I was in court the other day. The judge had a computer on her desk. In a window, in the middle of the desktop, was an 80x25 textmode Wordperfect 5.1.
I don't use it because I've never done anything serious with a wordprocessor, and Open Office does just fine for me. The legal world, however, is still stuck on WP 5.1, because it absolutely fucking works.
I've had Linux installs fall flat on their face after a few years of not being kept current followed by an attempt to upgrade the system to current-spec code.
Generally, I can revive these systems, given enough time. But lately I find it's easier to start them over with a fresh install.
It's possible, though obviously insane, to do the following:
Install Windows 3.1. Upgrade to Windows 95. Upgrade to Windows 98SE. Upgrade to Windows ME. Upgrade to Windows XP. Upgrade to Windows Vista. Upgrade to Windows 7.
But that's just retarded, as anyone here (with their blinders off) should be able to recognize. Real men don't upgrade operating systems -- they just buy the upgrade kit (because it's cheaper), and Google a good method for doing a clean install with it.
In my own experience: I bought a laptop with XP, almost four years ago. It's a good machine, and was pretty quick at the time. I kept XP around because most of the stuff I need for my daily work needs Windows, though I'd really be a lot more pleased to see Slackware, Gentoo, or FreeBSD on the machine. When Vista was released, I upgraded (er - I did a fresh install). Vista worked fine, though I also doubled the RAM to 2 gigs and gave it a bigger, faster hard drive at the same time. A year or two later, the hard drive crashed -- probably from being used too much outside in sub-zero Ohio winters. I had a choice: Reinstall the backup DVD of a fresh, clean, working Vista install in less than an hour, or download Windows 7.
I decided to see what Windows 7 was all about.
Spent a day or so shoving my usual software back into the machine, and it all works fine. The box suspends, hibernates, and resumes faster than it ever did with XP or Vista, both of which had occasional issues. Performance once it is running is good. I haven't been tempted -- yet -- to disable the Readyboost service, as I often do on Vista machines to improve speed. It bluescreened exactly one time, while copying files from the old (crashed, littered with bad sectors) hard drive over a cheap IDE-USB adapter, and I don't think it's been rebooted since (aside from updates).
It just works. So far. I hate Windows, but 7 seems to be OK.
Where I work, I spec'd the mail server to have lots of storage.
It gets backed up daily, offsite, with an automated rsync script. If the backup takes hours, that's fine, as long as it's fewer than 24 of them.
I encourage (and sometimes force) people to use IMAP instead of POP3, so that their mail is all on the server, and thus backed up.
I coach people to save things in folders in their mail client, instead of downloading them to their desktop PC (which may or may not be company-owned), because that way I can make sure that backups of their stuff actually fucking happen.
There's lots of internal bandwidth -- even though it's just a switched 100mbps network. There's plenty of external upstream to keep remote IMAP users happy.
The mail server (postfix+amavis+dovecot) Just Works(tm) . Years of uptime transpire between planned reboots for maintenance and kernel security fixes.
Several years ago, I increased the maximum size of an email from whatever Postfix's default was, to something like 25 megabytes. I have yet to see any detrimental effect in this.
Am I really alone? I mean, sure, my site is smaller than a lot of them, but if it's affordable for my several-dozen users, then similar design principals ought also to be affordable for places with hundreds or thousands of users. For fuck's sake: Google does stuff like this for millions of users.
I guess the point is this: I often use my phone in the car, with it on a mini-USB charger. This often involves putting the phone to my ear, for quick calls where I don't want to goof around with powering up a Bluetooth headset. The coiled charge cord does create a fair bit of connector stress as I stretch it out to my driving position. Usually, it relieves this stress by unplugging itself -- which isn't a very good solution.
Throwing an additional connector on there gives the stuff even less opportunity for stress relief. It doesn't really matter, to me, if it's 2.5mm or 3.5mm. These are just my opinions as a wire and connector snob.:) (And if I had my way about it, every modern connector would be round, and resemble either a coaxial DC plug, or some variation on the 2.5mm/3.5mm/.25 inch TRS standard. Every. Single. One. No more flashlights and cramped spaces when plugging in connectors on the back of gear.)
I didn't realize that iPod's have three different sets of contacts for different voltages. But looking at the the pinouts, it really is a stupid design. The S-video thing is just as stupid -- AFAIK, one doesn't even need a low-pass filter to sum the two signals together with good quality.
