> If you don't accept testimonials as proofs, you
won't accept any proof as proof.
The cry of the quack everywhere.
Yeah, I know Slashdot was just trollin' when they decided to
interview Chiu, and I think both the "Ask Alex Chiu" and his responses constitute some of the funniest shit
I've read this week.
But to get serious for a minute: If you're interested in
quackery, check out
http://www.quackwatch.com
for all the dirt.
In the case of Alex, he hits pretty much all the
buttons on
How Quackery Sells. Bravo!
> How do scientists know the diameter of earth? Must a scientist walk
the equator in order to tell you the exact diameter of earth?
[... ]
We gather
scientific evidence (testimonials) in order to base our assumptions.
Alex, m'boy, you wouldn't know the scientific
method if it propositioned you on the street and asked for a rim-job.
> Gotta say Tivo just lost a customer though. I had been thinking about buying one for awhile, had it
budgeted for next month. But a bullshit move like this... *sigh*.
Ditto.
Next project: Spend some time checking out the status of Linux drivers for my ATI All-in-Wonder-128 32M card.
And if that doesn't give me any love, install Windoze Scripting Host and goof around with VB for a weekend using the bundled software. I can capture and encode MPEG2 video in real time when the box is suitably overclocked. All that's missing is a 30G hard drive and a user interface.
A little bit of Perl (yeah, it'll run in a Windoze console too!) and a suitable lynx -dump to and I've got my TV listings.
> but there's a risk here that
viruses will have a nice little hook for gleaning information from every document you open in a
SmartTags-aware program.
Hmm, COM objects and other fun stuff...
The scary thing is that the privacy-invasion implications are probably thought of as a "feature", not a bug.
(After all, this is a company whose own marketroid said it's a way to "improve" sites that don't have enough URLs for Bill's satsifaction...)
> Security and monopolistic activities aside, how many people actually want these features?
Hell, no, but that's beside the point. We're not the customer. OEM computer manufacturers and Microsoft's marketing products are the customers, Windows users are the product.
Umm, while we're at it - didn't we rake Dejanews over the coals for this last year?
When I put up a web page about modems, I don't want every instance of the word modem to go
to a specific modem manufacturer chosen my MSFT.
Which brings me to the privacy implications.
What do you want to bet that it doesn't go to a <A
HREF="http://www.usrobotics.com">modem%lt;/A> manufacturer's site, but it actually goes to a <A HREF="http://msid.msn.com/tracker.cgi?smart_tag=mo dem&GUID=[your GUID]">Microsoft-operated tracking site</A> before redirecting you to the advertised site?
(And what do you want to bet that when you mouseover to see where the smart tag goes, you don't see the actual MSFT tracker, but just the words "Smart Tag! Find out about 'Modem'" in the status bar?)
> [The EMI/Roxio deal doesn't matter when compared to] the real show is the DRM technologies built into Windows XP. Think XP will let
you burn DRM-locked files with its built-in CD burner? Remember that within a year, almost every
consumer PC will be shipping with XP installed.
You're absolutely right.
To which, I say "Bring it on".
In the days of Win9x, the answer to problems like this was "Go here and download the warezed version."
In the days of WinXP, the answer may well be "You can't. DRM's built into the OS. Here's a Linux CD, try this instead."
Re:Do you have Britney Spears home address?
on
Napster Going Legit
·
· Score: 1
> I can walk into any High School in any major city in America and find at least 3 girls who can sing as well
as she does. > > You would be surprised how good you can make a mediocre singer sound if you mic them just right, use
the best take, and choose your effects carefully.
<JOB_ROLE=RIAA_AGENT VOICE_MODE=HONEST>
What kind of talent scout are you? Sure, they can sing, but what about their tits?
</JOB_ROLE>
(Or is that what you meant by "choosing their effects carefully";-)
Re:Do you have Britney Spears home address?
on
Napster Going Legit
·
· Score: 3
> One thing jumped out at me that I wasn't looking for- since around 1990, the pattern of platinum and
multiplatium record sales is that the act sells multiplatinum while the label is pushing it, and then
when the label stops, the act entirely stops selling. In previous decades, records had the capacity to
sell from back catalog. That doesn't particularly happen anymore. People _are_ losing interest in
bands once the label stops pumping them into the stores, by the RIAA's own figures.
Interesting.
I've been wondering about that for a while.
I've got a wide range of musical tastes. Funk, disco, new wave, house, techno, industrial, EBM, metal, rap, hip-hop, hell, I'll listen to damn near anything except what's on today's Top-40. I've been listening to some of those genres for 20 years.
Drop by alt.binaries.sounds.mp3.19[5678]0s. You'll see lots of music from each decade.
Drop by alt.binaries.sounds.mp3.rap-hiphop or.dance and things are different. You'll see almost nothing older than the mid-90s. Most of the stuff is last year or newer.
