've found Winamp 5 stealing 99% of my CPU when it's asked to load a file (any file, MP3, PLS, M3U, anything) from a directory containing a very large amount (1000+) of subdirs and files. It doesn't release CPU until you kill it or hit escape, after which it will take a few minutes for it to close the open diag.
Interesting. When you mentioned that I tries this on my machine with a 46gb folder of about 6500 songs. Using winamp 2.8 (the last good one) it took only a couple of seconds longer than normal to begin playing music when I told it to open the entire folder and begin playing.
I'm always hearing about these new music players but I've still not found one better than winamp. I tried the later media player as well that (finally) has "seamless" song selection playback but even that doesn't work as well as the plugin for winamp.
WMP10 looks like wmp9, which is to say it looks like shit. Why do they have this ridiculous need to put that hideous frame around their player? Why is wmp the one piece of software that seems to be exempt from the standard windows gui? And why does ms take so goddamn much bloat to do this when winamp2.x did the same thing with almost no bloat at all?
I don't know any places around here that get paid when someone hauls off their used vegetable oil and there's a whole mile of fast food places just around the corner from where I sit. And recycling it is NOT just a simple matter of "filtering it." Vegetable oil is an organic product that does not last forever. It WILL go rancid and using it for cooking speeds up this process greatly. About the only way you could keep up this process of use and recycle is if you were born without a sense of smell (or just without sense period).
Some used cooking oil does get filtered and shipped abroad for use in food products. But most places I know (including mcd, bk, kfc etc) still have to arrange to have it hauled off and the best they can manage so far is to break even.
a) where do you find an LCD panel rated at 45,000 hrs? I've never even seen a backlight ccfl rated at that, much less an lcd panel.
b) lifecycle numbers are under bias. FWIW many electrolytic capacitors are also rated for 1000 hr lifecycles, and you don't see many tv sets just blowing up after 6 months. "Lifetime" typically means "this much time until specifications change X%." For capacitors it's typically a 20% change in value, and this change is not linear - the greatest change comes in the first 100 hrs or so and degrades slower after that.
Given "normal" program material and use in a true color display "1000 hrs" absolutely does NOT mean "it dies in 40 days." It means after 1000 hrs under bias any given pixel element will lose 50% of its brightness. In a 1/64 duty cycle system this means you can multiply those 40 days by 64 - about 2500 days, or 7 years.
As someone else has pointed out, the real challenge is getting a reliable means of producing panels with consistent degradation of all pixels over time. If you have 10% of the red oleds fading after 800 hrs and 20% of the green elements fading after 1200 hrs you're going to have a display with splotches of color that, over years, becomes worse and worse.
Still, this is no worse than LCDs that typically require repair after just a couple of years because their backlight (or the inverter driving it) has failed. At best you can hope for a warning as the color gradually turns pink - or maybe you just turn it on one day and find the screen is "dead." Or your projection set - those bulbs are often a couple hundred bucks, and damn few are rated at more than 2000 hrs lifetime. Given all that, this 1000hrs don't seem bad at all.
organic leds are not exactly like silicon leds. They are apparently quite a bit more capacitive than the "conventional" LEDs you are likely thinking of ( which would be expected since they are, after all, made from plastic - a material used to make capacitor dielectrics). this capacitance will either slow down refresh cycles or drive up power consumption. In a home unit you could probably live with the added power consumption to get a great display, in a notebook that might be a bit more of an issue.
But even if you can live with higher power dissipation, that power has to "dissipate" somewhere. On a glass display, the only place that power could go is through the glass itelf or maybe on a heatsink across the back.
It seems certain this technology will become inexpensive enough to compete. I just wish they would hurry up about it...
How are linux users being treated second class? What they are providing is very comparable to what they are providing under Windows: a binary library for supported other pieces of software. That's what the Windows users are getting.
Good point. What I should have demanded in that post is not source code, but documentation. And if you think windows users don't have that either, well, good point again - but again I'll point out that the PC became what it did because it was well documented. If I just want proprietary hardware that "just works" and I don't give a shit how it does it or how well it's supported I'll buy a mac and be done with it.
IOW I gave up ATI some time ago and don't look back. Looking forward I don't see any Via parts in my future either - if that means limiting myself to intel, amd or sis motherboards and graphics that's damn skippy with me - the system I have now plays high rez video just fine with a humble 1600xp cpu and if I want or need a feature added I can hack away at the driver (or pay someone else) until I get it.
