So, you are saying that USENET has changed from an informal discussion group to a searchable perpetual repository of technical support Q&As, plus a repository of background information on people who were foolish enough in the 1980s to post under their own names. I agree. The part I don't understand how you think that constitutes "saving" USENET. USENET didn't use to be much of an on-line community compared to some of the others, but it was a community. Once it became archival, anonymous, and searchable, that went away. Who, after all, wants their every word recorded and replayed into perpetuity?
Re:services like Deja/Google killed USENET
on
How Google Saved USENET
·
· Score: 1, Offtopic
Ah, I see, another idiotic, anonymous Slashdot moderator got to my post.
I'm tired of moderators moderating posts down because they disagree with the content, so I'm reposting this. Go ahead, moderate it down again, I have lots of points. But I suggest rather than having some gut reaction to this, you think about it and, if you disagree with it, post some reasoned response.
I have been using USENET for 20 years, so I am affected by this, and I have seen USENET slowly fall apart. USENET was always a bit rough and had a lot of noise, but people did get to know each other personally and professionally. Today, USENET is nearly completely useless for any kind of social functions, and the huge expansion of people posting, anonymous/pseudonymous postings, and the need to post anonymously because of searchable archives is largely responsible. There is no forum like USENET was 20 years ago anymore.
USENET used to be an informal discussion forum, like something where you might talk with others like you would around the water cooler. Google, AOL, and similar services have greatly expanded the user base for USENET, which means that it isn't much of a community anymore. And by archiving and republishing in perpetuity, thinking people have to watch carefully what they say, or they just post anonymously (or don't participate at all anymore).
This was probably an unavoidable turn of events. Nevertheless, whether it is Google or some other company, I consider it wrong for them to republish this stuff, in particular as part of a commercial venture. It's the equivalent of digging out old security surveillance tapes and broadcasting them for the amusement of the masses. It's wrong, and the fact that people find some sort of voyeuristic delight in it doesn't change that. The backup tapes that Google used should have been destroyed.
We had this in 1981: expiration headers. The expectation was that articles would expire within a few weeks, and some articles had explicit expiration dates. But Deja/Google just decided that those didn't apply to them and if you wanted your article to expire, you really had to add their new header to your messages. What guarantee is there that in another 20 years, some other company isn't going to decide that "x-no-archive" really doesn't apply to them?
The appearance of Deja/Google archives killed USENET because it has shown that there are no guarantees: only a fool would now engage in any kind of controversial discussion on USENET under their own name.
Erm... things you publish aren't automatically in the public domain. Besides, I wasn't making a legal point, I was making a point about the social impact of archiving and republishing informal discussion groups into perpetuity.
The information is preserved for posterity, not for making money or other commercial exploits.
Oh? When did Google become a non-profit foundation for the preservation of historical electronic information? And what does historical preservation have to do with publishing a searchable database to the web?
I can't really believe you think we'd be better off destroying information instead of preserving it!
Well, we can preserve a lot more information. For example, we can install video recorders all around your house. I'm sure people around the world would find it amusing, and 200 years from now, historians will love those sorts of documents.
Well, I didn't criticize Apple over their strategy--maybe there is nothing they can do. I'm just saying that if they can't figure out how to offer some really nifty sub-$1000 machines (comparable to what you get in the PC world) and a wider range of laptops, I think they will not achieve much more market share than they have. Maybe Apple could partner with Sony or IBM to broaden their product selection?
Google/Deja has killed USENET, not saved it
on
How Google Saved USENET
·
· Score: 2, Insightful
USENET used to be an informal discussion forum, like something where you might talk with others like you would around the water cooler. Google, AOL, and similar services have greatly expanded the user base for USENET, which means that it isn't much of a community anymore. And by archiving and republishing in perpetuity, thinking people have to watch carefully what they say, or they just post anonymously (or don't participate at all anymore).
This was probably an unavoidable turn of events. Nevertheless, whether it is Google or some other company, I consider it wrong for them to republish this stuff, in particular as part of a commercial venture. It's the equivalent of digging out old security surveillance tapes and broadcasting them for the amusement of the masses. It's wrong, and the fact that people find some sort of voyeuristic delight in it doesn't change that. The backup tapes that Google used should have been destroyed.