That's not to say that I expect to see Public Enemy's "Nation of Millions" posted every week (though Pink Floyd fans seem to have no trouble reposting Dark Side of the Moon with astonishing regularity:), but the lack of history in the genre groups is startling.
Some of this is due to things like (e.g. dance) the importance of the DJ over the actual tracks and the relative unavailability of the latest mix from DJ $SOME_GUY_IN_IBIZA each week. But it's a little unnerving to see almost a complete absence of "old-school" music in these two sub-genres, as both rap and house/techno variants have histories going back to the early 80s.
The old geezer in me says "Young punks just want the new stuff, they've got no respect for the classics that defined their genre. Buncha cable kiddies who think anything invented more than a week ago is obsolete and not worth listening to anymore".
But I don't trust myself when I start thinkin' like a geezer. It usually means I'm missing something important.
I think the poster to whom I'm replying has stumbled onto the important something I missed.
Suppose we take a typical US dialup user and have them use the hours between midnight and 0600 for my USENET MP3 l33ching, That's six hours a day, or 180 hours a month, to download stuff.
I'm sure my ISP would be pissed at you (and righteously so!) if you were to use six hours a day during peak time, but by doing bulk downloads in the off-hours, you're not costing them anything - the modem pools are mostly empty, and even your local phone company's network is unloaded.
Six hours at 48000 bps = about 12M per hour of MP3s from USENET once you subtract out the uuencoding bloat and allow for some latency. 72M per day, or about 2G per month. If we assume 160kbps MP3s, that's about an hour's worth of new music every day. Egad.
(Given a 2-3G per month cap, why bother with broadband?;-)
> Thats's why Telstra should try to get as many servers as possible on their network. Servers run by their customers,
that is.
Great point!
Consider the MP3 leech from an ISP's point of view. Would you rather have 10000 users downloading (transiting) 3GB per month of MP3s from Supernews or Newsguy, or would you rather have them doing it from your own news server?
One way, you pay 300GB per day (a full NNTP feed) for 9TB per month transit to your NNTP server. After that, it's all internal traffic to you.
The other way, you pay 3GB per month times 1000 users, for 30TB per month to someone else's NNTP server (or to P2P servers located outside your network, etc...)
As "bad" as the situation is wrt USENET and binaries, for a sufficiently-large ISP, there may be a business case for upgrading/maintaining your own news server farm, rather than expecting your users to outsource it.
The same applies to P2P servers within the network:
Suppose you had a news server with only one day retention, but 100 of your 10000 USENET users each downloaded 1% (2.5G per day) of the feed and redistributed it by setting up their own NNTP servers for "Stuff They Liked". If each of these 100 individuals, who would only need to eat 2.5G per day off your news server (and it's all LAN traffic, so you don't pay for their transit), dropped $150 on a 30G hard drive, you could effectively have 10 days' retention, and take some of the load away from your news server.
I'm sure similar economies apply to Gnutella and more pure P2P solutions. I'm just using NNTP as an example.
> If the recording industry was
destroyed, how many people would have heard early black Jazz performers?
True in the early part of the 20th century, when electronic distribution at zero cost was impossible, leaving for-pay distribution at moderate cost as the only way for the "Rock and Roll" derived from early Jazz/R&B musicians to make its way into the ears of Joe Consumer and sweep the nation in the 1950s.
False in the early part of the 21st century, when electronic distribution at zero cost was widely available, and the recording industry served only to prevent Joe Consumer from hearing the early Tibetan Electronica artists whose work set the stage for the solar-system-shaking grooves that rocked the colonies of Luna and Mars in the 2180s.
> The only response to all of the systems that will restrict your fair use (and I don't know if this scheme does) is to keep yer
old stuff in working order! For once obsolescence(sp?) may be a GOOD thing, because they can't add copy-protect
nonsense to the stuff you already own.
Suppose they pulled the plug on useful hardware today. Every manufacturer on the planet goes to copy control, no exceptions, and nobody ever cracks it. (This is of course, impossible. But let's go with the doomsday scenario for a bit.)
Given the number of ATs and XTs at my local surplus store, and the time it's taken to have them gradually replaced with 386s and 486s, I'd say we have at least a 10-15 year supply of useful hardware ahead of us.
And given the age of some of the older pieces of hardware in my collection, I'd say we have at least another 10 years, probably more, before that supply of useful hardware starts to fail.
If Copy Control Doomsday happened tomorrow, we'd have about 25 years before we had to worry. Spare parts purchased now, run for 6 months (to shake out any cases of "infant mortality") and stored in anti-static bags, will be just as good 25 years from now as they are today.
For less than $1000, you can buy enough hard drive storage and multiple sets of spare parts to store 100G of MP3s and have at least one system capable of playing them back for the rest of your natural life.