This is a trait that, aside from Ati, seems to be almost uniquely asian. If I want to work with some great uber intel or amd or motorola chip all I need do is contact the maker and request info. Aside from patented and trademarked stuff like dolby dsp source code they really don't give a shit who I am nor will they demand to see my wallet before accomodating them. Intel of late seems to be going back to that very arrogant model they adopted in the eighties and now are apparently witholding info on certain features of their cpu chips in regard to wireless networking - so perhaps I should even leave them out of this commentary (or mroe accurately, put them aside with via and the rest)
What would you rather have: partial support that lets you use the hardware, even if it's imperfect, or no support at all?
Hardware with no documentation is, essentially, no support at all. A pos driver that works with "some systems" (and most often not very well) is NOT an acceptable solution - not to me and not to most any home user who wants a computer that just works.
If you want to reward a company with your business it should be the ones who provide accurate documentation on their chips that will allow others to make use of them and the community to support them. To buy into any other system is to accept a "license" on hardware itself - if that's your game you might as well just stick with windows or go get a mac because it's all the same. If linux continues to grow as it has makers like via et all will simply have no choice then but to catch up.. but giving them money now sets the very bad precedent of telling them "ok, here's the vaseline.. have at it, skippy."
VIA provides this shared library (libddmpeg.so) as a binary file for certain supported distributions. Although there are some attempts to reverse engineer the library, the source code is not available. So if you need to use this library you must use one of the supported distributions.
(Editor's note: The source code is available to large OEM customers under NDA/licensing agreements. It is not available to end-users.)
This is what I'm talking about. Linux users do NO ONE any favors when they adopt this sort of crap. Via is not the only maker of small motherboards. When they stop treating linux users like sewcond class citizens I'll happily go looking here - until then Via will get not one penny of my money. Not for one of these mini systems nor for any other motherboard or peripheral.
Better still, just document the damn hardware so we can use it.
They're cool little machines, but we need to be putting more pressure on ALL these hardware companies to document their shit. Providing comprehensive documentation on the use of the hardwaer they spent so many man-years developing doesn't protect them from their competitors at all - it jsut lowers the value of their product. In this case, no matter how small it ain't worth it - I can get a (well supported linux compatible) matx card that's nearly as small and it will let me choose the cpu - including an underclocked athlon that would still smoke that underpowered C3.
It's not the motherboard that's "inductive" (although everything does have inductance, capaitance, etc). What matters here is the power supply itself. Most every pc power supply (I would say every one of them, but there's probably one or two out there that would prove this wrong) uses switching circuitry to chop down a "bulk" supply of 70-200VDC into something the computer can use. This "bulk supply" was, in the beginning, created by simply rectifying the AC line directly and dumping into a fat capacitor. Most TV sets and CRT monitors still do this.
But better power supplies don't simply use a cheap diode/cap bulk supply anymore because this puts a HUGE load on the AC line - especially when it's first turned on and that fat cap has to charge up. This capacitive load also causes non-ideal power factor alignment the entire time the system is on, which means your effective energy consumption (what it reads at the meter) may not actually reflect actual use in the system. Rsistive (neutral) to slightly inductive loads are generally favored by the power company and their meter equipment, and there are real benefits to optimizing power factor of euqipment (especially if you're in an office or hosting location with hundreds or thousands of computers). So, most of the better PC power supplies for PCs employ PFC, or power factor correction. Still, not all of them do.. it's something worth watching in the specifications when shopping for a supply (or a CRT monitor) for a new system.
I wonder if they fixed this version? Because I've been futzing around with the 10.0 "community version" that was released earlier and I've never been able to get it to actually mount an encrypted drive - home or otherwise. I've scoured newsgroups for help and found nothing but years old howtos on using loopback support with the older kernels. I'm told there's a bug in modloop (I think it was) but because the "update your system" wizard also cannot seem to get through to some server in de (no matter which server I try initially) I can never get any insight into what packages actually need patching.
I tried mandrake 7 and 8 and they were both a bit wanting but overall pretty good - at least I was able to connect and pull in system updates. But I've reinstalled this goddamned OS at least thirty times (yes, literally) on two different machines and it's always broken roughly the same way. Since I get no error messages and the discs checkout fine with the disc checking tool I can only assume this is another example of why.0 releases are bad news.
That said, tomorrow it's off to the library to fetch the new fedora core.
Still looking for that linux release that will let me confidently wipe my windows partition for good...
it's not about being able to surf on your tv, it's about being able to have all your shit in one place.