But just accepting different voltages, I consider that to be a good thing. I like being able to shove random voltages into devices without them blowing up, and I'm happy to pay extra for that.
You're describing a connection standard which is very expensive, which if ever it is adopted by a couple of manufacturers, will be untenable in the market and will fail. It's also ugly. And it has issues with stress: A 3.5mm connector can rotate freely, relieving axial stress along the cable, or tensional stress if a right-angle-connector is used, but is otherwise rigid. An iPod connector has a lot of flex in one dimension. A USB connection is pretty non-versatile -- it has no mechanism for stress relief. Any combination of these will be very inflexible, and more likely to break.
That said, I only personally know two iPods: My iPod Touch, and my wife's second-gen Nano. Both accept anything from 5V to 30VDC for charging, according to the etching on the backs of the units. I charge them frequently from USB ports, but I also have an aftermarket charger which outputs 10.5V -- both work fine.
If you think you know differently, then please be more elaborate in your conjecture.
On this side of the pond, we use things called "credit cards," but they're not generally actually credit cards: They're debit cards. It's just an easy way to suck some money directly from one's checking account, without carrying cash.
However, network-wise, they show up just like any other Mastercard or Visa, with all of the vendor fees associated with that.
The value added is that I don't have to carry cash. I don't have to write checks. And, lately, it's often faster than cash. There's a place near my house that I buy coffee from some mornings, and I can slide my Visa Debit Card through the reader, get a receipt, and be on my way faster than I can come up with the correct change (or wait for change to be made for me). Over on your side of the pond, I'm sure you've got roughly equivalent tech, between EBT and smart cards.
And then there's actual credit cards. We abuse the hell out of those, but they have some good purposes. Think something along the lines of being broke until the end of the month, with a dead transmission in the car, a 30 mile commute, and no public transportation. Paying for a tranny rebuild on a credit card is expensive but can be both necessary and far cheaper than finding a new job.
I've created a few clever things, also, that the world may never get to see because of the difficulty of obtaining a patent.
Everyone's all about patent reform these days: So, let's get to it. Reform patents so that small inventors can realistically afford to patent their creations again.
I build the one with coathangers, screws, some old paneling, aluminum foil, and a 2x4 just after I got my 52" LCD home and was feeling dispondant about its weak tuner. I had this stuff laying around, and was able to assemble it with a power screwdriver and a pneumatic stapler in about 30 minutes. For feedline, I used a couple of scraps of wire from bulb cat5 to feed a 300-75 Ohm transformer, and connected that with a dozen or so feet of RG-6 into the TV. I secured it to a piece of scrap 1x8 with a single screw for a base, allowing it to rotate without beating up the table too horribly.
It worked fine. No dropouts or weirdness. It was very directional, which is generally good, but did require adjustment to get different stations. On the other hand, I was using it on a table in the front, ground-level room of my house, about four feet above the ground. It'd have certainly worked better if it had some real elevation, or if it were outside.
My wife threw it away after we got HD cable. And now that we have Uverse instead (which rocks), I need to build another one -- Uverse lacks one of the local PBS stations that I like to watch, which this antenna picked up fine.
But again, it was basically free, and I've got materials... It's a good excuse to buy a USB HDTV tuner, plug the antenna (aimed at the PBS tower) into it, and see just how difficult it is to make it all work with my MythTV on my Gentoo box.
But I don't want "Dad, I'll be home late from school today," or "How's Mom doing?" to be part of the national archives. It's none of anyone's business.
Having separate phones for separate tasks takes care of this neatly.
Nobody else replied to you, so I figure I might as well give it a shot:
I'd guess that (typically) a Windows box with wired and wireless connections bridged together could cause issues if both interfaces are connected to the same network. It's just a guess, though - I've never tried it.
I have, however, definitely connected to my own network with both wired and wireless connections. Why? Well, I wanted more speed than 802.11g offers so I could copy some big files faster, and I didn't want to dick around with turning off WLAN. So I plugged it in, Windows did its magic detection boojigity, the laptop started seeing the wired interface as the default one, and all was well. I've done this a few times, actually, and it's always worked fine -- even with dumb switches.
I used to keep pet fish. Not just any fish - I had an elaborate saltwater setup with lots of invertebrates, and another freshwater tank with a couple of big Oscars. The Oscars, in particular, were very friendly and moody, almost person-like in their actions. To fail to bond with an Oscar is to fail to be human - they're like dogs in their emotional tenacity.
