For those of you looking for more technical explanations, NTRU's website has a detailed discussion of their algorithm. The algorithm was published at CRYPTO96, so while it hasn't been thoroughly tested yet (and the la st I heard is that there are some implementation problems), it has been out for a while, and looked at by the best (Shamir, Coppersmith). This is no TriStrata.
As for its use, most of you are forgetting that the average person is willing to pay for convenience. Sure, it's easy to intercept the signal at the soundcard, or record it off your speakers, but the average person isn't going to go to that trouble, provided that the price is reasonable ($20/mo for on-demand access to the majors' catalogs, e.g.).
Doesn't even need to be that much. I used to burn CDs on my P75 wihout problems. I'd get occasional buffer underruns when doing other stuff at the same time, but since this is going to be a dedicated burning machine, you're not going to have much else happening anyway. Also, it was an older generation burner with a small (512K, IIRC) cache.
I have a dedicated burning machine, which is a 486/100, 20MB RAM. My CD-R drive is a Mitsumi 4802TE (4x write, EIDE), with a 2MB buffer. I use cdrecord for burning, and have never seen it report a cache less than 80% full. (i.e. a 512K buffer would be fine) My point is that if it's a dedicated burning machine, you don't need a PII or SCSI...
GramoFile is built exactly for removing clicks and pops from digitized LPs. Here's the homepage for GramoFile:
http://panic.et.tudelft.nl/~costar/gramofile/
GramoFile will not only remove clicks and pops, but it does so on the fly, and it will also automatically split the tracks for you... it's a neat program. Unfortunatley, it doesn't do noise reduction. You can check out:
http://www.sci.fi/~mjkoskin/
For dnr and denoi, two command-line denoising programs. I've used them a little, and they seemed okay, but I haven't gotten around to really testing them out.
You can also apparently use Broadcast 2000 for noise reduction, but I couldn't get it to work. It's homepage is:
I picture it as a star-rating for "maturity" and overall "quality" and a brief, independent description of the music. Not a full, serious review, but enough to give people the idea of whether they might like it, and whether it's worth downloading.
This implies that there's some absolute standard of musical quality, which is clearly untrue.
It's clearly untrue across genres, but it's not clearly untrue within a genre (though that depends upon the genre -- techno might be an exception). I think it's pretty easy to distinguish between serious, professional bands (yes, there are quite a few on mp3.com) and half-assed, amateur garage bands (yes, there are many of these on mp3.com).
I wouldn't personally trust anybody random stranger's taste in music enough to even bother to download what they considered worth listening to.
Which is one of the reasons an independent site would have regular reviewers, each of whom gives e.g. a description of their musical tastes, and a list of their favorite bands. Readers would be able to get an idea of how close their tastes match up with the reviewer's, and then weigh the reviewer's opinion accordingly. This is more-or-less what I've been doing through USENET and discussion boards for years, and with pretty good success. There are a number of people whose opinions I trust enough to buy on their recommendation alone. There are many, many more whose opinions I don't trust.
mp3.com keeps track of the most popular downloads, which is a pretty good idea. It seems that that would help you zero in on the "good" stuff, right? Unfortunately, I found that it wasn't very useful. Perhaps most people download indiscriminately, or perhaps they have poor taste.
...or perhaps they look through the lists starting with the most downloaded stuff, and never get past the first few pages... (after all, if it were any good, lots of people would've downloaded it, right?) Of course, it doesn't help that a lot of bands mischaracterize their own music, and "spam" it to completely unrelated categories... Also, since I have a fast connection, I tend to download everything from a particular artist before listening to it... and then I listen to it while downloading stuff from other artists. I never preview stuff before downloading it...
What I've really been longing for is a series of independent websites that act as indices into the massive mp3.com archive. An independent site could post reviews from people they pay to wade through all the noise, searching for the elusive signal.
I agree, though I don't think it's necessary to pay people. I'd volunteer to help out with a bunch of quick reviews a month... and I'm sure a lot of other people would too.
Such a system would have to be carefully designed of course, but even an independent opinion on what a band sounds like (not even a full review) and their relative "maturity" would be very useful.