That said, if it says Sony on the front you can be sure it's an overpriced pos. I don't game and can't comment on the virtues of the playstation, but I do know every Sony appliance I ever bought died an early death. My $179 6 head Sharp VCR lasted years and kicked ass. It was such a great machine I actually lamented its death. Meanwhile, my $229 Sony four head had a terrible fucking picture with anything other than absolutely perfect tapes - which it regularly ate from the time it was a mere three months old.
Every Sony I ever bought fell apart quicker than either the device it replaced or the device that replaced it and their longevity only goes to prove the old adage about suckers being in infinite supply.
Yes, before 1985 "windows" was a generic term describing the holes in one's wall. So what? Before 1964 the word "Mustang" was also rather generic, as was the term "Camaro" before 1967 or so. Do you think Ford (or anyone else besides GM) could get away with calling their car a Camaro? Or anyone but Ford a Mustang?
Last time that site was mentioned here it was offline for nearly a month. I had five CDs in the encode queue when the site was slashdotted last month and I was only able to start downloading again little more than a week ago. Had great service with them for months (years already?) and one cover story blows them out of the water for more than three weeks. I'd prefer if you'd go back to keeping this our little secret... at least until I get the rest of my Shakespeare's Sister and Moloko.
Why has no one mentioned the difference between servers and relays? It's no secret there are lots of open relays in that part of the world - I want to know how these folks think they have determined the actual "servers" are there and not the relays.
Lots of open usenet servers in china - but how is their propogation? Lots of open mail relays, but I very much doubt the source of the spam are these machines in PROC.
I have seen more meaningful "articles" get modded no higher than 1 here on/. I'm serious about this, and not saying this just because I'm some gnome zealot (I'm typing you this right now from a MS desktop of my own). But that "article" was what... six paragraphs? I won't go on about how you managed to spread those measly few paragraphs over two pages but will point out that, for all the "information" it provided, it still managed to be about five paragraphs too long.
I'm also surprised this got/.ed - it's an opinion piece to be sure, but amounts in content to no more than some silly usenet rant -- and, frankly, I've seen plenty of those full of vitriol that still maanged to say more than this (ahem) "article."
Wow, that's a lot of storage, is that in one drive?
If it's like the other Vaio "video pcs" I've had, it is indeed all one drive - most likely a custom made "Bigfoot" drive that spins a princely 4500RPM.
If it is, then what if the drive fails?
You won't notice anyway, since it will work just as well without them as it did the day it was new.
The Iraqi prisoner pictures is about the WORSE example you could have chosen. I mean, they came out on mainstream media MONTHS after they were taken. Hell, they might as well have been taken with a 1940 vintage "Brownie" box camera and shipped to the US in a bottle...
Meanwhile, many of us see movies weeks before they're even released to theatres and watch TV shows the day after they air via internet exchanges. Just the other day someone promised to post a TV program that had JUST aired "as soon as the encoding is done" which, in this case, was about four hours.
I buy and sell shit via the internet in the blink of an eye. Just the other day I bought another CD from magnatune and the only reason it took me a day to get it was because of my hideously slow dialup connection and my insistence on getting the highest practical quality (FLAC).
ALL these examples and the best you can come up with is to mention an "old guard" news source releasing months old photos only AFTER they had "cleared it with washington?" Yeesh.
I just added it up: I've spent (just on thinkpad parts) less than $1000 in the last three months. I have assembled from these purchases 6 thinkpads - one with a 12" screen, five with 13" screens. Most have 300MHz CPUs although one is 366MHz and one is a 500MHz PIII t600x model. (Well, that's sorta a lie - it's actually a 600x motherboard with the rest of the stuff from a DOA 266MHz model I picked up for about 50 bucks.) I also have an assload of spare parts including a stack of those hard drive covers that are currently going for about twenty bucks each. The two that have "shipped" went with decent batteries and 2GB hard drives - the rest I'll offer without battery warranty with larger (6GB) hard drives. Easy 50% profit on each; won't get rich on that, but it's enough to pay for the hobby and have a few extra bucks each month to put in the Benz kitty.
If I wanted to put together four more machines that I would back with a proper 90 day warranty it would cost maybe another $400 for warrantied hard drives and new batteries. That's less than $1500, although selling them at $325 a pop it would suck up most of my profit. Most folks I know use their laptops plugged in all the time and don't care about good batteries and I stand behind the rest of the unit.
BTW it's "CD/2nd HDD/Floppy/2nd Battery" - they all fit interchangeably in the same slot.