On the other hand, I like to catch fish. I also like to eat them. I don't see this as a conflict at all.
I also currently have a domestic pet rat. The largest cat in the house would love to eat it, and has tried to on more than one occasion when the rat's been out of his cage with a certain ferocity that I've only ever seen on the Discovery channel. The cat goes straight for the neck. I don't let these altercations persist with my pet rat. But if I had an infestation of wild rats in my house, why sure, I'd kill them all.
For that matter, as much as I like housecats as pets, I do wonder how they'd taste.
"Today, about 25 percent of BitTorrent traffic is encrypted," says Schulze.
So. What he really means to say is this: By allowing ONLY encrypted traffic with your BitTorrent client, you'll still be able to accomplish connections with at least about 25% of available peers.
This is, I'd guess, quite good enough. I think I'll go adjust Azureus now.
Perhaps you're right and I'm misremembering the details, but I certainly recall the battle. This was a long, long time ago.
Might be that the method I was using involved l3dec instead of mpg123, which came later and was vastly more efficient. I still think I've spent some quality time tweaking buffering with mpg123, which (according to the changelog) didn't have output buffering until 0.56.
And the point is the same: It wasn't always easy. And it's not always easy, now.
Eh? I'm afraid you don't fully understand how a shock absorber works. I'll leave that as an effort for you to figure out on your own. But, in general practice, stiffer shocks are used on performance cars (more damping) and softer shocks on bigger, more luxury cars (less damping). They also get softer as they wear out.
I used to have a 1985 Buick LeSabre with lousy, worn out shocks, and passengers praised the quality of the ride, which was rather comfortable. But it was a terrifying vehicle to drive, with the whole thing rocking like a boat as it went down the road in an uncontrolled fashion and the tires being loaded unpredictably in turns, because the tired old shocks weren't eating as much energy as they should've been.
Current shock absorbers are 0% efficient -- all of the energy they consume is converted to heat. So, shocks get hot. 100% of this energy comes, eventually, from the burning of gasoline.
If they were instead 10% efficient at producing electricity, they'd be (gasp) 10% more efficient than they are now. It doesn't matter that 10% efficiency is horrible - it's still less horrible than 0%.
And let's not oversimplify things, either. There's no compelling reason to believe that this is a simple system with no smarts about it at all. It seems to me that there is ample opportunity to have a controller between the shock absorbing generator and whatever load is attached, and that it would be a straight-forward process to manipulate the loading of the shock in response to road conditions or driver preference.
TFA talks about a smoother ride -- I see this as very practical and straight-forward. The system can simply load the hell out the shock generators on relatively smooth surfaces (stiffer shocks) and decrease loading dynamically as conditions merit (softer shocks for potholes). There are conventional shock absorbers in common use which can do some of this by way of having special valving which reduces damping in response to sudden suspension movements.
Fancy shock absorbers on high-end cars (think 7-series BMW) use a special oil which changes viscosity in response to an electrically-generated magnetic field. These of course have negative efficiency. Turning this same functionality into something which can produce energy instead of merely consume it (or which perhaps does a little of both, depending on the situation) might not ever pay for itself, but it's worth investigating to see if it can.
There's a lot of electrical accessories in a modern car which use a significant amount of energy. Blowers, fans, defrosters, stereos, lighting, heated seats, wipers, so on, so forth. It'd be nice if they were more efficient to operate, wouldn't it?
(I also understand back EMF, and fail to see how it might produce any ill effects that a diode bridge cannot resolve.)
Indeed. Playon is non-free, but the cost is not a real issue. The quality is just fine, even on my well-adjusted 52" Samsung LCD, for a streaming movie that I did not have to download in advance or wait for it to show up in the mail. It's a lot softer than a DVD, but blocky and frame aberrations artifacts are generally not large issues. And, despite being anal about audio and video quality, I find that within a few minutes of watching a streaming Netflix film via Playon and a PS3, that I forget that I'm doing so, and just watch the movie.
It also streams Hulu just fine.
I'm not afraid of spending money on software that actually works, yaknow? And the same registered copy of Playon will also concurrently stream (with a little more CPU, but CPU is bloody cheap right now) to my brother in law's Xbox 360 upstairs. It just works. It might even work under Wine, but I don't even think it matters anymore: Any geek who plays games (and therefore has a PS3 or 360) probably has a relatively stoutly-configured gaming machine running Windows on the network, and that machine will run Playon. If time is money, then Playon is cheap.