Well, I was posting from a $1000 low-end laptop with a 14.4" screen, a 1GHz AMD processor, 256M of RAM, and a DVD drive, bought about 6 months ago on sale. Even allowing for a smaller screen, the closest Apple configuration at the time was about $1800. Even six months later, now that we have 14" iBooks, the closest Apple offering has still costs $1800. Low-end PCs aren't "Yugos", they are pretty comfortable and affordable economy and compact cars. Apple makes nice hardware, but you do pay a premium.
Well, take a look here:
Compaq Presario, 1.5GHz P4, 512M, 40Gb, CDRW, Windows XP for $1000. Or here: T4150 Minitower, 1.5GHz P4, 256M, 40Gb, CDRW, Windows XP. $749
And for an LCD monitor, here: Envision 15" for $349. This is CompUSA, so these aren't even rock-bottom prices; you can get these brands even cheaper.
As for DVD writer, I was comparing entry-level machines. If you want to add a DVD-writer, the cost differential is roughly the same in the PC and Mac world.
I can't buy the "$799 iMac" at CompUSA (I think it's a special educational deal) and it's a pretty limited machine even compared to $800 PCs. At 128M, it isn't really good for running Apple's now-standard OSX. That price is basically a closeout price for a discontinued model.
And, no, the TiBook is not an ultralight laptop by any stretch of the imagination. Maybe it's "acceptable" to you, but that doesn't make Apple's selection any less limited. In the PC world, you can get laptops weighing around 2 pounds (1kg), laptops with 10h battery life, and laptops with a wide variety of pointing devices. They sell because lots of people want those features.
I don't think the price difference has evaporated. Let's look at the "entry-level" iMac, a $1300 machine. For around $1300, you get very nice laptops with 14"-15" screens. Or, for less than $1000, you can get a fast desktop PC with a 15" LCD screen. Furthermore, at the very low end of the market, CRTs are still the way to go because they are cheaper, and Apple is cutting themselves off from that market. The iMac is a neat machine, but it isn't cheap and it isn't low-end. For practical purposes, Apple doesn't have a low-end machine, and that's a problem for them.
Apple reveals a bigger iBook, a not-so-low-end desktop machine with a built-in LCD screen at a not-so-entry-level price, and some software (did I miss anything?). That's nice and a logical evolution, but it doesn't extend Apple's product range or widen Apple's appeal. A few people may buy the iMac for its funky design, and others will likely not touch it. Some people will be convinced by OSX's (now standard on all models) better software quality compared to Windows, but most people probably don't know or don't care.
Notably absent was an Apple PDA, where Apple could have demonstrated fundamentally new technology (OSX, in contrast, has been is a great engineering effort, but is an evolutionary development from NeXTStep and MacOS).
The markets apparently weren't "blown away" either, as Apple stock is down for the day as of 3pm EST. I think Apple overhyped this one.
Apple is, and remains, a high-end, high-quality vendor for a niche market with a particular taste. In comparison, no matter how nice a car an RX-7 may be, not everybody will want to drive one. Apple will not take over the world, at least not with its current product range. Most sorely missing from their product range is a smaller, cheap desktop (a sub-$1000 cube, maybe) and an ultralight laptop.
Come on--you just moderate things down because you don't like it. If you disagree, provide some arguments. Like, where is Apple's 3 pound laptop? Where is Apple's stereo component-sized computer? Where is Apple's sleek black machine? Where is Apple's wooden box? Where is Apple's 8 processor machine? Apple does have a very limited product range and that limits their appeal, no matter how well Apple's techno-plexiglass geek chic may fit into your decor.
The performance DOES actually represent a real-world system, as long as you have reasonable expectations of the real world system. I.e. do not train your system on a broadcast news corpus and then feed it Jerry Springer:-).
Unfortunately, no matter how carefully you design your database in 2001, you can't predict how people speak or how they are recorded in 2003. And current speech recognition systems do not adapt well enough automatically.
As to whether the task is worth doing -- well, that's for neither of us to answer, but the market + Virage, et al -- right?:-)
The market has spoken loud and clear so far, which is why this sort of thing is not a major line of business for any of the players that started it.