I probably shouldn't be giving away the store like this here, but these are real gems. The 570E may be a bit thinner, but it's also less supported - there doesn't seem to be as many of them around, so the 570 unique parts are much more expensive. And no, I agree about not needing a smoking P4 laptop - I'd much rather have that extra money for something more useful - like, I dunno, a decent used car that won't suck my wallet dry at the pump.:)
Many apologies to the downtime of the publicly available IIP network - the hardware supporting the end server has been destroyed - and the programming and support for this project has been severely lacking.
Wanna guess why? Not some conspiracy - just because no one used the thing. So what if you have an "invisible mesh of servers" if there are only 1000 users? Doesn't take long to nail down 1000 IPS when they're all connected to one another 24/7. We discussed this in the support forum many times and, so far as I can recall, nobody ever came up with a reasonable solution to prevent people from disconnecting and reconnecting. Do that enough times, pretty soon you've got everyone's address.
anyone could use an anonymous proxy to cover the IP
This is how people get arrested. How do you know you can trust the proxy? If you are connected to a box cracked by a worm or some script kiddy, odds are it's some windows box on a cable modem. It's likely to have ZERO security and if it has a relatively new OS on it it would take about thirty seconds to crack into it and get a list of ALL the machines with active connections to it. If it's an older machine then, darn it, you gotta take an extra minute or two to upload a toolkit.
Lather, rinse, repeat.
And what if it's not a script kiddy box at all, but a share on some vigilante's honeypot?
Trusting proxy servers while doing illegal shit on IRC - if you don't really, really, really know what you are doing - is just a fast pass to the local lockup.
Apparently you don't watch the evening news (never mind read the messages these folks exchange). If you're busted and they find ANYTHING that will give them reasonable cause to suspect you are hiding something hardcore (or even just actionable in their jurisdiction) they'll take your computers, your monitors, routers, switches, mice, keyboards, your DVD players, your game machines - you'll be lucky if they leave you with a clock radio and a telephone.
There are many open source projects using usenet forums and lots of companies do that on closed NNTP servers - but that ain't usenet. Use of NNTP does not equate with USENET. There's something to be said for product support (which I already pointed out) but a company would be supremely stupid to develop their new proprietary products in an openly accessible USENET discussion group.
Interesting. When you mentioned that I tries this on my machine with a 46gb folder of about 6500 songs. Using winamp 2.8 (the last good one) it took only a couple of seconds longer than normal to begin playing music when I told it to open the entire folder and begin playing.
I'm always hearing about these new music players but I've still not found one better than winamp. I tried the later media player as well that (finally) has "seamless" song selection playback but even that doesn't work as well as the plugin for winamp.
WMP10 looks like wmp9, which is to say it looks like shit. Why do they have this ridiculous need to put that hideous frame around their player? Why is wmp the one piece of software that seems to be exempt from the standard windows gui? And why does ms take so goddamn much bloat to do this when winamp2.x did the same thing with almost no bloat at all?
And by the way, making diesel fuel provides plenty of soap as well. And did you know you can make napalm from kerosene and concentrated orange juice?
Some used cooking oil does get filtered and shipped abroad for use in food products. But most places I know (including mcd, bk, kfc etc) still have to arrange to have it hauled off and the best they can manage so far is to break even.
b) lifecycle numbers are under bias. FWIW many electrolytic capacitors are also rated for 1000 hr lifecycles, and you don't see many tv sets just blowing up after 6 months. "Lifetime" typically means "this much time until specifications change X%." For capacitors it's typically a 20% change in value, and this change is not linear - the greatest change comes in the first 100 hrs or so and degrades slower after that.
Given "normal" program material and use in a true color display "1000 hrs" absolutely does NOT mean "it dies in 40 days." It means after 1000 hrs under bias any given pixel element will lose 50% of its brightness. In a 1/64 duty cycle system this means you can multiply those 40 days by 64 - about 2500 days, or 7 years.
As someone else has pointed out, the real challenge is getting a reliable means of producing panels with consistent degradation of all pixels over time. If you have 10% of the red oleds fading after 800 hrs and 20% of the green elements fading after 1200 hrs you're going to have a display with splotches of color that, over years, becomes worse and worse.
Still, this is no worse than LCDs that typically require repair after just a couple of years because their backlight (or the inverter driving it) has failed. At best you can hope for a warning as the color gradually turns pink - or maybe you just turn it on one day and find the screen is "dead." Or your projection set - those bulbs are often a couple hundred bucks, and damn few are rated at more than 2000 hrs lifetime. Given all that, this 1000hrs don't seem bad at all.