> Let me see here...your new cd burning software will not burn songs unless they are digitally signed. In what way is this
wrong?
My old CD burning software didn't care about copy control.
My old CD burning software did things my new CD burning software doesn't do.
My old CD burning software was more functional - I could do more things with it than I can the new version.
What's wrong about it is that there are people trying to pass off downgrades as upgrades.
If your local Porsche dealer said "By the way, the new model Porsche has a rev-limiter hooked up to a GPS system that prevents you from going faster than 55 mph! It's so much better than last year's model!", you'd slap him silly, and you'd be right to slap him silly whether you ever intended to drive over the speed limit or not.
> My solution to running a fan at lower voltage would be to run it off the 12V supply with an appropriate
series resistor. Calculate the value from the power rating of the fan. There's no need to get the 5V supply
involved. > >
You might find that the fan has trouble starting up on the lower voltage supply, but runs fine once it gets
moving. In this case, a big capacitor in parallel with the resistor will supply the full voltage to the fan for a
short time after the power is applied.
Actually, I had exactly the problem when I tried running a huge-ass 120mm 24V fan at 12V. The resistor-capacitor circuit wouldn't help there, but it'd solve the equivalent problem for 7V fans quite nicely. Thanks!
> If you want to bring real probability theory in, you can even show that with even odds but differences in
fortune (bank has infinity for all practical purposes, you have xxx USD), the bank is also guarenteed to win
all your pennies. So even with even odds, your chances to win are smaller than the banks, let alone with
several greens and maximal bets.
Actually, he did. "What are the green squares for?" was replied with an example along the lines of "Suppose you doubled your bet with every spin of the roulette wheel (and there were no green squares), every time you lost. You'd be guaranteed to win. The problem is that if you double your bet every time, you've gotta have more money than I [Dad, playing the house] do. The green squares make sure that even a guy with a million bucks can't try this."
Of course, we did it out with pen and paper and "real numbers" ($2, $4, $8, $16...) instead of exponents, but the effect was the same.
> I couldn't believe the $20 and $50 minimums at tables, and that there were no $5 tables!
(Yeah, it was Reno, and they had a $5 table. I could hack a $20 minimum with $100 to play with, but anything beyond that is out of my league. It stops being fun when it's about real money. If I wanna do that, I'll play the stock market:)
> I like the games themselves, but I don't like the depressingly mundane
(yet garish) surroundings.
The "entertainment" in those places is always
meant to appeal to one or two previous generations
> I don't doubt for one moment that the mafia runs the games, or that the regulators know that. My take on it is
that the legal, fair games are nevertheless so profitable that the mafia wants in.
Hey, what's wrong with the mob?
Compare the payouts on the mob-run "numbers game" in with the payout ratio of the government-run lotteries. The mob offers better odds, according to what appears to be expert testimony on the subject:
MR. KARCHER: --but I'm told every major gaming site where there's enough employees to have a
numbers, a meaningful numbers game played at a work site, it's preferable to a lottery because the
payout is better. The payout is always constant in an illegal lottery and the illegals numbers game,
the games that were played -- first of all, the illegal numbers game were never able to go into lotto
because there was never enough sophistication so it's always just been a three digit game, the
numbers business. And on job sites or in factories it is still played and the payout remains constant.
Whereas, in, as I understand it, the three digit game is a function of how many people have that
number so the payout can be as low as two or three hundred dollars. Is that correct? At least it is in
New Jersey. In other words, if you played the numbers illegally you would get a $500 payout no
matter what, out of 999 numbers. If you had that number you would get a $500 payout, whereas you
might run the risk in a legal lottery, a legal numbers game of being paid out only two hundred and
some odd dollars or three hundred and some odd dollars. So, I'm told that some people still prefer to
play the old fashioned game.
Given the choice, it looks like the mob offers a superior product in the gambling market. Which reminds me, both the government and the Mafia have programmes whereby they offer "protection" you didn't ask for in exchange for "protection money" you have to pay. Both have the power to see that Bad Things Happen to you if you don't pay up. What's the difference, really? (There are days when I think the only reason the government fights organized crime is because it can't stand the competition;-)
>I think I get your point, but there is no 12 in a deck of cards.
But the computer said there was!;-)
Actually, I was aiming for (+1, Funny), but someone could get an easy (+1, Insightful) by pointing out that what I just described as "preposterous, who'd be dumb enough to play cards when you can't see the cards being dealt and you don't have the source code of the algorithm that deals the cards?" is exactly what players of VLTs (Video Lottery Terminals) and electronic slot machines are doing right now in the real world.
>If they taught gambling in schools the gambling industry would be out of business.
Amen.
Actually, I have fond memories of my Dad teaching me (at age 7-8) how to play poker, blackjack, and roulette.
