Always thought/usr was for, well, users, the stuff in/bin and/sbin mostly being used by programs prior to the mounting of the/usr partition (if there is one)./usr/local then refers to the local versions of user software.
User programs that use the etc hierarchy always use/etc for the system etc files, but for a while it was fashionable to have/usr/etc store the "not any of the above" files specific to userspace applications. The same then applied to/usr/local/etc for local versions of user tools.
Programs that needed their own tree, like X11, OpenLook, or whatever, created their own directory off/usr and built exactly the same layout for themselves on a local basis.
This all makes perfect sense, requires no acronymitis, and explains a lot of how Unix got along for so long without a "standard base" specification. If anything, attempts to eliminate some of the directory hierarchies in modern Unix software is actually making it much harder to find anything and much riskier to install software, due to the increased risk of namespace collisions. As none of the older packages considered there to be any risk - they were off in their own isolated namespace - none of the older packages take any care over their naming conventions.
I'm restricted by NDA and by corporate policy in what I can say, so I won't comment on the article itself. I will say though that most newspapers are nothing more than paid advertising that run a few real articles as typing exercises.
Either way, however, I do not believe for one moment that AAC and Ogg Vorbis are the only unencumbered audio codecs out there. Besides, last I heard, Ogg Vorbis had limited value when you start pushing the envelope on sounds, whilst AAC is only good at certain frequency bands and can't encode sound "textures" at all. True audiophiles loath and detest CDs because 16 bits stereo at 44.1 KHz over variable latency circuits accumulates just too much distortion. Professional digital instruments now all work at 24 bits, and top-end analog-to-digital converters are 26-bit. (IIRC, the top-of-the-top-end is 26-bit, 1 gigahertz.) Professional sound systems are now 11 speakers plus 1 subwoofer, and most movie DVDs these days are at least 5.1.
These are the targets to hit. Never mind uncle Billy's 78s, or cousin Joe's 45s. Sure, encoding those is cool, but your average user is infinitely more likely to be watching the latest movie or a trailer for the IMAX down the street.
IMHO, MP3 is so wide of the real mark that it's not worth bothering to consider as competition. People listen to MP3s for the same reason they watch camcorder-copied movies - it sucks so bad it's not funny but realistic alternatives aren't always available. If you gave them a realistic alternative that could play professional-quality sound at very close to professional quality without requiring them to buy terabytes of extra space, I'm as certain as I can be that people would adopt it. What is the point in them buying high-end professional sound cards if there are no sounds to play? It's as simple as that.
Arxiv doesn't appear to carry the paper, and only two papers in it relate to mock theta functions at all. One of them is a transformation formula for second-order mock theta functions and the other talks about mock theta functions as quantum invariants, whatever that means. A glance at the paper suggests that mock theta functions relate to a key element in topology, but my maths isn't nearly good enough to tell you exactly what is being described.
It would have to be imaginary, or complex. But that's a bit of a tangent from the point. The TFA is obtuse, cos() it doesn't exp()lain anything much. It would seem that the Slashdot crowd are caught on it Hooke, line and sinker, though. Of course, any maths problem is as easy as Pi, if you use sufficiently advanced techniques. However, if the problem cannot be differentiated meaningfully, can it be integral?
If they rewrote XRoach as a secure Java applet, then the cockroach would climb around the screen and hide behind windows until you squash it by clicking on it. Hmmm. Actually, that's not a bad idea - can someone on the Slash code development team add this?
When there's money left over, the budget is reduced the following fiscal year by the excess amount the prior fiscal year. That's the way the UK Government funds things, which is why there is a desperate scramble at the end of the year to buy useless, expensive items to flush the rest of the budget. Take a look at the spending pattern of any of the UK county councils.
Ah, well if we're including Linux, then I had SLS Linux running on a Viglen 386SX-16 with 5 megabytes of RAM (5? Yes, well, Viglen was always known for being a little odd. You had the 4 MB of extended memory, but it then let you use the original 1 MB of RAM as well.)
Actually, it worked pretty well. I had X and OpenLook running and was able to run a Netrek server with 19 robot players and myself on a regular Netrek client. That's not bad going, given that Netrek was not the most elegant of programs. (Hell, olvwm was not exactly a masterpiece of coding. I'm amazed to this day that I could get a full GUI environment that showed no obvious lag on such a system.)