So, yeah: Me, too. I keep watching these folks fumble around with streaming media, thinking they're on the bleeding edge, when I've been doing this stuff for a long time.
Which is not to say that I don't like my Linux machines; I just don't like transcoding streaming media with them because it's inconvenient and, at best, barely works.
I distinctly remember ugly hacks to get MP3s to play smoothly and reliably on my (absurdly stable and still running) P120 Linux box. It went something like this:
nice -n-10 (mpg123 "hello i am an mp3.mp3" | bplay -b 4096 -)
XMMS (or whatever it was named back then) didn't work much at all -- the box didn't have enough oomph. Winamp, under Windows, wasn't very reliable. It took a Gods-small and efficient mp3 player, at a real command line, without X running, along with a program designed specifically only to buffer audio, for it to function reliably.
It chewed up more than 90% CPU according to top. And yes, I was pleased with that -- it worked.
Nowadays, I get occasional skips under Windows on my 2.4GHz quad-core Q6600 box. And similar skips and strangeness on my Athlon XP 1900+ Linux box. Both of which, one would think, would be adequate to play a fucking mp3 without hacks and tweakage. *sigh*
With just a little bit of spare CPU on a Windows box, my PS3 has wife-compatible Hulu and Netflix. And it's also a wife-capable CD player. And a wife-capable DVD player. And a wife-capable Blu-Ray player. And even the wife can figure out how to stream music and shows from the Mediatomb box onto the PS3. I understand you can also play games on a PS3, and I'm sure nobody believes me, but even my wife has been seen doing so.
It's not currently doing any OTA stuff, but I don't have any hardware to support it just yet . . .
Slashdot Mirror
I was in court the other day. The judge had a computer on her desk. In a window, in the middle of the desktop, was an 80x25 textmode Wordperfect 5.1.
I don't use it because I've never done anything serious with a wordprocessor, and Open Office does just fine for me. The legal world, however, is still stuck on WP 5.1, because it absolutely fucking works.
I've had Linux installs fall flat on their face after a few years of not being kept current followed by an attempt to upgrade the system to current-spec code.
Generally, I can revive these systems, given enough time. But lately I find it's easier to start them over with a fresh install.
It's possible, though obviously insane, to do the following:
Install Windows 3.1. Upgrade to Windows 95. Upgrade to Windows 98SE. Upgrade to Windows ME. Upgrade to Windows XP. Upgrade to Windows Vista. Upgrade to Windows 7.
But that's just retarded, as anyone here (with their blinders off) should be able to recognize. Real men don't upgrade operating systems -- they just buy the upgrade kit (because it's cheaper), and Google a good method for doing a clean install with it.
In my own experience: I bought a laptop with XP, almost four years ago. It's a good machine, and was pretty quick at the time. I kept XP around because most of the stuff I need for my daily work needs Windows, though I'd really be a lot more pleased to see Slackware, Gentoo, or FreeBSD on the machine. When Vista was released, I upgraded (er - I did a fresh install). Vista worked fine, though I also doubled the RAM to 2 gigs and gave it a bigger, faster hard drive at the same time. A year or two later, the hard drive crashed -- probably from being used too much outside in sub-zero Ohio winters. I had a choice: Reinstall the backup DVD of a fresh, clean, working Vista install in less than an hour, or download Windows 7.
I decided to see what Windows 7 was all about.
Spent a day or so shoving my usual software back into the machine, and it all works fine. The box suspends, hibernates, and resumes faster than it ever did with XP or Vista, both of which had occasional issues. Performance once it is running is good. I haven't been tempted -- yet -- to disable the Readyboost service, as I often do on Vista machines to improve speed. It bluescreened exactly one time, while copying files from the old (crashed, littered with bad sectors) hard drive over a cheap IDE-USB adapter, and I don't think it's been rebooted since (aside from updates).
It just works. So far. I hate Windows, but 7 seems to be OK.
Forgive me, but:
Where I work, I spec'd the mail server to have lots of storage.
It gets backed up daily, offsite, with an automated rsync script. If the backup takes hours, that's fine, as long as it's fewer than 24 of them.
I encourage (and sometimes force) people to use IMAP instead of POP3, so that their mail is all on the server, and thus backed up.
I coach people to save things in folders in their mail client, instead of downloading them to their desktop PC (which may or may not be company-owned), because that way I can make sure that backups of their stuff actually fucking happen.