It's not about software, it's just that those kinds of designs only fit into 5% of the country's living rooms. Seriously, Apple's very limited product design range (one big G4 clunker, an oddball looking iMac, and two laptops with one choice of pointing device) really limits the adoption of their systems. Intel system, in contrast, come in hundreds of different shapes and sizes, from stylish to practical to industrial. Add to that that $1200 is rather high for an entry-level system, and it's not difficult to see why Apple has only a niche market and won't grow beyond it.
In addition, teaching involves research, and if university research ends up being proprietary, there are horrendous conflicts of interest that greatly interfere with the academic functions of the university. In order to be able to carry out the business of teaching, university research must be in the public domain, as it has traditionally been.
You can find the patent here. I find both the patent and the claims that Torng invented OOE kind of dubious. The idea of out-of-order execution goes back to at least the 1960s and has cropped up in several different areas of computer science. It may not have been applied much to microprocessors because until the late 1980s, it wasn't all that practical, and EEs are often unfamiliar with ideas in CS anyway. Overall, at least at first glance, this patent looks to me like so many recent patents where people patent ideas that have been around for a long time but simply hadn't been used in practice for practical reasons. Note that Torng has no prior patents in processor design, so it doesn't look like he had been inventing in the field much before then.
I certainly wouldn't assume outright that HP wilfully infringed this patent--HP may well have legitimate reasons to believe that this idea is in the public domain by now. The courts will have to work that out.
(BTW, the guy's name is "Torng"; shouldn't you know the name of your professor?)
Yes, I'm watching BBC on a US TV. I'm probably behind the times on that.
Quite. The system has been in place in Europe for at least 20 years, probably closer to 30 years. It may actually predate closed captioning in the US, which I believe was deployed in 1980.
Also, the published results of numbers of companies are in the 90% range. These are ajudicated results -- not marketing material.
Yes, but they are results on "standard databases". Unfortunately, those do not represent real-world performance of deployed systems, no matter how carefully the databases are designed.
I agree with you that speech rec may be one technique -- but it's the only one that scales.
Of course, that presumes that the task (indexing broadcast video) is worth doing in the first place, which I am not convinced it is. In almost other video indexing tasks, the video clips come with plenty of metadata. And the only reason that isn't included in broadcast video is because broadcasters have no incentive for doing it (why make copying any easier than it already is).
Funny, last time I watched BBC (a few days ago) there was no closed captioning.
If you are watching European television on a US set, you won't get it. CC is some limited, oddball US hack. Europe uses Teletext, which provides not only closed captioning, but also news, weather, subtitles, and program information.
As far as CC getting keywords right -- have you LOOKED at closed captioning lately?
Sure. Looks fine to me. Speech recognition systems are actually being trained with it.
Speech recognition currently attains 90%+ accuracy on a problem like broadcast news,
I think that's a bit optimistic. In any case, my point was not that people shouldn't use speech recognition, but that it's one of many techniques, and that all of them have been worked on by many groups for years. The real problem with these kinds of systems is that they are less useful than you might think.
Not only are people bothering to write viruses for it, the popular press now refers to Linux as in "programs written for Linux, an alternative to Microsoft's Windows".
Don't fiddle around with special E-mail clients, use an encrypted file system.
If you must, Emacs/XEmacs can be set up to automatically decrypt/encrypt on load/save, and that should work with any of the Emacs/XEmacs mail clients. The packages are crypt.el or jka-compr.
That's utter BS. There are plenty of reasons you may want to encrypt your E-mail archives even if it's transmitted in plain text. Perhaps you keep them on a laptop and worry about it getting stolen. Perhaps you use a secure VPN for getting your corporate mail and now want to secure the on-disk storage.
Furthermore, for any reasonable cryptosystem, having even tons of plaintex and encrypted text available is not sufficient to recover the key.
Guess what--you are completely wrong. European television, of course, has closed captioning (it's transmitted differently, but it's the same effect) for the hearing impaired. I believe several Asian countries do as well. And you can bet that CC gets the keywords right. Furthermore, CC gets every speaker in every scene. Even if CC is somewhat of a a shorthand, that's a lot better than speech recognition, which misses entire scenes.