Probably the most surreal experience, though, was running Windows under dosemu under Linux.
Under the law, you can't go around make defamatory claims for the sake of it. The first amendment does not protect slander or libel, although in the US it can be tough to prove either of these. In this case, and with the rulings from the SCO case to back them up, it would seem possible to maybe get a restraining order put on Balmer. Any failure by Microsoft to prove or explain their case at that point could be seriously damaging to Microsoft, simply because businesses would start to associate their claims with SCO's claims.
Unfortunately, this is unlikely to happen. Most don't have the money or the inclination to go against an 800 lb. chair-wielding gorilla, and I can't blame them.
Part of the problem is that the Open Source community has opted against unionizing. This matters. It has left the community as a whole extremely vulnerable and has left it with no meaningful voice. We cannot collectively act against bullying and intimidation if there is no mechanism for collective action of this kind. The price we are paying for our individual freedom is our freedom as a community. Our ability to act collectively in programming is precisely why Open Source is as sophisticated, rich and diverse as it is. Those who act as self-isolated individualists are generally the ones who get scorned and ignored, their products soon shown to be inferior.
Yet we somehow persist in believing, despite the harassment, despite the intimidation, despite every scrap of evidence that we have to date, that we can simply wish away the abuse of the megacorps.
The letter, as others have noted, is so juvenile as to not be worth discussing. However, the concept of unified action is at least going in the right direction. Mere lobbying won't work. Have you ever seen anyone successfully lobby Microsoft? The European Union only succeeded by threatening to bleed Microsoft dry if it didn't comply, and even then Microsoft did everything in its power to obey the letter of the law only as little as humanly possible without being dragged into court by its collective toenails.
No, merely asking the Evil Empire to play nice isn't going to achieve a damn thing, no matter how many names are put to what piece of paper. If you want Microsoft to cut the crap, then you need to have the strength to back up your request. Right now, we have... exactly nothing. There will be no chorus of heavenly angels throwing hellfire and brimstone at Redmond, there will be no Congressional investigations, you won't even get IBM to contribute to your tombstone.
Yes, opposition to this abuse is essential, but show ME the code. Show me how the community is to achieve this, when it took three companies two decades to bring Microsoft to account over DR-DOS. How do we achieve what the DOJ were unable to do? When we have refused ourselves the right to be heard?
Thanks for those references. Of course, most of this is redundant, as the military already have a program on hypersonic scramjets. Use a jet first-stage, essentially the same method as used by the X-Prize winner, and you can get far greater efficiency than from a rocket alone. (Basically the design proposed by the Advanced Rocket Launch Assist - ARLA - system.) The problem is starting off, but the military have long-ago been using turbine-assisted ramjets to solve that problem.
The idea behind assisted ramjets is that you only need the air to be travelling at a high speed, not the vehicle. A standard ramjet will operate at airspeeds of 400 MPH, well within the capabilities of a simple propeller system to provide. A hydrogen-fuelled ramjet will max out at mach 7, which is about the speed a scramjet should ignite. Just stage 'em up. By the time the scramjet finishes, you'll be close to escape velocity and will barely need the rocket stage at all.
Teletext was a broadcast system, and broadcast systems do suck. Broadcast is wasteful of resources when few people want something, in the same way point-to-point is wasteful when more than a few want something. Multicast only uses resources if someone at the other end uses it.
Teletext was also a cyclic system. Pages were broadcast in sequence, until the one you wanted appeared. This is OK when there are a few pages - ie: when latency vastly exceeds the cycle time - but it doesn't scale well. Multicast delivery works by delivering content that is being requested to those who have requested it (and nobody else). If done right, it is not based on push/pull concepts, but is based around pooling common requests and pooling common delivery. If you want analogies, then it's efficient mass transit, not television.
And you'd better tell Nokia that their content delivery system sucks, because they clearly don't appear to know it all by themselves.
