearbox. This change was the only reason I even had an opportunity to see what the AT shift lever looked like...
BMW 325I 1995 THROTTLE OPERATES SLOWLY WHEN COLD. THIS CONDITION CAN CAUSE THE ACCELERATOR TO HOLD THE THROTTLE PARTIALLY OPEN, RESULTING IN LOSS OF VEHICLE CONTROL INCREASING THE RISK OF A VEHICLE ACCIDENT.
I made that up, of course -- AFAIK there's no NHTSA verbiage about the problem. But I just wanted to chime in and say throttle cables can have issues, too, and they can exist outside of the realm of NHTSA reports. I experience it every day that it's cold outside, every time I shift gears, until the radiant heat of the engine warms up the cable and/or return spring (and after that it works properly).
In my case, the car is old, has lots of miles on it, and is simply due for a cable replacement. Parts wear out, etc. It's no big deal to me (I know how to use the clutch pedal, the brakes are awesome, and I trust the rev limiter to keep the engine from eating itself if it came to that), so I'll just fix it sometime after it warms up outside.
For the less-initiated, I can easily see how issues like this could become a real and unexpected problem, however.
IIRC, that individual had already roasted the brakes in the course of several half-assed attempts at slowing the vehicle.
Stopping a car stuck at WOT once is easy: Mash the brake pedal hard, and stand on it until stopped while letting ABS sort out whatever difficulties that might entail. Eventually, shift into neutral and/or kill the engine whenever available brain power permits -- either before or after the car comes to a safe rest.
Stopping a car stuck at WOT using more than one half-assed attempt to slow down, as that individual tried, is hard: Mash the pedal, car slows down but doesn't stop as swiftly as normal. Panic. Release pedal, try again. Pump brakes. Act stupid. Dial 911. Try again. Etc.
Each time you do that, the brakes get warmer, until the brake fluid boils and pads turn to muck. In my experience, an average not-so-special passenger car only has a couple of 80-90MPH stops in quick succession before the brakes turn to goo. So, after sufficient fucking around with the brake pedal (instead of actually being stubborn and insistent on it stopping AT ONCE), the pedal gets soft and the brakes don't even appreciably slow the vehicle -- maybe not at all at WOT. After that, panic more. Add this new panic to the existing panic, and shifting to neutral might rationally be the last thing you'll be thinking of. (IIRC from the 911 recording, he started praying instead.)
Also, remember, some of these (most? all?) Toyotas had a funky push-button start configuration which required one to push and hold a button for some number of seconds before the engine would shut down while in gear. This point is obviously an RTFM sort of problem, but that doesn't mean that it's not a problem....
Perhaps I can.
It really doesn't seem to be so hard.
I mean, really: Did you even stop to look around at the world before you wrote that?
Because, frankly, I think the concept that you're attempting to quantify is bullshit.
Really, I do.
Peace.
If that's the case, then you haven't yet met Crestron.
(Nope, I'm not a shill, though I did ostensibly have a job programming Crestron gear a decade or so ago, mostly for private home theater use. I'm just trying to convey the concept that a $599 for a control system isn't really so far-fetched, if the featureset is right. In fact, $599 might be cheap.)
Are you suggesting that "podcast" be deprecated and replaced with "broadcast"?
If so I, for one, am all for it.
Does audiogalaxy have clients for multiple platforms, a multi-platform server, an awesome web interface, and a GPL behind it?
I can't tell. All I see on audiogalaxy.com is a thing that says "Sign up! It's FREE!"
With subsonic, I don't have to sign up anywhere; I don't have to use subsonic's "services." I just toss the author a few bucks*, install the software, forward the requisite port on my router, and call it a day. Subsonic has its roots in an open-source web-based media player, dating to well before the day of the smartphone, whereas Audiogalaxy has its heritage rooted in a heavily-litigated P2P clusterfuck.
(*Note: Just because it's free software, doesn't mean that it's afraid of money.)
When I was a kid, we had an XT made by AT&T. It included a hefty set of manuals, on heavy stock, in 3-ring binders, with 5-sided boxes to keep them tidy on the shelf. There was an MS-DOS manual, a GW-BASIC manual, and a system manual -- the latter of which I never tore into much -- and maybe another book, too. (It's been a long time.)
It wasn't anything as voluminous as the set of VMS manuals that a girl I dated later had on her bookshelf, but it was plenty to keep my 8- or 9-year-old brain occupied as bedtime reading.
IIRC, the programming manual for my Soundblaster 1.5 was available separately, for a fee. The manual that was included with the card was pretty sparse by the standards of the day, but compared to now, was a volume of information.
Modems, at least, included real books. Nowadays, winmodems generally work just fine, but trying to find a list of supported AT commands is like picking ticks from the ass of a rhino.
Not all Dells are crapfested. Most of those that we buy at work have pretty lean installs, and we don't do anything particularly special when we buy them except order from Small Business instead of Home.
At home, my wife and I each have an Alienware box. Neither of those came with any extra crap -- there's a couple of little support-related widgets, and it is otherwise a clean install of Windows (plus or minus some drivers). (Alienware, of course, is owned by Dell.)
Meanwhile, Dell generally does at least ship an unfucked copy of Windows with their PCs, making a clean reinstall at least straight-forward and cheap, so your options 1 and 2 are a little blurry...
I have limited experience with the other big OEMs, simply because most of the Dell hardware I've had my hands on seems to last more-or-less forever, and I've therefore not had good a reason to explore other options.
My local municipality collects income tax. It's a simple tax: 1%. It usually fits onto a simple, one-page form. But there's still some data entry and calculations for exemptions and crap and so, like anything else more complicated than taking a leak, it could be improved.
For the 1999 tax year, they issued a PDF tax form that automagically did the simple math for me, just by filling out the values in Adobe Reader/Acrobat/X/whatever it was then.
It worked well. My brain already hurt from filing Federal and State taxes that evening, and it was refreshing and very surprising to have something easy to work with for the city by the time I got down to them. I filled in a couple of fields, printed the resultant form, dropped it off at the local tax office, and it was fast and simple and done.
For 2010, they use the same form, with a different tax rate (1.25%), but they dropped the code that does the math (they must have downsized the one clever person who figured it out for 2009). All it is for 2010 is a simple PDF that one can print and then fill out by hand.
I'll reiterate: By hand. It is dumb.
Yes, it's a corner-case. Yes, it would perhaps be better-solved with an online form on a web site (eliminating both paper and physical delivery). But yes, it was useful.
It's improper to say things like "nobody uses this," when some people do.
Now, if you ask me if such things belong in a Portable Document Format file to begin with, I guess I'd have to say "no." And if you ask me if it's a terrible burden to write this stuff out by hand in this instance, I'd also say "no." But it was handy, and it did work well.
On the other hand, I'm not aware of any other cross-platform semi-open system that can handle document layout, user entry on forms, and math, all from a single concise file.
Good point.
We might be running short on IPV4 addresses, but I think we'll have plenty to support such a system as you describe for as long as it is necessary, once companies begin return their existing IPV4 subscriber blocks.
There's no compelling reason, long-term, for a home user to have an ISP connection that includes a routable IPV4 address, so there should be giant swaths of address space being returned to the pool once companies like Comcast are finished with them after transition.
I dare say that enough addresses will become available as IPV6 rolls out properly, that there will be sufficient quantities of IPV4 addresses for every public-facing server in the world for a long, long time.
Meanwhile, a poster in another thread has written a bit about NAT-PT, which would (in theory) solve the problem at the end-user's router, including the case of my NTP example.
It seems to me that it is an eventuality that we just won't need IPV4 on the public network at all.