After spending some time taking my pennies, we both noticed that he was doing well, and so he started teaching me why the house always won. "Daddy, what are the green numbers on the roulette wheel for?" is actually a pretty natural question for a seven-year-old.
So we switched sides, and I played the "house" in blackjack and roulette the next time. All of a sudden, my "luck" changed and I "won" my pennies back. And he taught me that luck had nothing to do with it. If you play enough hands and have an advantage, you will win. And the rules are set up so that the house has an advantage.
Was it real probability theory? Not really. But when a kid's just discovered fractions, it's pretty easy to show him that 18/38 is less than 18/36. If betting on black/red is like flipping a coin, it's a fair game. Except that two out of every 38 spins, the coin is designed to land on its edge.
Since then, I've gone to casinos and had plenty of fun. But I've always gone in with cash, and with the notion that my $40 in gambling money is purchasing me a few hours of fun at the blackjack table - in the same way that the same $40 could purchase me few hours of fun in a bar, at the movies, or at the arcade.
Player draws: Twelve of Clubs.
Player has:
Four of Hearts, Seven of Clubs, Twelve of Clubs.
Total=22.
Player Busts.
Dealer draws: Ace of Hearts.
Dealer stands.
Dealer turns over unknown card: Four of Clubs.
Dealer has: Four of Clubs, Six of Diamonds, Nine of Hearts. Total=21.
Dealer wins.
Wager again? Yes/No.
("Aaw, shit, again?!?!?! I always draw a twelve when I'm winning!")
> I personally think there's a quite a bit of "spyware" going on in IE,
You can deactivate this (mis-)feature under Internet_Options -> Advanced -> Search from the Address bar -> (Radio Button) "Do not search from the Address bar".
Of course, 99% of people don't do that, because it's the default, as are things like "Browsing -> Automatically check for IE updates" and "Enable Install On Demand" (which just sounds like an accident waiting to happen).
Back on topic, these sites get so much traffic because they're the default home pages. If users are too dumb to change the default home page (and I've seen usability studies that show many users don't even know the default home page can be changed - they think of the default home page as "The Internet"), they're too dumb to secure themselves against stuff like this.
This isn't an anti-M$ rant. Nutscrape blows hot donkey meat when it does the same thing with "What's Related".
For lack of a better term, I call it "stealth spyware".
(Aside: I agree with the original poster - just turn off Javashit unless you're going to your bank or broker! Any site that requires it tends not to be worth visiting.)
But if that's not enough - just firewall out the offending domain.
My first experience with pop-unders was on clicktv.com, a TV listings site. In order to see what episode of Star Drek is being aired, you need Javashit enabled. Unfortunately, this also cranks up the popunders.
A glance at the HTML source reveals that the page with the TV listings comes from http://www.clicktv.com, but the Pop-under comes from http://media.fastclick.net. (Clever pigfux0rz, they even thought to dc.write ('<scr'+'ipt language=Javashit' to dodge proxies that filter out the string script language="Javashit")
So I said "fuck 'em", and stuck media.fastclick.net into my C:\WINDOWS\HOSTS,/etc/hosts or Junkbuster proxy file depending on my platform or whether I was running Junkbuster, and the problem went away.
General solution: treat the source of pop-ups and pop-unders the way you would Doubleclick. Firewall 'em. Most of the time, it's an HTTP transaction to a tracker that loads the offending Javashit. Firewall the tracker and the Javashit doesn't show up. If the Javashit can't load, the window don't pop.
> You shouldn't need to underclock your computer to turn the fan off if it is a
Pentium 166. An AMD will make a little more heat and may need to be slowed down a little, though.
Depends on the AMD;-)
From personal experience, a K6-III-333 (grey-market laptop part available at Fry's) or K6-2+-450 (make sure it's the "2+" version) underclocked to 200 MHz will outperform (the hell out of) a P166. I run one in an I-Opener, cooled passively, and the mobo temperature never goes above 46C. (And that's with a big power transistor about two inches away... without that transistor, I think it'd be around 40C.)
If your P166 box happens to be running an ASUS TX-97 series board, search for beta BIOS 0112. This version allows you to run VCore down to 2.0V, 2.1V, and 2.2V. That won't help you run your P166, but the K6 CPUs I mentioned require the lower voltages, and are available for around $30-40. And they all have lots (256K on the K6-3-333, 128K on the K6-2+) of on-die cache compared to the P166.
Same MHz, better performance, lower voltage, lower temperatures. AMD CPUs that never made it into laptops are great!
> No, don't do this. When the fan breaks it allows +12V to flow directly into sensitive +5V components. This destroys at least
the motherboard and most likely some of the expansion cards. I am not kidding!
I hadn't thought about that - the guy's got a point. A short on the fan would draw lots of current, and could pull +5V line up to +12V. It wouldn't have to get there before the P/S shut itself down - even +6 or 7V would probably be enough to do damage.