IBM has a lot of OS hackers, and they have little left to do. AIX is kind of a dead end and IBM management is pushing Linux more and more. OS/2 is dead for practical purposes. They can't just all pack up and go into management. Many of these people think that Linux can't be any good because it is open source and wasn't written by people like them.
So, what do they do? The desparately try to find a justification for their existence and they pitch various projects to their management. "Media OS", "low latency", "very high bandwidth", "digital rights management", and "working with Sony", is what stuck. Among the few choices that they had to justify working on a new proprietary OS, that's probably the best they could do.
I doubt it will come to much. In the best case (for them), the PS/3 may run this thing, but PS/2 already runs a proprietary OS. But Linux will be able to handle all those problems as well or better by the time the system comes out, and you will see Linux in more and more media and consumer devices. That's not to say Linux is perfect, just that what these people seem to be proposing doesn't improve on Linux in the areas where it needs improving.
Please be more careful when making statements about other people's claims. Here is what I wrote:
The way it should be done is that you spend some time ahead of time and work out APIs that you can live with for decades. When you do make significant API changes, you give people the tools to convert their code (deprecation, automatic translation). Saddling the platform and their code with dozens of incompatible variants is not the right approach.
This doesn't mean that nothing ever gets added or that all requirements are anticipated decades in advance. It means that when you add functionality, you try to do it in such a way that you won't have to add another variant of the same thing a couple of years later. And if you do make a mistake, you admit it and clean up the mess, rather than retaining junk for ever and ever.
Now, which APIs have those properties? Many actually: the core UNIX APIs, the C standard library, the UNIX/POSIX shell environment, X11, Motif, and Postscript, to name just a few. Well written C/UNIX code from the early 1980's is still functioning, well-written UNIX code today.
This is in contrast to the DOS/Windows approach, where, over the years, you got lots of different approaches for things like opening files, creating windows, handling events, etc. Code written for Microsoft platforms in the early 1980's looks nothing like code you would write for Microsoft platforms today, even if it may still compile in some cases. Heck, even most code written for Microsoft platforms in the mid-1990s looks nothing like what you would do today, using Microsoft's "new" APIs and "new" paradigms. This is particularly bad given that Microsoft's systems haven't been exactly breaking new ground--Microsoft had no excuse for getting any of this wrong in the first place.
Slashdot Mirror
So, you are saying that USENET has changed from an informal discussion group to a searchable perpetual repository of technical support Q&As, plus a repository of background information on people who were foolish enough in the 1980s to post under their own names. I agree. The part I don't understand how you think that constitutes "saving" USENET. USENET didn't use to be much of an on-line community compared to some of the others, but it was a community. Once it became archival, anonymous, and searchable, that went away. Who, after all, wants their every word recorded and replayed into perpetuity?
Ah, I see, another idiotic, anonymous Slashdot moderator got to my post.
I have been using USENET for 20 years, so I am affected by this, and I have seen USENET slowly fall apart. USENET was always a bit rough and had a lot of noise, but people did get to know each other personally and professionally. Today, USENET is nearly completely useless for any kind of social functions, and the huge expansion of people posting, anonymous/pseudonymous postings, and the need to post anonymously because of searchable archives is largely responsible. There is no forum like USENET was 20 years ago anymore.
The appearance of Deja/Google archives killed USENET because it has shown that there are no guarantees: only a fool would now engage in any kind of controversial discussion on USENET under their own name.
The information is preserved for posterity, not for making money or other commercial exploits.
Oh? When did Google become a non-profit foundation for the preservation of historical electronic information? And what does historical preservation have to do with publishing a searchable database to the web?
I can't really believe you think we'd be better off destroying information instead of preserving it!
Well, we can preserve a lot more information. For example, we can install video recorders all around your house. I'm sure people around the world would find it amusing, and 200 years from now, historians will love those sorts of documents.