USENET issued "death sentences" (ie: disconnects) to ISPs that tolerated spammers. Shortlay after, ISPs stopped popularizing USENET. USENET is under-utilized for the simple reason that ISPs want the spammers' money more than they want to provide useful services. I could list a dozen protocols and services from "the old days" that would be a major improvement over current methods that are being mis-applied. And that is the key. There is nothing wrong with current methods, if they are used in the manner they are suited for, for the types of problems they work best with. You do not use a sledgehammer as a substitute crowbar, you do not use a Swiss Army knife (as multi-function as it is) as a substitute pick-axe. So why use PtP for a role it was never intended nor designed for?
Tier 1 backbone providers (AT&T, Sprint, and so on) all have PIMv2 enabled on the backbones, probably in sparse mode, along with most tier 2 backbone providers. Dense mode (which is the same model DVMRP used) doesn't make any sense for the sorts of software people are actually using, so most people ignore it. The third PIMv2 method (bi-directional multicast) would however make the most sense if you were to have P2P applications make use of multicast.
Modern IP multicast - ie: ignoring DVMRP and MOSPF - isn't too bad when it comes to distribution. Pruning and grafting of branches is now more-or-less solid ground. IGMPv3/MLPv2 support authentication extensions, source-specific multicast and other cool stuff. IPv6 multicast is intended to be interoperable with Infiniband multicast (though as the number of users of either is extremely limited, this one is of limited value for right now). As a raw transport, it's not shabby. Now, to get anything useful in the way of information systems done, you need to layer a reliable multicast transport on top of that. SRM and NORM are the big two players in this arena, with Open Source implementations of both. FLUTE was a potential big player, for file distribution, but there is a patent on multicast file sharing which has shut down most FLUTE implementations.
(Possibly one reason P2P software doesn't include a multicast option is precisely because of that patent. It's bad enough being vilified by the **IA lawyertroids, but being sued for patent infringement as well would likely cause some serious problems. It's not through a lack of means. A lack of programmers familiar with multicast might also be a problem.)
Oh, I dunno. Multicast information delivery is significant enough that the entire Internet backbone now deploys PIMv2 multicast routing as standard. You think they did this out the kindness of their hearts? Hardly. You think they did this for revenue? As most ISPs receive multicast but never forward it to customers, I'd love to see what this revenue would be.
Multicast information delivery systems have been tried? When? Where? Show me a single ISP that has delivered multicast to residential customers and a single non-trivial example of reliable multicast for information delivery. I consider the mbone utils of VIC and VAT to be trivial examples. Show me something real. A distributed database engine, a replication server that pushes a filesystem to all mirror sites simultaneously, an MPI implementation that uses reliable multicast for collective operations, a multicast SMTP server for sending mail to multiple destinations in a single transfer. If you can't show me the apps and you can't show me real-world residential users, then you can't possibly claim multicast information systems have ever been tried.
(Oh, and I don't consider multicast mosaic to be a meaningful example, either. It was never widely distributed, by the time anyone knew it even existed the world had switched to other browsers, and those who developed it did not exactly go out of their way to tell anyone it was there or what it did. Even if they had, the big thing of the time was Netscape - Mosaic was essentially dead. Why would anyone use a browser that was inferior in virtually every respect, simply to use a distribution model that their ISP would have blocked anyway?)
Slashdot Mirror
...is when they'll make something with negative reflectivity. A perfect vaccuum doesn't reflect, so zero reflectivity isn't terribly impressive.
My guess is that by "black", they only mean for a limited range of frequencies. In the infrared, it's probably some sort of violent crimson.
...planet. Y'know, it doesn't have the same ring(s) to it.
User programs that use the etc hierarchy always use /etc for the system etc files, but for a while it was fashionable to have /usr/etc store the "not any of the above" files specific to userspace applications. The same then applied to /usr/local/etc for local versions of user tools.
Programs that needed their own tree, like X11, OpenLook, or whatever, created their own directory off /usr and built exactly the same layout for themselves on a local basis.
This all makes perfect sense, requires no acronymitis, and explains a lot of how Unix got along for so long without a "standard base" specification. If anything, attempts to eliminate some of the directory hierarchies in modern Unix software is actually making it much harder to find anything and much riskier to install software, due to the increased risk of namespace collisions. As none of the older packages considered there to be any risk - they were off in their own isolated namespace - none of the older packages take any care over their naming conventions.