Yes, as far as the internet is concerned. For some (most?) computer-illiterate people, "AOL == the WWW == the internet". Just because AT&T or Microsoft has a good brand name doesn't mean that their online service will benefit from that brand name. How many customers does AT&T WorldNet have these days? How about MSN?
AOL has, of course, been actively lobbying Congress for open cable lines, presumably so that they could use them. When they bought Time Warner, they quit lobbying Congress, presumably because they now owned a cable network and didn't need. Today, they open up their lines. Very interesting. Are they being un-hypocritical or do they have ulterior motives.
Of course AOL has ulterior motives. They think this will help them make money. That's the only reason they're doing it.
I expected AOL to open TW's cable lines, and I'll explain why. My numbers may be off, but I think they're close enough that you'll get the idea.
AT&T and @Home have about 75% of the cable ISP market. RoadRunner has about 25%. AOL figures that if they open their cable lines, they'll lose about 50% of their customers to other ISPs (because they have the strongest brand, and they can offer the best pricing). Of course, they're still making money on those customers, just not as much. So they stand to lose somewhat less than 12.5% of the market. Where I think they gain is that they can force the FCC to open AT&T's lines, thereby gaining about half of AT&T's customers (brand is king), or about 37.5% of the market. How will this happen? First, AOL implementing an open cable system shows that it's possible, so AT&T's objection on those grounds is thrown out. Further, I think they're hoping that AT&T/@Home customers will start objecting to their exclusive arrangement ("My brother gets AOL, why can't I?"). Finally, if prices are lower, AOL can show that the customer is helped by opening the cable lines. That might be enough to force the FCC to step in...
Ob. Disclaimer: IANAB (I am not a businessman). The above could be complete hooey. But it makes sense to me...
VA Linux priced at $30, but started trading at $300. It hit an intra-day high of $320 on its first day of trading.
RedHat split 2:1 on Jan. 10. It's actual peak was around $300 in early December. In addition, as late as November (it IPO'ed in August (?)), it was trading for less than $50 (split-adjusted).
Microsoft owns a silent, non-voting share of Inprise last time I checked.
This is true. MS's investment in Inprise was part of a settlement of the lawsuit Inprise filed against MS for stealing their developers.
I think there are many more reasons to be skeptical about this (somewhat shady) merger
I'm curious if you could elaborate on what you consider "shady" about the deal.
here's no reason at all why Corel (productivity/graphics company with a Linux distro targetted at beginners' desktops) needs to be in the same company with Inprise (tools/middleware company with upcoming Linux products targetted at high-end professional developers). There's just no fit there.
Sure there is... let me try to set it out the way I see it... (bear with me, I'm thinking this through as I go...)
What Corel brings: name brand office suite, desktop-oriented Linux distribution, high-end graphics apps
What Inprise brings: open source database (Interbase), app server, developers tools
While they may not "need" to be in the same company, I think combining them makes sense, because:
Inprise's developer tools will be building apps for Linux desktops. Corel is shooting for the Linux desktop market, so it makes sense that they'd want Inprise's developer tools to be "optimized" for Corel's distro.
Corel/Inprise is in the position to provide a total package to a small business. They have a desktop OS, a database, an app server, a server OS, and an office suite... all of which will presumably be designed to work well together.
The way I see it, Corel/Inprise has about six months to a year to get their act together... when MS starts dropping support for NT 4.x, small businesses currently using NT 4.x will have to upgrade. If they're a really small business (i.e. too small for a full-time sysadmin), then a complete package like what Corel/Inprise can deliver is about their only alternative to Win2000. Further, it could be a very compelling alternative, since it'll be a lot cheaper (both outright, and due to the fact that the hardware won't have to be upgraded as well).
So basically, Corel is shooting for the business/consumer desktop, and Inprise helps them on both fronts, and doesn't hurt them anywhere. About the only negative to the whole deal I see is Cowpland.
On the day of the IPO, when its price rises to ridiculous levels, short sell it.
Nice idea, but you can't short it until 30 days after its IPO. If (by some miracle) LinuxOne hasn't crashed and burned by then, then it is not being valued on fundamentals (heh, as if RHAT was...), and these types of stocks make lousy short candidates.