I came close to this when a +12V fan failed catastrophically. The cheap-azz bearings caused the fan blade to wiggle out of position. A blade hit one of the holes in the case and got stuck, which presumably shorted out the fan motor a few seconds later.
I'm gonna hit the surplus stores and hunt down some 12V fuses. I'm thinking that something at 1.5-2 times the rated fan current (probably 250mA tops) in series with the fan before hooking it up should do the trick. The goal is to have the fuse blow on short before the P/S +5V line gets drawn too high.
And with the price of fuses at a surplus store, I could probably do some destructive testing on a junk mobo to verify the fuse is doing the right thing. Lord knows I have enough "dead" (they still turn, but they whine like hell because of dead bearings) fans in my junk parts box!
1) Undervolt your fans. Most 12V fans can run at 7V. (Hook +5V to the GND wire on the fan, and +12V to the +12V wire on the fan).
The fan will run slower, but cooler.
2) Work on airflow near your fan blades.
A fan with a great big sheet of metal with punched holes in it will be loud. The same fan with the metal grill removed will be quieter. The same fan with the metal grill and some extra space around it (because these fans typically blow air out in a cone on a 45-degree angle away from the center of the fan) will run even more quietly.
Still need finger protection? Get a real fan grill - the old-sk00l things that looked like three or four concentric circles of wire stuck together with a couple of cross-wise pieces of wire.
3) Rule of thumb - low RPM = low noise. If you don't get enough airflow (for cooling purposes) when you undervolt your 80mm case fan, carve up the case and add a big-ass 120mm fan. An undervolted 120mm fan can often move as much air as a typical 80mm fan running at +12V. If your local surplus store is well-stocked, you might even find some +24V fans that run at +12V. (But be sure to test them first;-)
Well, there are a few ideas to start with. I'm sure others will follow up.
Slashdot Mirror
The cry of the quack everywhere.
Yeah, I know Slashdot was just trollin' when they decided to interview Chiu, and I think both the "Ask Alex Chiu" and his responses constitute some of the funniest shit I've read this week.
But to get serious for a minute: If you're interested in quackery, check out http://www.quackwatch.com for all the dirt.
In the case of Alex, he hits pretty much all the buttons on How Quackery Sells. Bravo!
> How do scientists know the diameter of earth? Must a scientist walk the equator in order to tell you the exact diameter of earth? [ ... ]
We gather
scientific evidence (testimonials) in order to base our assumptions.
Alex, m'boy, you wouldn't know the scientific method if it propositioned you on the street and asked for a rim-job.
Ditto.
Next project: Spend some time checking out the status of Linux drivers for my ATI All-in-Wonder-128 32M card.
And if that doesn't give me any love, install Windoze Scripting Host and goof around with VB for a weekend using the bundled software. I can capture and encode MPEG2 video in real time when the box is suitably overclocked. All that's missing is a 30G hard drive and a user interface.
A little bit of Perl (yeah, it'll run in a Windoze console too!) and a suitable lynx -dump to and I've got my TV listings.
Hmm, COM objects and other fun stuff...
The scary thing is that the privacy-invasion implications are probably thought of as a "feature", not a bug.
(After all, this is a company whose own marketroid said it's a way to "improve" sites that don't have enough URLs for Bill's satsifaction...)
Hell, no, but that's beside the point. We're not the customer. OEM computer manufacturers and Microsoft's marketing products are the customers, Windows users are the product.
Umm, while we're at it - didn't we rake Dejanews over the coals for this last year?
When I put up a web page about modems, I don't want every instance of the word modem to go to a specific modem manufacturer chosen my MSFT.
Which brings me to the privacy implications.
What do you want to bet that it doesn't go to a ;o dem&GUID=[your GUID]">Microsoft-operated tracking site</A>
<A HREF="http://www.usrobotics.com">modem%lt;/A>
manufacturer's site, but it actually goes to a
<A HREF="http://msid.msn.com/tracker.cgi?smart_tag=m
before redirecting you to the advertised site?
(And what do you want to bet that when you mouseover to see where the smart tag goes, you don't see the actual MSFT tracker, but just the words "Smart Tag! Find out about 'Modem'" in the status bar?)
You're absolutely right.
To which, I say "Bring it on".
In the days of Win9x, the answer to problems like this was "Go here and download the warezed version."
In the days of WinXP, the answer may well be "You can't. DRM's built into the OS. Here's a Linux CD, try this instead."
>
> You would be surprised how good you can make a mediocre singer sound if you mic them just right, use the best take, and choose your effects carefully.
<JOB_ROLE=RIAA_AGENT VOICE_MODE=HONEST>
What kind of talent scout are you? Sure, they can sing, but what about their tits?
</JOB_ROLE>
(Or is that what you meant by "choosing their effects carefully" ;-)
Interesting.