My mistake. That really was an anonymous howard.
You're lucke that you posted that anonymously. :)
'nuff said.
Find me a multicast-capable bus that electrically scales and I'll find you an answer.
I'm restricted by NDA and by corporate policy in what I can say, so I won't comment on the article itself. I will say though that most newspapers are nothing more than paid advertising that run a few real articles as typing exercises.
This is most definitely not point-to-point.
These are the targets to hit. Never mind uncle Billy's 78s, or cousin Joe's 45s. Sure, encoding those is cool, but your average user is infinitely more likely to be watching the latest movie or a trailer for the IMAX down the street.
IMHO, MP3 is so wide of the real mark that it's not worth bothering to consider as competition. People listen to MP3s for the same reason they watch camcorder-copied movies - it sucks so bad it's not funny but realistic alternatives aren't always available. If you gave them a realistic alternative that could play professional-quality sound at very close to professional quality without requiring them to buy terabytes of extra space, I'm as certain as I can be that people would adopt it. What is the point in them buying high-end professional sound cards if there are no sounds to play? It's as simple as that.
Arxiv doesn't appear to carry the paper, and only two papers in it relate to mock theta functions at all. One of them is a transformation formula for second-order mock theta functions and the other talks about mock theta functions as quantum invariants, whatever that means. A glance at the paper suggests that mock theta functions relate to a key element in topology, but my maths isn't nearly good enough to tell you exactly what is being described.
It would have to be imaginary, or complex. But that's a bit of a tangent from the point. The TFA is obtuse, cos() it doesn't exp()lain anything much. It would seem that the Slashdot crowd are caught on it Hooke, line and sinker, though. Of course, any maths problem is as easy as Pi, if you use sufficiently advanced techniques. However, if the problem cannot be differentiated meaningfully, can it be integral?
If they rewrote XRoach as a secure Java applet, then the cockroach would climb around the screen and hide behind windows until you squash it by clicking on it. Hmmm. Actually, that's not a bad idea - can someone on the Slash code development team add this?
I think I met a regular user once. I'm not entirely sure, though, as they seemed rather odd.
When there's money left over, the budget is reduced the following fiscal year by the excess amount the prior fiscal year. That's the way the UK Government funds things, which is why there is a desperate scramble at the end of the year to buy useless, expensive items to flush the rest of the budget. Take a look at the spending pattern of any of the UK county councils.
Actually, it worked pretty well. I had X and OpenLook running and was able to run a Netrek server with 19 robot players and myself on a regular Netrek client. That's not bad going, given that Netrek was not the most elegant of programs. (Hell, olvwm was not exactly a masterpiece of coding. I'm amazed to this day that I could get a full GUI environment that showed no obvious lag on such a system.)
Probably the most surreal experience, though, was running Windows under dosemu under Linux.
Unfortunately, this is unlikely to happen. Most don't have the money or the inclination to go against an 800 lb. chair-wielding gorilla, and I can't blame them.
Part of the problem is that the Open Source community has opted against unionizing. This matters. It has left the community as a whole extremely vulnerable and has left it with no meaningful voice. We cannot collectively act against bullying and intimidation if there is no mechanism for collective action of this kind. The price we are paying for our individual freedom is our freedom as a community. Our ability to act collectively in programming is precisely why Open Source is as sophisticated, rich and diverse as it is. Those who act as self-isolated individualists are generally the ones who get scorned and ignored, their products soon shown to be inferior.
Yet we somehow persist in believing, despite the harassment, despite the intimidation, despite every scrap of evidence that we have to date, that we can simply wish away the abuse of the megacorps.
The letter, as others have noted, is so juvenile as to not be worth discussing. However, the concept of unified action is at least going in the right direction. Mere lobbying won't work. Have you ever seen anyone successfully lobby Microsoft? The European Union only succeeded by threatening to bleed Microsoft dry if it didn't comply, and even then Microsoft did everything in its power to obey the letter of the law only as little as humanly possible without being dragged into court by its collective toenails.