There's lots of internal bandwidth -- even though it's just a switched 100mbps network. There's plenty of external upstream to keep remote IMAP users happy.
The mail server (postfix+amavis+dovecot) Just Works(tm) . Years of uptime transpire between planned reboots for maintenance and kernel security fixes.
Several years ago, I increased the maximum size of an email from whatever Postfix's default was, to something like 25 megabytes. I have yet to see any detrimental effect in this.
Am I really alone? I mean, sure, my site is smaller than a lot of them, but if it's affordable for my several-dozen users, then similar design principals ought also to be affordable for places with hundreds or thousands of users. For fuck's sake: Google does stuff like this for millions of users.
*shrug*
Cthulhu. 'Nuff said.
FATALITY
I guess the point is this: I often use my phone in the car, with it on a mini-USB charger. This often involves putting the phone to my ear, for quick calls where I don't want to goof around with powering up a Bluetooth headset. The coiled charge cord does create a fair bit of connector stress as I stretch it out to my driving position. Usually, it relieves this stress by unplugging itself -- which isn't a very good solution.
Throwing an additional connector on there gives the stuff even less opportunity for stress relief. It doesn't really matter, to me, if it's 2.5mm or 3.5mm. These are just my opinions as a wire and connector snob. :) (And if I had my way about it, every modern connector would be round, and resemble either a coaxial DC plug, or some variation on the 2.5mm/3.5mm/.25 inch TRS standard. Every. Single. One. No more flashlights and cramped spaces when plugging in connectors on the back of gear.)
I didn't realize that iPod's have three different sets of contacts for different voltages. But looking at the the pinouts, it really is a stupid design. The S-video thing is just as stupid -- AFAIK, one doesn't even need a low-pass filter to sum the two signals together with good quality.
But just accepting different voltages, I consider that to be a good thing. I like being able to shove random voltages into devices without them blowing up, and I'm happy to pay extra for that.
You're describing a connection standard which is very expensive, which if ever it is adopted by a couple of manufacturers, will be untenable in the market and will fail. It's also ugly. And it has issues with stress: A 3.5mm connector can rotate freely, relieving axial stress along the cable, or tensional stress if a right-angle-connector is used, but is otherwise rigid. An iPod connector has a lot of flex in one dimension. A USB connection is pretty non-versatile -- it has no mechanism for stress relief. Any combination of these will be very inflexible, and more likely to break.
That said, I only personally know two iPods: My iPod Touch, and my wife's second-gen Nano. Both accept anything from 5V to 30VDC for charging, according to the etching on the backs of the units. I charge them frequently from USB ports, but I also have an aftermarket charger which outputs 10.5V -- both work fine.
If you think you know differently, then please be more elaborate in your conjecture.
On this side of the pond, we use things called "credit cards," but they're not generally actually credit cards: They're debit cards. It's just an easy way to suck some money directly from one's checking account, without carrying cash.
However, network-wise, they show up just like any other Mastercard or Visa, with all of the vendor fees associated with that.
The value added is that I don't have to carry cash. I don't have to write checks. And, lately, it's often faster than cash. There's a place near my house that I buy coffee from some mornings, and I can slide my Visa Debit Card through the reader, get a receipt, and be on my way faster than I can come up with the correct change (or wait for change to be made for me). Over on your side of the pond, I'm sure you've got roughly equivalent tech, between EBT and smart cards.
And then there's actual credit cards. We abuse the hell out of those, but they have some good purposes. Think something along the lines of being broke until the end of the month, with a dead transmission in the car, a 30 mile commute, and no public transportation. Paying for a tranny rebuild on a credit card is expensive but can be both necessary and far cheaper than finding a new job.
Yeah, yeah. Fine, fine.
I like Linux too.
But this sort of thing isn't new in the Windows world. Managing file ownership and permission hasn't changed much since NT 4.
It hasn't gotten worse since then. The problem is that it hasn't gotten better, either.
Roger that.
I've created a few clever things, also, that the world may never get to see because of the difficulty of obtaining a patent.
Everyone's all about patent reform these days: So, let's get to it. Reform patents so that small inventors can realistically afford to patent their creations again.
*shrug*
If I could plonk down $110.00, in advance, to legally have 1,000 digital songs of my choosing, I'd do so immediately and without hesitation.
Fine. You want to recycle my old beer cans?
They're over there, in a pile. At the landfill. Go get 'em.