But even if you can live with higher power dissipation, that power has to "dissipate" somewhere. On a glass display, the only place that power could go is through the glass itelf or maybe on a heatsink across the back.
It seems certain this technology will become inexpensive enough to compete. I just wish they would hurry up about it...
Good point. What I should have demanded in that post is not source code, but documentation. And if you think windows users don't have that either, well, good point again - but again I'll point out that the PC became what it did because it was well documented. If I just want proprietary hardware that "just works" and I don't give a shit how it does it or how well it's supported I'll buy a mac and be done with it.
IOW I gave up ATI some time ago and don't look back. Looking forward I don't see any Via parts in my future either - if that means limiting myself to intel, amd or sis motherboards and graphics that's damn skippy with me - the system I have now plays high rez video just fine with a humble 1600xp cpu and if I want or need a feature added I can hack away at the driver (or pay someone else) until I get it.
This is a trait that, aside from Ati, seems to be almost uniquely asian. If I want to work with some great uber intel or amd or motorola chip all I need do is contact the maker and request info. Aside from patented and trademarked stuff like dolby dsp source code they really don't give a shit who I am nor will they demand to see my wallet before accomodating them. Intel of late seems to be going back to that very arrogant model they adopted in the eighties and now are apparently witholding info on certain features of their cpu chips in regard to wireless networking - so perhaps I should even leave them out of this commentary (or mroe accurately, put them aside with via and the rest)
What would you rather have: partial support that lets you use the hardware, even if it's imperfect, or no support at all? Hardware with no documentation is, essentially, no support at all. A pos driver that works with "some systems" (and most often not very well) is NOT an acceptable solution - not to me and not to most any home user who wants a computer that just works.
If you want to reward a company with your business it should be the ones who provide accurate documentation on their chips that will allow others to make use of them and the community to support them. To buy into any other system is to accept a "license" on hardware itself - if that's your game you might as well just stick with windows or go get a mac because it's all the same. If linux continues to grow as it has makers like via et all will simply have no choice then but to catch up.. but giving them money now sets the very bad precedent of telling them "ok, here's the vaseline.. have at it, skippy."
No thanks. I'll keep my money and my vaseline.
(Editor's note: The source code is available to large OEM customers under NDA/licensing agreements. It is not available to end-users.)
This is what I'm talking about. Linux users do NO ONE any favors when they adopt this sort of crap. Via is not the only maker of small motherboards. When they stop treating linux users like sewcond class citizens I'll happily go looking here - until then Via will get not one penny of my money. Not for one of these mini systems nor for any other motherboard or peripheral.
Vote with your dollars. And be vocal about it.
They're cool little machines, but we need to be putting more pressure on ALL these hardware companies to document their shit. Providing comprehensive documentation on the use of the hardwaer they spent so many man-years developing doesn't protect them from their competitors at all - it jsut lowers the value of their product. In this case, no matter how small it ain't worth it - I can get a (well supported linux compatible) matx card that's nearly as small and it will let me choose the cpu - including an underclocked athlon that would still smoke that underpowered C3.
But better power supplies don't simply use a cheap diode/cap bulk supply anymore because this puts a HUGE load on the AC line - especially when it's first turned on and that fat cap has to charge up. This capacitive load also causes non-ideal power factor alignment the entire time the system is on, which means your effective energy consumption (what it reads at the meter) may not actually reflect actual use in the system. Rsistive (neutral) to slightly inductive loads are generally favored by the power company and their meter equipment, and there are real benefits to optimizing power factor of euqipment (especially if you're in an office or hosting location with hundreds or thousands of computers). So, most of the better PC power supplies for PCs employ PFC, or power factor correction. Still, not all of them do.. it's something worth watching in the specifications when shopping for a supply (or a CRT monitor) for a new system.
I tried mandrake 7 and 8 and they were both a bit wanting but overall pretty good - at least I was able to connect and pull in system updates. But I've reinstalled this goddamned OS at least thirty times (yes, literally) on two different machines and it's always broken roughly the same way. Since I get no error messages and the discs checkout fine with the disc checking tool I can only assume this is another example of why .0 releases are bad news.
That said, tomorrow it's off to the library to fetch the new fedora core.
Still looking for that linux release that will let me confidently wipe my windows partition for good...