I've been wondering about that for a while.
I've got a wide range of musical tastes. Funk, disco, new wave, house, techno, industrial, EBM, metal, rap, hip-hop, hell, I'll listen to damn near anything except what's on today's Top-40. I've been listening to some of those genres for 20 years.
Drop by alt.binaries.sounds.mp3.19[5678]0s. You'll see lots of music from each decade.
Drop by alt.binaries.sounds.mp3.rap-hiphop or .dance and things are different. You'll see almost nothing older than the mid-90s. Most of the stuff is last year or newer.
That's not to say that I expect to see Public Enemy's "Nation of Millions" posted every week (though Pink Floyd fans seem to have no trouble reposting Dark Side of the Moon with astonishing regularity :), but the lack of history in the genre groups is startling.
Some of this is due to things like (e.g. dance) the importance of the DJ over the actual tracks and the relative unavailability of the latest mix from DJ $SOME_GUY_IN_IBIZA each week. But it's a little unnerving to see almost a complete absence of "old-school" music in these two sub-genres, as both rap and house/techno variants have histories going back to the early 80s.
The old geezer in me says "Young punks just want the new stuff, they've got no respect for the classics that defined their genre. Buncha cable kiddies who think anything invented more than a week ago is obsolete and not worth listening to anymore".
But I don't trust myself when I start thinkin' like a geezer. It usually means I'm missing something important.
I think the poster to whom I'm replying has stumbled onto the important something I missed.
3072 MB / 720 hours
3145728 KByte / 43200 mins
25165824 kbit / 2592000 secs
9.7 kbps
Which reminds me.
Suppose we take a typical US dialup user and have them use the hours between midnight and 0600 for my USENET MP3 l33ching, That's six hours a day, or 180 hours a month, to download stuff.
I'm sure my ISP would be pissed at you (and righteously so!) if you were to use six hours a day during peak time, but by doing bulk downloads in the off-hours, you're not costing them anything - the modem pools are mostly empty, and even your local phone company's network is unloaded.
Six hours at 48000 bps = about 12M per hour of MP3s from USENET once you subtract out the uuencoding bloat and allow for some latency. 72M per day, or about 2G per month. If we assume 160kbps MP3s, that's about an hour's worth of new music every day. Egad.
(Given a 2-3G per month cap, why bother with broadband? ;-)
Great point!
Consider the MP3 leech from an ISP's point of view. Would you rather have 10000 users downloading (transiting) 3GB per month of MP3s from Supernews or Newsguy, or would you rather have them doing it from your own news server?
One way, you pay 300GB per day (a full NNTP feed) for 9TB per month transit to your NNTP server. After that, it's all internal traffic to you.
The other way, you pay 3GB per month times 1000 users, for 30TB per month to someone else's NNTP server (or to P2P servers located outside your network, etc...)
As "bad" as the situation is wrt USENET and binaries, for a sufficiently-large ISP, there may be a business case for upgrading/maintaining your own news server farm, rather than expecting your users to outsource it.
The same applies to P2P servers within the network:
Suppose you had a news server with only one day retention, but 100 of your 10000 USENET users each downloaded 1% (2.5G per day) of the feed and redistributed it by setting up their own NNTP servers for "Stuff They Liked". If each of these 100 individuals, who would only need to eat 2.5G per day off your news server (and it's all LAN traffic, so you don't pay for their transit), dropped $150 on a 30G hard drive, you could effectively have 10 days' retention, and take some of the load away from your news server.
I'm sure similar economies apply to Gnutella and more pure P2P solutions. I'm just using NNTP as an example.
True in the early part of the 20th century, when electronic distribution at zero cost was impossible, leaving for-pay distribution at moderate cost as the only way for the "Rock and Roll" derived from early Jazz/R&B musicians to make its way into the ears of Joe Consumer and sweep the nation in the 1950s.
False in the early part of the 21st century, when electronic distribution at zero cost was widely available, and the recording industry served only to prevent Joe Consumer from hearing the early Tibetan Electronica artists whose work set the stage for the solar-system-shaking grooves that rocked the colonies of Luna and Mars in the 2180s.
Suppose they pulled the plug on useful hardware today. Every manufacturer on the planet goes to copy control, no exceptions, and nobody ever cracks it. (This is of course, impossible. But let's go with the doomsday scenario for a bit.)
Given the number of ATs and XTs at my local surplus store, and the time it's taken to have them gradually replaced with 386s and 486s, I'd say we have at least a 10-15 year supply of useful hardware ahead of us.
And given the age of some of the older pieces of hardware in my collection, I'd say we have at least another 10 years, probably more, before that supply of useful hardware starts to fail.
If Copy Control Doomsday happened tomorrow, we'd have about 25 years before we had to worry. Spare parts purchased now, run for 6 months (to shake out any cases of "infant mortality") and stored in anti-static bags, will be just as good 25 years from now as they are today.