Well, I didn't criticize Apple over their strategy--maybe there is nothing they can do. I'm just saying that if they can't figure out how to offer some really nifty sub-$1000 machines (comparable to what you get in the PC world) and a wider range of laptops, I think they will not achieve much more market share than they have. Maybe Apple could partner with Sony or IBM to broaden their product selection?
This was probably an unavoidable turn of events. Nevertheless, whether it is Google or some other company, I consider it wrong for them to republish this stuff, in particular as part of a commercial venture. It's the equivalent of digging out old security surveillance tapes and broadcasting them for the amusement of the masses. It's wrong, and the fact that people find some sort of voyeuristic delight in it doesn't change that. The backup tapes that Google used should have been destroyed.
Well, I was posting from a $1000 low-end laptop with a 14.4" screen, a 1GHz AMD processor, 256M of RAM, and a DVD drive, bought about 6 months ago on sale. Even allowing for a smaller screen, the closest Apple configuration at the time was about $1800. Even six months later, now that we have 14" iBooks, the closest Apple offering has still costs $1800. Low-end PCs aren't "Yugos", they are pretty comfortable and affordable economy and compact cars. Apple makes nice hardware, but you do pay a premium.
As for DVD writer, I was comparing entry-level machines. If you want to add a DVD-writer, the cost differential is roughly the same in the PC and Mac world.
And, no, the TiBook is not an ultralight laptop by any stretch of the imagination. Maybe it's "acceptable" to you, but that doesn't make Apple's selection any less limited. In the PC world, you can get laptops weighing around 2 pounds (1kg), laptops with 10h battery life, and laptops with a wide variety of pointing devices. They sell because lots of people want those features.
I don't think the price difference has evaporated. Let's look at the "entry-level" iMac, a $1300 machine. For around $1300, you get very nice laptops with 14"-15" screens. Or, for less than $1000, you can get a fast desktop PC with a 15" LCD screen. Furthermore, at the very low end of the market, CRTs are still the way to go because they are cheaper, and Apple is cutting themselves off from that market. The iMac is a neat machine, but it isn't cheap and it isn't low-end. For practical purposes, Apple doesn't have a low-end machine, and that's a problem for them.
Notably absent was an Apple PDA, where Apple could have demonstrated fundamentally new technology (OSX, in contrast, has been is a great engineering effort, but is an evolutionary development from NeXTStep and MacOS).
The markets apparently weren't "blown away" either, as Apple stock is down for the day as of 3pm EST. I think Apple overhyped this one.
Apple is, and remains, a high-end, high-quality vendor for a niche market with a particular taste. In comparison, no matter how nice a car an RX-7 may be, not everybody will want to drive one. Apple will not take over the world, at least not with its current product range. Most sorely missing from their product range is a smaller, cheap desktop (a sub-$1000 cube, maybe) and an ultralight laptop.
Come on--you just moderate things down because you don't like it. If you disagree, provide some arguments. Like, where is Apple's 3 pound laptop? Where is Apple's stereo component-sized computer? Where is Apple's sleek black machine? Where is Apple's wooden box? Where is Apple's 8 processor machine? Apple does have a very limited product range and that limits their appeal, no matter how well Apple's techno-plexiglass geek chic may fit into your decor.
Unfortunately, no matter how carefully you design your database in 2001, you can't predict how people speak or how they are recorded in 2003. And current speech recognition systems do not adapt well enough automatically.
As to whether the task is worth doing -- well, that's for neither of us to answer, but the market + Virage, et al -- right? :-)
The market has spoken loud and clear so far, which is why this sort of thing is not a major line of business for any of the players that started it.
It's not about software, it's just that those kinds of designs only fit into 5% of the country's living rooms. Seriously, Apple's very limited product design range (one big G4 clunker, an oddball looking iMac, and two laptops with one choice of pointing device) really limits the adoption of their systems. Intel system, in contrast, come in hundreds of different shapes and sizes, from stylish to practical to industrial. Add to that that $1200 is rather high for an entry-level system, and it's not difficult to see why Apple has only a niche market and won't grow beyond it.
In addition, teaching involves research, and if university research ends up being proprietary, there are horrendous conflicts of interest that greatly interfere with the academic functions of the university. In order to be able to carry out the business of teaching, university research must be in the public domain, as it has traditionally been.