I build the one with coathangers, screws, some old paneling, aluminum foil, and a 2x4 just after I got my 52" LCD home and was feeling dispondant about its weak tuner. I had this stuff laying around, and was able to assemble it with a power screwdriver and a pneumatic stapler in about 30 minutes. For feedline, I used a couple of scraps of wire from bulb cat5 to feed a 300-75 Ohm transformer, and connected that with a dozen or so feet of RG-6 into the TV. I secured it to a piece of scrap 1x8 with a single screw for a base, allowing it to rotate without beating up the table too horribly.
It worked fine. No dropouts or weirdness. It was very directional, which is generally good, but did require adjustment to get different stations. On the other hand, I was using it on a table in the front, ground-level room of my house, about four feet above the ground. It'd have certainly worked better if it had some real elevation, or if it were outside.
My wife threw it away after we got HD cable. And now that we have Uverse instead (which rocks), I need to build another one -- Uverse lacks one of the local PBS stations that I like to watch, which this antenna picked up fine.
But again, it was basically free, and I've got materials... It's a good excuse to buy a USB HDTV tuner, plug the antenna (aimed at the PBS tower) into it, and see just how difficult it is to make it all work with my MythTV on my Gentoo box.
A dry Martini, with a big, fat, oil-soaked, blue-cheese stuffed olive at the end.
That's what else.
But wait! There's more!
If you are successfully seduced by our satin-clad, big-titted spokesperson, you to can have Rosetta Stone! Just five easy payments of...
Fine, dandy.
But I don't want "Dad, I'll be home late from school today," or "How's Mom doing?" to be part of the national archives. It's none of anyone's business.
Having separate phones for separate tasks takes care of this neatly.
Nobody else replied to you, so I figure I might as well give it a shot:
I'd guess that (typically) a Windows box with wired and wireless connections bridged together could cause issues if both interfaces are connected to the same network. It's just a guess, though - I've never tried it.
I have, however, definitely connected to my own network with both wired and wireless connections. Why? Well, I wanted more speed than 802.11g offers so I could copy some big files faster, and I didn't want to dick around with turning off WLAN. So I plugged it in, Windows did its magic detection boojigity, the laptop started seeing the wired interface as the default one, and all was well. I've done this a few times, actually, and it's always worked fine -- even with dumb switches.
I used to keep pet fish. Not just any fish - I had an elaborate saltwater setup with lots of invertebrates, and another freshwater tank with a couple of big Oscars. The Oscars, in particular, were very friendly and moody, almost person-like in their actions. To fail to bond with an Oscar is to fail to be human - they're like dogs in their emotional tenacity.
On the other hand, I like to catch fish. I also like to eat them. I don't see this as a conflict at all.
I also currently have a domestic pet rat. The largest cat in the house would love to eat it, and has tried to on more than one occasion when the rat's been out of his cage with a certain ferocity that I've only ever seen on the Discovery channel. The cat goes straight for the neck. I don't let these altercations persist with my pet rat. But if I had an infestation of wild rats in my house, why sure, I'd kill them all.
For that matter, as much as I like housecats as pets, I do wonder how they'd taste.
I honestly don't see a conflict here.
"Today, about 25 percent of BitTorrent traffic is encrypted," says Schulze.
So. What he really means to say is this: By allowing ONLY encrypted traffic with your BitTorrent client, you'll still be able to accomplish connections with at least about 25% of available peers.
This is, I'd guess, quite good enough. I think I'll go adjust Azureus now.
Thanks for the tip!
That's what boots are for.
Perhaps you're right and I'm misremembering the details, but I certainly recall the battle. This was a long, long time ago.
Might be that the method I was using involved l3dec instead of mpg123, which came later and was vastly more efficient. I still think I've spent some quality time tweaking buffering with mpg123, which (according to the changelog) didn't have output buffering until 0.56.
And the point is the same: It wasn't always easy. And it's not always easy, now.
Eh? I'm afraid you don't fully understand how a shock absorber works. I'll leave that as an effort for you to figure out on your own. But, in general practice, stiffer shocks are used on performance cars (more damping) and softer shocks on bigger, more luxury cars (less damping). They also get softer as they wear out.
I used to have a 1985 Buick LeSabre with lousy, worn out shocks, and passengers praised the quality of the ride, which was rather comfortable. But it was a terrifying vehicle to drive, with the whole thing rocking like a boat as it went down the road in an uncontrolled fashion and the tires being loaded unpredictably in turns, because the tired old shocks weren't eating as much energy as they should've been.