I'm thinking that this would be a great stock to short the afternoon of the IPO.
It would be... if it were possible to short an IPO on the first day. It's not. You have to wait something like 30 (60?) days. By then, LinuxOne will have crashed and burned. If it hasn't, shorting is an intriguing idea...
As for secret communications between the companies, IIRC that's collusion. I don't know what the punishments for collusion are off the top of my head, but they're most likely to be pretty nasty.
Given MS's reaction the last time they had a run-in with the DOJ, (Didn't Gates brag about MS not changing their practices at all?) I wouldn't put this type of collusion past them.
In any case, as I mentioned, proving such collusion would be much harder than the DOJ's current case. For one thing, MS might be smart engouh this time to not leave an email trail. Further, since the CEOs of the new companies are likely to be from MS, they could very well be friends. What's to stop friends from getting together every Friday night for a dinner party? How could you prove that collusion is going on at those parties (legally)?
Having multiple companies working on Windows compatible OS's will *not* fragment it.
I think that it would just be a matter of time before all of the companies involved are adding their own undocumented API calls, etc... they need something to give them an edge. This would end up making all of the versions "Windows compatible", but some of them more compatible than others. Each company would have a version of Office that works best on its version of Windows, etc.
Just like, if the Windows API was documented and WINE supported it perfectly, that wouldn't result in fragmentation.
That's only because one company controls the API, and the other is playing catch-up. The situation is more likely to be like what happened to UNIX, or more along the lines of what's happened to HTML: you have two or more companies adding their proprietary "enhancements" to the core API, resulting in a bunch of versions of Windows which all handle the current stuff in more-or-less the same way, but for which new technologies are handled completely differently.
The purpose is to benefit consumers. And consumers certainly will benefit
I don't know about that.... one definite side-effect of this kind of breakup is end-user confusion as to what the differences are between versions of Windows. If you're correct and Windows doesn't fragment, then the differences between the various versions are mostly superficial. How does one go about choosing which version of Windows when one has nothing to base one's decision upon?
>could somebody share their professional opinion of what a breakup would result in?
Well, I don't have a professional opinion, but I'll give it a go... From what I've read, there are two breakup scenerios being thrown around:
Split MS into n "Baby Bills": smaller versions of MS, each with the Windows source code, and each with (some subset of) the various other departments (hardware, Office suites, games, etc.).
Split MS into some combination of distinct companies, based upon some partitioning of the departments (operating systems, hardware, games, office suites, etc.).
The basic effect of 1 I see is that Windows (and perhaps Office, etc...) become fragmented. I don't see much of a direct benefit to Linux/Be/etc, until the versions of Windows become drastically different (which won't happen for many years, if at all).
The effects of option 2 are harder to predict, but it opens the door for things like MS Office being ported to Linux/Be/etc. The big problem with 2 is making sure that the separation takes place in more than name. I've heard (completely unsubstantiated and probably paranoid) rumors that secret channels of communication have already been set up in case of a break-up to keep the app programmers up-to-date with the new undocumented features of the OS, etc. Whether or not that's the case, it would be very hard to show e.g. that MS-App's decision to not port Office to Linux was based only upon Linux's market share, and not at least partly upon some agreement with MS-OS to help retain their desktop monopoly.
I don't know about most radio stations, but none of the three radio stations I've had contact with use any kind of file servers to play music from. They all use standard CD players, and possibly cassettes and vinyl.
I don't see why the record companies aren't actively pushing such solutions. It seems like it would cut their distribution costs dramatically, as well as increasing their sales (since many of those promo-only CDs somehow end up in used CD stores...).
I wouldn't be too sure of that... I remember reading a story about the guy who does the voice for the Taco Bell chihuahua. He makes well into six figures, just for doing a few commercials, the voices for the talking chihuahuas that TB sold, etc...
Anyway, as previously mentioned by a bunch of posters, it's very difficult to be able to do two dozen distinctly different voices. If you're going to do a show like the Simpsons, where you have a huge number of characters, would you hire a dozen different people to do voices, or would you hire one person (Harry Shearer, e.g.) who can do not only the two dozen voices that need doing now, but pretty much any new characters you come up with as well?