For less than $1000, you can buy enough hard drive storage and multiple sets of spare parts to store 100G of MP3s and have at least one system capable of playing them back for the rest of your natural life.
My old CD burning software didn't care about copy control.
My old CD burning software did things my new CD burning software doesn't do.
My old CD burning software was more functional - I could do more things with it than I can the new version.
What's wrong about it is that there are people trying to pass off downgrades as upgrades.
If your local Porsche dealer said "By the way, the new model Porsche has a rev-limiter hooked up to a GPS system that prevents you from going faster than 55 mph! It's so much better than last year's model!", you'd slap him silly, and you'd be right to slap him silly whether you ever intended to drive over the speed limit or not.
Download all patches for Windows-based CD-burning software today.
Install Linux tomorrow.
>
> You might find that the fan has trouble starting up on the lower voltage supply, but runs fine once it gets moving. In this case, a big capacitor in parallel with the resistor will supply the full voltage to the fan for a short time after the power is applied.
Actually, I had exactly the problem when I tried running a huge-ass 120mm 24V fan at 12V. The resistor-capacitor circuit wouldn't help there, but it'd solve the equivalent problem for 7V fans quite nicely. Thanks!
Actually, he did. "What are the green squares for?" was replied with an example along the lines of "Suppose you doubled your bet with every spin of the roulette wheel (and there were no green squares), every time you lost. You'd be guaranteed to win. The problem is that if you double your bet every time, you've gotta have more money than I [Dad, playing the house] do. The green squares make sure that even a guy with a million bucks can't try this."
Of course, we did it out with pen and paper and "real numbers" ($2, $4, $8, $16...) instead of exponents, but the effect was the same.
(Yeah, it was Reno, and they had a $5 table. I could hack a $20 minimum with $100 to play with, but anything beyond that is out of my league. It stops being fun when it's about real money. If I wanna do that, I'll play the stock market :)
> I like the games themselves, but I don't like the depressingly mundane (yet garish) surroundings. The "entertainment" in those places is always meant to appeal to one or two previous generations
You have a gift for understatement :)
Hey, what's wrong with the mob?
Compare the payouts on the mob-run "numbers game" in with the payout ratio of the government-run lotteries. The mob offers better odds, according to what appears to be expert testimony on the subject:
SOURCE: Q&A session on the National Gambling Impact Survey, March 17, 1998
Given the choice, it looks like the mob offers a superior product in the gambling market. Which reminds me, both the government and the Mafia have programmes whereby they offer "protection" you didn't ask for in exchange for "protection money" you have to pay. Both have the power to see that Bad Things Happen to you if you don't pay up. What's the difference, really? (There are days when I think the only reason the government fights organized crime is because it can't stand the competitionBut the computer said there was! ;-)
Actually, I was aiming for (+1, Funny), but someone could get an easy (+1, Insightful) by pointing out that what I just described as "preposterous, who'd be dumb enough to play cards when you can't see the cards being dealt and you don't have the source code of the algorithm that deals the cards?" is exactly what players of VLTs (Video Lottery Terminals) and electronic slot machines are doing right now in the real world.
Amen.
Actually, I have fond memories of my Dad teaching me (at age 7-8) how to play poker, blackjack, and roulette.
After spending some time taking my pennies, we both noticed that he was doing well, and so he started teaching me why the house always won. "Daddy, what are the green numbers on the roulette wheel for?" is actually a pretty natural question for a seven-year-old.
So we switched sides, and I played the "house" in blackjack and roulette the next time. All of a sudden, my "luck" changed and I "won" my pennies back. And he taught me that luck had nothing to do with it. If you play enough hands and have an advantage, you will win. And the rules are set up so that the house has an advantage.
Was it real probability theory? Not really. But when a kid's just discovered fractions, it's pretty easy to show him that 18/38 is less than 18/36. If betting on black/red is like flipping a coin, it's a fair game. Except that two out of every 38 spins, the coin is designed to land on its edge.
Since then, I've gone to casinos and had plenty of fun. But I've always gone in with cash, and with the notion that my $40 in gambling money is purchasing me a few hours of fun at the blackjack table - in the same way that the same $40 could purchase me few hours of fun in a bar, at the movies, or at the arcade.
Four of Hearts, Seven of Clubs.
Dealer is showing:
Unknown card, Six of Diamonds
Will you: Hit, Stand, or Double-down?
(Double-down)
Host: dealer.casino.nv.us contacted...
18K read (at 943 bytes/sec)
Player draws: Twelve of Clubs.
Player has: Four of Hearts, Seven of Clubs, Twelve of Clubs. Total=22.
Player Busts.
Dealer draws: Ace of Hearts.
Dealer stands.