No, merely asking the Evil Empire to play nice isn't going to achieve a damn thing, no matter how many names are put to what piece of paper. If you want Microsoft to cut the crap, then you need to have the strength to back up your request. Right now, we have... exactly nothing. There will be no chorus of heavenly angels throwing hellfire and brimstone at Redmond, there will be no Congressional investigations, you won't even get IBM to contribute to your tombstone.
Yes, opposition to this abuse is essential, but show ME the code. Show me how the community is to achieve this, when it took three companies two decades to bring Microsoft to account over DR-DOS. How do we achieve what the DOJ were unable to do? When we have refused ourselves the right to be heard?
The idea behind assisted ramjets is that you only need the air to be travelling at a high speed, not the vehicle. A standard ramjet will operate at airspeeds of 400 MPH, well within the capabilities of a simple propeller system to provide. A hydrogen-fuelled ramjet will max out at mach 7, which is about the speed a scramjet should ignite. Just stage 'em up. By the time the scramjet finishes, you'll be close to escape velocity and will barely need the rocket stage at all.
...the new laser pointers blew so many wholes in the whiteboard that the diagrams can no longer be read.
Teletext was also a cyclic system. Pages were broadcast in sequence, until the one you wanted appeared. This is OK when there are a few pages - ie: when latency vastly exceeds the cycle time - but it doesn't scale well. Multicast delivery works by delivering content that is being requested to those who have requested it (and nobody else). If done right, it is not based on push/pull concepts, but is based around pooling common requests and pooling common delivery. If you want analogies, then it's efficient mass transit, not television.
And you'd better tell Nokia that their content delivery system sucks, because they clearly don't appear to know it all by themselves.
USENET issued "death sentences" (ie: disconnects) to ISPs that tolerated spammers. Shortlay after, ISPs stopped popularizing USENET. USENET is under-utilized for the simple reason that ISPs want the spammers' money more than they want to provide useful services. I could list a dozen protocols and services from "the old days" that would be a major improvement over current methods that are being mis-applied. And that is the key. There is nothing wrong with current methods, if they are used in the manner they are suited for, for the types of problems they work best with. You do not use a sledgehammer as a substitute crowbar, you do not use a Swiss Army knife (as multi-function as it is) as a substitute pick-axe. So why use PtP for a role it was never intended nor designed for?
Modern IP multicast - ie: ignoring DVMRP and MOSPF - isn't too bad when it comes to distribution. Pruning and grafting of branches is now more-or-less solid ground. IGMPv3/MLPv2 support authentication extensions, source-specific multicast and other cool stuff. IPv6 multicast is intended to be interoperable with Infiniband multicast (though as the number of users of either is extremely limited, this one is of limited value for right now). As a raw transport, it's not shabby. Now, to get anything useful in the way of information systems done, you need to layer a reliable multicast transport on top of that. SRM and NORM are the big two players in this arena, with Open Source implementations of both. FLUTE was a potential big player, for file distribution, but there is a patent on multicast file sharing which has shut down most FLUTE implementations.
(Possibly one reason P2P software doesn't include a multicast option is precisely because of that patent. It's bad enough being vilified by the **IA lawyertroids, but being sued for patent infringement as well would likely cause some serious problems. It's not through a lack of means. A lack of programmers familiar with multicast might also be a problem.)
Multicast information delivery systems have been tried? When? Where? Show me a single ISP that has delivered multicast to residential customers and a single non-trivial example of reliable multicast for information delivery. I consider the mbone utils of VIC and VAT to be trivial examples. Show me something real. A distributed database engine, a replication server that pushes a filesystem to all mirror sites simultaneously, an MPI implementation that uses reliable multicast for collective operations, a multicast SMTP server for sending mail to multiple destinations in a single transfer. If you can't show me the apps and you can't show me real-world residential users, then you can't possibly claim multicast information systems have ever been tried.
(Oh, and I don't consider multicast mosaic to be a meaningful example, either. It was never widely distributed, by the time anyone knew it even existed the world had switched to other browsers, and those who developed it did not exactly go out of their way to tell anyone it was there or what it did. Even if they had, the big thing of the time was Netscape - Mosaic was essentially dead. Why would anyone use a browser that was inferior in virtually every respect, simply to use a distribution model that their ISP would have blocked anyway?)
Hmmm. I thought the distinction between client-side and server-side had been declared a violation of the RPC convention.