>It's the old, tired-but-true Pascal's Wager that is used so often against atheists: "If you >atheists are right and there is no God, then when I die, nothing will happen. But if we Christians >are right and our God does exist, then when you die, you're going to fry for eternity. Therefore, >worship God, because He's a safe bet!" What a lousy reason to believe.
Actually, it's even worse than that... if the only reason you believe in god is Pascal's Wager, then you don't truly "believe" in god. You're essentially faking it... hoping to sneak into heaven by fooling god. If god is willing to condemn you to the bowels of hell for not believing in him, what do you think he'd do to you if he found out you thought you could fool him?
Slashdot Mirror
As for its use, most of you are forgetting that the average person is willing to pay for convenience. Sure, it's easy to intercept the signal at the soundcard, or record it off your speakers, but the average person isn't going to go to that trouble, provided that the price is reasonable ($20/mo for on-demand access to the majors' catalogs, e.g.).
#strings /sbin/shutdown
...
/dev/%s
...
tyler
Oh hello Mr. Tyler - going DOWN?
@(#) shutdown 2.73 25-Nov-1997 MvS
(Yes, yes, I know... use the source... well, I'm too lazy... you guys figure it out.)
On an unrelated note, early versions of MATLAB (before it became a commercial package) would reply to the command "fuck" with "your place or mine?"
It's spelled: Millennium.
http://www.geocities.com/SunsetStrip/Towers/4513/
click on the "Escape from darkness" button... :)
I have a dedicated burning machine, which is a 486/100, 20MB RAM. My CD-R drive is a Mitsumi 4802TE (4x write, EIDE), with a 2MB buffer. I use cdrecord for burning, and have never seen it report a cache less than 80% full. (i.e. a 512K buffer would be fine) My point is that if it's a dedicated burning machine, you don't need a PII or SCSI...
http://panic.et.tudelft.nl/~costar/gramofile/
GramoFile will not only remove clicks and pops, but it does so on the fly, and it will also automatically split the tracks for you... it's a neat program. Unfortunatley, it doesn't do noise reduction. You can check out:
http://www.sci.fi/~mjkoskin/
For dnr and denoi, two command-line denoising programs. I've used them a little, and they seemed okay, but I haven't gotten around to really testing them out.
You can also apparently use Broadcast 2000 for noise reduction, but I couldn't get it to work. It's homepage is:
http://heroine.linuxave.net/bcast2000.html
I picture it as a star-rating for "maturity" and overall "quality" and a brief, independent description of the music. Not a full, serious review, but enough to give people the idea of whether they might like it, and whether it's worth downloading.
This implies that there's some absolute standard of musical quality, which is clearly untrue.
It's clearly untrue across genres, but it's not clearly untrue within a genre (though that depends upon the genre -- techno might be an exception). I think it's pretty easy to distinguish between serious, professional bands (yes, there are quite a few on mp3.com) and half-assed, amateur garage bands (yes, there are many of these on mp3.com).
I wouldn't personally trust anybody random stranger's taste in music enough to even bother to download what they considered worth listening to.
Which is one of the reasons an independent site would have regular reviewers, each of whom gives e.g. a description of their musical tastes, and a list of their favorite bands. Readers would be able to get an idea of how close their tastes match up with the reviewer's, and then weigh the reviewer's opinion accordingly. This is more-or-less what I've been doing through USENET and discussion boards for years, and with pretty good success. There are a number of people whose opinions I trust enough to buy on their recommendation alone. There are many, many more whose opinions I don't trust.
What I've really been longing for is a series of independent websites that act as indices into the massive mp3.com archive. An independent site could post reviews from people they pay to wade through all the noise, searching for the elusive signal.
I agree, though I don't think it's necessary to pay people. I'd volunteer to help out with a bunch of quick reviews a month... and I'm sure a lot of other people would too.
Such a system would have to be carefully designed of course, but even an independent opinion on what a band sounds like (not even a full review) and their relative "maturity" would be very useful.