I certainly wouldn't assume outright that HP wilfully infringed this patent--HP may well have legitimate reasons to believe that this idea is in the public domain by now. The courts will have to work that out.
(BTW, the guy's name is "Torng"; shouldn't you know the name of your professor?)
Quite. The system has been in place in Europe for at least 20 years, probably closer to 30 years. It may actually predate closed captioning in the US, which I believe was deployed in 1980.
Also, the published results of numbers of companies are in the 90% range. These are ajudicated results -- not marketing material.
Yes, but they are results on "standard databases". Unfortunately, those do not represent real-world performance of deployed systems, no matter how carefully the databases are designed.
I agree with you that speech rec may be one technique -- but it's the only one that scales.
Of course, that presumes that the task (indexing broadcast video) is worth doing in the first place, which I am not convinced it is. In almost other video indexing tasks, the video clips come with plenty of metadata. And the only reason that isn't included in broadcast video is because broadcasters have no incentive for doing it (why make copying any easier than it already is).
If you are watching European television on a US set, you won't get it. CC is some limited, oddball US hack. Europe uses Teletext, which provides not only closed captioning, but also news, weather, subtitles, and program information.
As far as CC getting keywords right -- have you LOOKED at closed captioning lately?
Sure. Looks fine to me. Speech recognition systems are actually being trained with it.
Speech recognition currently attains 90%+ accuracy on a problem like broadcast news,
I think that's a bit optimistic. In any case, my point was not that people shouldn't use speech recognition, but that it's one of many techniques, and that all of them have been worked on by many groups for years. The real problem with these kinds of systems is that they are less useful than you might think.
Not only are people bothering to write viruses for it, the popular press now refers to Linux as in "programs written for Linux, an alternative to Microsoft's Windows".
If you must, Emacs/XEmacs can be set up to automatically decrypt/encrypt on load/save, and that should work with any of the Emacs/XEmacs mail clients. The packages are crypt.el or jka-compr.
Furthermore, for any reasonable cryptosystem, having even tons of plaintex and encrypted text available is not sufficient to recover the key.
Guess what--you are completely wrong. European television, of course, has closed captioning (it's transmitted differently, but it's the same effect) for the hearing impaired. I believe several Asian countries do as well. And you can bet that CC gets the keywords right. Furthermore, CC gets every speaker in every scene. Even if CC is somewhat of a a shorthand, that's a lot better than speech recognition, which misses entire scenes.
So, what do they do? The desparately try to find a justification for their existence and they pitch various projects to their management. "Media OS", "low latency", "very high bandwidth", "digital rights management", and "working with Sony", is what stuck. Among the few choices that they had to justify working on a new proprietary OS, that's probably the best they could do.
I doubt it will come to much. In the best case (for them), the PS/3 may run this thing, but PS/2 already runs a proprietary OS. But Linux will be able to handle all those problems as well or better by the time the system comes out, and you will see Linux in more and more media and consumer devices. That's not to say Linux is perfect, just that what these people seem to be proposing doesn't improve on Linux in the areas where it needs improving.
This doesn't mean that nothing ever gets added or that all requirements are anticipated decades in advance. It means that when you add functionality, you try to do it in such a way that you won't have to add another variant of the same thing a couple of years later. And if you do make a mistake, you admit it and clean up the mess, rather than retaining junk for ever and ever.
Now, which APIs have those properties? Many actually: the core UNIX APIs, the C standard library, the UNIX/POSIX shell environment, X11, Motif, and Postscript, to name just a few. Well written C/UNIX code from the early 1980's is still functioning, well-written UNIX code today.
This is in contrast to the DOS/Windows approach, where, over the years, you got lots of different approaches for things like opening files, creating windows, handling events, etc. Code written for Microsoft platforms in the early 1980's looks nothing like code you would write for Microsoft platforms today, even if it may still compile in some cases. Heck, even most code written for Microsoft platforms in the mid-1990s looks nothing like what you would do today, using Microsoft's "new" APIs and "new" paradigms. This is particularly bad given that Microsoft's systems haven't been exactly breaking new ground--Microsoft had no excuse for getting any of this wrong in the first place.