Current shock absorbers are 0% efficient -- all of the energy they consume is converted to heat. So, shocks get hot. 100% of this energy comes, eventually, from the burning of gasoline.
If they were instead 10% efficient at producing electricity, they'd be (gasp) 10% more efficient than they are now. It doesn't matter that 10% efficiency is horrible - it's still less horrible than 0%.
And let's not oversimplify things, either. There's no compelling reason to believe that this is a simple system with no smarts about it at all. It seems to me that there is ample opportunity to have a controller between the shock absorbing generator and whatever load is attached, and that it would be a straight-forward process to manipulate the loading of the shock in response to road conditions or driver preference.
TFA talks about a smoother ride -- I see this as very practical and straight-forward. The system can simply load the hell out the shock generators on relatively smooth surfaces (stiffer shocks) and decrease loading dynamically as conditions merit (softer shocks for potholes). There are conventional shock absorbers in common use which can do some of this by way of having special valving which reduces damping in response to sudden suspension movements.
Fancy shock absorbers on high-end cars (think 7-series BMW) use a special oil which changes viscosity in response to an electrically-generated magnetic field. These of course have negative efficiency. Turning this same functionality into something which can produce energy instead of merely consume it (or which perhaps does a little of both, depending on the situation) might not ever pay for itself, but it's worth investigating to see if it can.
There's a lot of electrical accessories in a modern car which use a significant amount of energy. Blowers, fans, defrosters, stereos, lighting, heated seats, wipers, so on, so forth. It'd be nice if they were more efficient to operate, wouldn't it?
(I also understand back EMF, and fail to see how it might produce any ill effects that a diode bridge cannot resolve.)
Indeed. Playon is non-free, but the cost is not a real issue. The quality is just fine, even on my well-adjusted 52" Samsung LCD, for a streaming movie that I did not have to download in advance or wait for it to show up in the mail. It's a lot softer than a DVD, but blocky and frame aberrations artifacts are generally not large issues. And, despite being anal about audio and video quality, I find that within a few minutes of watching a streaming Netflix film via Playon and a PS3, that I forget that I'm doing so, and just watch the movie.
It also streams Hulu just fine.
I'm not afraid of spending money on software that actually works, yaknow? And the same registered copy of Playon will also concurrently stream (with a little more CPU, but CPU is bloody cheap right now) to my brother in law's Xbox 360 upstairs. It just works. It might even work under Wine, but I don't even think it matters anymore: Any geek who plays games (and therefore has a PS3 or 360) probably has a relatively stoutly-configured gaming machine running Windows on the network, and that machine will run Playon. If time is money, then Playon is cheap.
So, yeah: Me, too. I keep watching these folks fumble around with streaming media, thinking they're on the bleeding edge, when I've been doing this stuff for a long time.
Which is not to say that I don't like my Linux machines; I just don't like transcoding streaming media with them because it's inconvenient and, at best, barely works.
YMMV. :)
Hear, hear.
I distinctly remember ugly hacks to get MP3s to play smoothly and reliably on my (absurdly stable and still running) P120 Linux box. It went something like this:
nice -n-10 (mpg123 "hello i am an mp3.mp3" | bplay -b 4096 -)
XMMS (or whatever it was named back then) didn't work much at all -- the box didn't have enough oomph. Winamp, under Windows, wasn't very reliable. It took a Gods-small and efficient mp3 player, at a real command line, without X running, along with a program designed specifically only to buffer audio, for it to function reliably.
It chewed up more than 90% CPU according to top. And yes, I was pleased with that -- it worked.
Nowadays, I get occasional skips under Windows on my 2.4GHz quad-core Q6600 box. And similar skips and strangeness on my Athlon XP 1900+ Linux box. Both of which, one would think, would be adequate to play a fucking mp3 without hacks and tweakage. *sigh*
Personally, I'm puzzled by the whole thing.
With just a little bit of spare CPU on a Windows box, my PS3 has wife-compatible Hulu and Netflix. And it's also a wife-capable CD player. And a wife-capable DVD player. And a wife-capable Blu-Ray player. And even the wife can figure out how to stream music and shows from the Mediatomb box onto the PS3. I understand you can also play games on a PS3, and I'm sure nobody believes me, but even my wife has been seen doing so.
It's not currently doing any OTA stuff, but I don't have any hardware to support it just yet . . .