Dealer turns over unknown card: Four of Clubs.
Dealer has: Four of Clubs, Six of Diamonds, Nine of Hearts. Total=21.
Dealer wins.
Wager again? Yes/No.
("Aaw, shit, again?!?!?! I always draw a twelve when I'm winning!")
You can deactivate this (mis-)feature under Internet_Options -> Advanced -> Search from the Address bar -> (Radio Button) "Do not search from the Address bar".
Of course, 99% of people don't do that, because it's the default, as are things like "Browsing -> Automatically check for IE updates" and "Enable Install On Demand" (which just sounds like an accident waiting to happen).
Back on topic, these sites get so much traffic because they're the default home pages. If users are too dumb to change the default home page (and I've seen usability studies that show many users don't even know the default home page can be changed - they think of the default home page as "The Internet"), they're too dumb to secure themselves against stuff like this.
This isn't an anti-M$ rant. Nutscrape blows hot donkey meat when it does the same thing with "What's Related".
For lack of a better term, I call it "stealth spyware".
But if that's not enough - just firewall out the offending domain.
My first experience with pop-unders was on clicktv.com, a TV listings site. In order to see what episode of Star Drek is being aired, you need Javashit enabled. Unfortunately, this also cranks up the popunders.
A glance at the HTML source reveals that the page with the TV listings comes from http://www.clicktv.com, but the Pop-under comes from http://media.fastclick.net. (Clever pigfux0rz, they even thought to dc.write ('<scr'+'ipt language=Javashit' to dodge proxies that filter out the string script language="Javashit")
So I said "fuck 'em", and stuck media.fastclick.net into my C:\WINDOWS\HOSTS, /etc/hosts or Junkbuster proxy file depending on my platform or whether I was running Junkbuster, and the problem went away.
General solution: treat the source of pop-ups and pop-unders the way you would Doubleclick. Firewall 'em. Most of the time, it's an HTTP transaction to a tracker that loads the offending Javashit. Firewall the tracker and the Javashit doesn't show up. If the Javashit can't load, the window don't pop.
Depends on the AMD ;-)
From personal experience, a K6-III-333 (grey-market laptop part available at Fry's) or K6-2+-450 (make sure it's the "2+" version) underclocked to 200 MHz will outperform (the hell out of) a P166. I run one in an I-Opener, cooled passively, and the mobo temperature never goes above 46C. (And that's with a big power transistor about two inches away... without that transistor, I think it'd be around 40C.)
If your P166 box happens to be running an ASUS TX-97 series board, search for beta BIOS 0112. This version allows you to run VCore down to 2.0V, 2.1V, and 2.2V. That won't help you run your P166, but the K6 CPUs I mentioned require the lower voltages, and are available for around $30-40. And they all have lots (256K on the K6-3-333, 128K on the K6-2+) of on-die cache compared to the P166.
Same MHz, better performance, lower voltage, lower temperatures. AMD CPUs that never made it into laptops are great!
I hadn't thought about that - the guy's got a point. A short on the fan would draw lots of current, and could pull +5V line up to +12V. It wouldn't have to get there before the P/S shut itself down - even +6 or 7V would probably be enough to do damage.
I came close to this when a +12V fan failed catastrophically. The cheap-azz bearings caused the fan blade to wiggle out of position. A blade hit one of the holes in the case and got stuck, which presumably shorted out the fan motor a few seconds later.
I'm gonna hit the surplus stores and hunt down some 12V fuses. I'm thinking that something at 1.5-2 times the rated fan current (probably 250mA tops) in series with the fan before hooking it up should do the trick. The goal is to have the fuse blow on short before the P/S +5V line gets drawn too high.
And with the price of fuses at a surplus store, I could probably do some destructive testing on a junk mobo to verify the fuse is doing the right thing. Lord knows I have enough "dead" (they still turn, but they whine like hell because of dead bearings) fans in my junk parts box!
The fan will run slower, but cooler.
2) Work on airflow near your fan blades. A fan with a great big sheet of metal with punched holes in it will be loud. The same fan with the metal grill removed will be quieter. The same fan with the metal grill and some extra space around it (because these fans typically blow air out in a cone on a 45-degree angle away from the center of the fan) will run even more quietly.
Still need finger protection? Get a real fan grill - the old-sk00l things that looked like three or four concentric circles of wire stuck together with a couple of cross-wise pieces of wire.
3) Rule of thumb - low RPM = low noise. If you don't get enough airflow (for cooling purposes) when you undervolt your 80mm case fan, carve up the case and add a big-ass 120mm fan. An undervolted 120mm fan can often move as much air as a typical 80mm fan running at +12V. If your local surplus store is well-stocked, you might even find some +24V fans that run at +12V. (But be sure to test them first ;-)
Well, there are a few ideas to start with. I'm sure others will follow up.