Yes, as far as the internet is concerned. For some (most?) computer-illiterate people, "AOL == the WWW == the internet". Just because AT&T or Microsoft has a good brand name doesn't mean that their online service will benefit from that brand name. How many customers does AT&T WorldNet have these days? How about MSN?
Of course AOL has ulterior motives. They think this will help them make money. That's the only reason they're doing it.
I expected AOL to open TW's cable lines, and I'll explain why. My numbers may be off, but I think they're close enough that you'll get the idea.
AT&T and @Home have about 75% of the cable ISP market. RoadRunner has about 25%. AOL figures that if they open their cable lines, they'll lose about 50% of their customers to other ISPs (because they have the strongest brand, and they can offer the best pricing). Of course, they're still making money on those customers, just not as much. So they stand to lose somewhat less than 12.5% of the market. Where I think they gain is that they can force the FCC to open AT&T's lines, thereby gaining about half of AT&T's customers (brand is king), or about 37.5% of the market. How will this happen? First, AOL implementing an open cable system shows that it's possible, so AT&T's objection on those grounds is thrown out. Further, I think they're hoping that AT&T/@Home customers will start objecting to their exclusive arrangement ("My brother gets AOL, why can't I?"). Finally, if prices are lower, AOL can show that the customer is helped by opening the cable lines. That might be enough to force the FCC to step in...
Ob. Disclaimer: IANAB (I am not a businessman). The above could be complete hooey. But it makes sense to me...
You can see all of the details at: Yahoo!.
This is true. MS's investment in Inprise was part of a settlement of the lawsuit Inprise filed against MS for stealing their developers.
I think there are many more reasons to be skeptical about this (somewhat shady) merger
I'm curious if you could elaborate on what you consider "shady" about the deal.
here's no reason at all why Corel (productivity/graphics company with a Linux distro targetted at beginners' desktops) needs to be in the same company with Inprise (tools/middleware company with upcoming Linux products targetted at high-end professional developers). There's just no fit there.
Sure there is... let me try to set it out the way I see it... (bear with me, I'm thinking this through as I go...)
- What Corel brings: name brand office suite, desktop-oriented Linux distribution, high-end graphics apps
- What Inprise brings: open source database (Interbase), app server, developers tools
While they may not "need" to be in the same company, I think combining them makes sense, because:- Inprise's developer tools will be building apps for Linux desktops. Corel is shooting for the Linux desktop market, so it makes sense that they'd want Inprise's developer tools to be "optimized" for Corel's distro.
- Corel/Inprise is in the position to provide a total package to a small business. They have a desktop OS, a database, an app server, a server OS, and an office suite... all of which will presumably be designed to work well together.
The way I see it, Corel/Inprise has about six months to a year to get their act together... when MS starts dropping support for NT 4.x, small businesses currently using NT 4.x will have to upgrade. If they're a really small business (i.e. too small for a full-time sysadmin), then a complete package like what Corel/Inprise can deliver is about their only alternative to Win2000. Further, it could be a very compelling alternative, since it'll be a lot cheaper (both outright, and due to the fact that the hardware won't have to be upgraded as well).So basically, Corel is shooting for the business/consumer desktop, and Inprise helps them on both fronts, and doesn't hurt them anywhere. About the only negative to the whole deal I see is Cowpland.
--
On the day of the IPO, when its price rises to ridiculous levels, short sell it.
Nice idea, but you can't short it until 30 days after its IPO. If (by some miracle) LinuxOne hasn't crashed and burned by then, then it is not being valued on fundamentals (heh, as if RHAT was...), and these types of stocks make lousy short candidates.
I'm thinking that this would be a great stock to short the afternoon of the IPO.
It would be... if it were possible to short an IPO on the first day. It's not. You have to wait something like 30 (60?) days. By then, LinuxOne will have crashed and burned. If it hasn't, shorting is an intriguing idea...
Given MS's reaction the last time they had a run-in with the DOJ, (Didn't Gates brag about MS not changing their practices at all?) I wouldn't put this type of collusion past them.