That said, if it says Sony on the front you can be sure it's an overpriced pos. I don't game and can't comment on the virtues of the playstation, but I do know every Sony appliance I ever bought died an early death. My $179 6 head Sharp VCR lasted years and kicked ass. It was such a great machine I actually lamented its death. Meanwhile, my $229 Sony four head had a terrible fucking picture with anything other than absolutely perfect tapes - which it regularly ate from the time it was a mere three months old.
Every Sony I ever bought fell apart quicker than either the device it replaced or the device that replaced it and their longevity only goes to prove the old adage about suckers being in infinite supply.
Last time that site was mentioned here it was offline for nearly a month. I had five CDs in the encode queue when the site was slashdotted last month and I was only able to start downloading again little more than a week ago. Had great service with them for months (years already?) and one cover story blows them out of the water for more than three weeks. I'd prefer if you'd go back to keeping this our little secret... at least until I get the rest of my Shakespeare's Sister and Moloko.
Lots of open usenet servers in china - but how is their propogation? Lots of open mail relays, but I very much doubt the source of the spam are these machines in PROC.
Hop over to easynews and ask the admins there how well that works for you once the MIB show up.
I'm also surprised this got /.ed - it's an opinion piece to be sure, but amounts in content to no more than some silly usenet rant -- and, frankly, I've seen plenty of those full of vitriol that still maanged to say more than this (ahem) "article."
If it's like the other Vaio "video pcs" I've had, it is indeed all one drive - most likely a custom made "Bigfoot" drive that spins a princely 4500RPM.
If it is, then what if the drive fails?
You won't notice anyway, since it will work just as well without them as it did the day it was new.
How pissed are you gonna be?
Not even surprised...
Meanwhile, many of us see movies weeks before they're even released to theatres and watch TV shows the day after they air via internet exchanges. Just the other day someone promised to post a TV program that had JUST aired "as soon as the encoding is done" which, in this case, was about four hours.
I buy and sell shit via the internet in the blink of an eye. Just the other day I bought another CD from magnatune and the only reason it took me a day to get it was because of my hideously slow dialup connection and my insistence on getting the highest practical quality (FLAC).
ALL these examples and the best you can come up with is to mention an "old guard" news source releasing months old photos only AFTER they had "cleared it with washington?" Yeesh.
If I wanted to put together four more machines that I would back with a proper 90 day warranty it would cost maybe another $400 for warrantied hard drives and new batteries. That's less than $1500, although selling them at $325 a pop it would suck up most of my profit. Most folks I know use their laptops plugged in all the time and don't care about good batteries and I stand behind the rest of the unit.
BTW it's "CD/2nd HDD/Floppy/2nd Battery" - they all fit interchangeably in the same slot.
I probably shouldn't be giving away the store like this here, but these are real gems. The 570E may be a bit thinner, but it's also less supported - there doesn't seem to be as many of them around, so the 570 unique parts are much more expensive. And no, I agree about not needing a smoking P4 laptop - I'd much rather have that extra money for something more useful - like, I dunno, a decent used car that won't suck my wallet dry at the pump. :)
Charge is stored in a capacitor. Doesn't matter how many hoops you care to jump through to get there, what I said was accurate. "Mr.E"
Yeah, it's posted AC but it's still funny...
Wanna guess why? Not some conspiracy - just because no one used the thing. So what if you have an "invisible mesh of servers" if there are only 1000 users? Doesn't take long to nail down 1000 IPS when they're all connected to one another 24/7. We discussed this in the support forum many times and, so far as I can recall, nobody ever came up with a reasonable solution to prevent people from disconnecting and reconnecting. Do that enough times, pretty soon you've got everyone's address.
This is how people get arrested. How do you know you can trust the proxy? If you are connected to a box cracked by a worm or some script kiddy, odds are it's some windows box on a cable modem. It's likely to have ZERO security and if it has a relatively new OS on it it would take about thirty seconds to crack into it and get a list of ALL the machines with active connections to it. If it's an older machine then, darn it, you gotta take an extra minute or two to upload a toolkit.
Lather, rinse, repeat.
And what if it's not a script kiddy box at all, but a share on some vigilante's honeypot?
Trusting proxy servers while doing illegal shit on IRC - if you don't really, really, really know what you are doing - is just a fast pass to the local lockup.
There are many open source projects using usenet forums and lots of companies do that on closed NNTP servers - but that ain't usenet. Use of NNTP does not equate with USENET. There's something to be said for product support (which I already pointed out) but a company would be supremely stupid to develop their new proprietary products in an openly accessible USENET discussion group.