In any case, as I mentioned, proving such collusion would be much harder than the DOJ's current case. For one thing, MS might be smart engouh this time to not leave an email trail. Further, since the CEOs of the new companies are likely to be from MS, they could very well be friends. What's to stop friends from getting together every Friday night for a dinner party? How could you prove that collusion is going on at those parties (legally)?
I think that it would just be a matter of time before all of the companies involved are adding their own undocumented API calls, etc... they need something to give them an edge. This would end up making all of the versions "Windows compatible", but some of them more compatible than others. Each company would have a version of Office that works best on its version of Windows, etc.
Just like, if the Windows API was documented and WINE supported it perfectly, that wouldn't result in fragmentation.
That's only because one company controls the API, and the other is playing catch-up. The situation is more likely to be like what happened to UNIX, or more along the lines of what's happened to HTML: you have two or more companies adding their proprietary "enhancements" to the core API, resulting in a bunch of versions of Windows which all handle the current stuff in more-or-less the same way, but for which new technologies are handled completely differently.
The purpose is to benefit consumers. And consumers certainly will benefit
I don't know about that.... one definite side-effect of this kind of breakup is end-user confusion as to what the differences are between versions of Windows. If you're correct and Windows doesn't fragment, then the differences between the various versions are mostly superficial. How does one go about choosing which version of Windows when one has nothing to base one's decision upon?
Well, I don't have a professional opinion, but I'll give it a go...
From what I've read, there are two breakup scenerios being thrown around:
- Split MS into n "Baby Bills": smaller versions of MS, each with the Windows source code, and each with (some subset of) the various other departments (hardware, Office suites, games, etc.).
- Split MS into some combination of distinct companies, based upon some partitioning of the departments (operating systems, hardware, games, office suites, etc.).
The basic effect of 1 I see is that Windows (and perhaps Office, etc...) become fragmented. I don't see much of a direct benefit to Linux/Be/etc, until the versions of Windows become drastically different (which won't happen for many years, if at all).The effects of option 2 are harder to predict, but it opens the door for things like MS Office being ported to Linux/Be/etc. The big problem with 2 is making sure that the separation takes place in more than name. I've heard (completely unsubstantiated and probably paranoid) rumors that secret channels of communication have already been set up in case of a break-up to keep the app programmers up-to-date with the new undocumented features of the OS, etc. Whether or not that's the case, it would be very hard to show e.g. that MS-App's decision to not port Office to Linux was based only upon Linux's market share, and not at least partly upon some agreement with MS-OS to help retain their desktop monopoly.
I don't know about most radio stations, but none of the three radio stations I've had contact with use any kind of file servers to play music from. They all use standard CD players, and possibly cassettes and vinyl.
I don't see why the record companies aren't actively pushing such solutions. It seems like it would cut their distribution costs dramatically, as well as increasing their sales (since many of those promo-only CDs somehow end up in used CD stores...).
>Plus I'm sure the money isn't that great,
I wouldn't be too sure of that... I remember reading a story about the guy who does the voice for the Taco Bell chihuahua. He makes well into six figures, just for doing a few commercials, the voices for the talking chihuahuas that TB sold, etc...
Anyway, as previously mentioned by a bunch of posters, it's very difficult to be able to do two dozen distinctly different voices. If you're going to do a show like the Simpsons, where you have a huge number of characters, would you hire a dozen different people to do voices, or would you hire one person (Harry Shearer, e.g.) who can do not only the two dozen voices that need doing now, but pretty much any new characters you come up with as well?
>It's the old, tired-but-true Pascal's Wager that is used so often against atheists: "If you
>atheists are right and there is no God, then when I die, nothing will happen. But if we Christians
>are right and our God does exist, then when you die, you're going to fry for eternity. Therefore,
>worship God, because He's a safe bet!" What a lousy reason to believe.
Actually, it's even worse than that... if the only reason you believe in god is Pascal's Wager, then you don't truly "believe" in god. You're essentially faking it... hoping to sneak into heaven by fooling god. If god is willing to condemn you to the bowels of hell for not believing in him, what do you think he'd do to you if he found out you thought you could fool him?