o Boot off Win98 floppy. o Answer "yes" when it asks you if you really want to re-install over the existing system o Sit there for 20 minutes while it installs. o Click on the "connect to internet" icon when install is done to reconfigure for your ISP o Go to Start->Windows Update to get any patches from the web-based interface.
No fdisking. No formatting. yea, a few reboots, but the system does them itself if you aren't there.
BTW, just for comparison, I'd like to see the equivalent Red Hat install path INCLUDING PATCHING THE KERNEL TO RUN USB:-)
Did you actually READ the other two articles he was referring to? He did exactly what you suggested, installed and used RedHat.
And as to your points about what the "average" user wants -- what you "believe" is probably not what is (hey, I don't mean this as an insult, it's true for most people). It's a good point -- there are 20 MILLION aol users. More than any Linux users I can count. For some people, it is the SOLE REASON THEY HAVE PURCHASED A COMPUTER. Scary, but true. I wouln't be so dismissive.
Interesting, I had a different reading of the article. I think he was saying that if a community of users is telling you that Microsoft (or Apple, or Sun, etc), is better because of x, maybe you should think a little bit about that and see if there is any truth to it -- NOT just write them off as a bunch of ignorant neanderthals. They might be telling you something VERY important, or giving you a clue as to the next important feature that Linux should have.
And before you cut me off at the legs, think about how not too long ago (like a year) writers like this would complain about how hard it was to install Linux, and that Windows installation was easier. Now the reviews are getting more positive in that respect -- we have simpler installers and pre-installed computers.
First, I doubt too many libraries subscribe to Hustler. Some may subscribe to Playboy, though -- it is still considered a rather prestigious publication for fiction writing. [don't laugh here, you KNOW what I mean...:-)]
But to address your main point, there is a HUGE difference between relying on a human to make those decisions and blindly turning over the reigns to a computer program written by a bunch of people you don't know.
As to political motivations, well, as I said before, librarians are human. It takes an awfully strong person to stand up to the person who writes your paycheck. But I have found that most librarians are highly ethical people with a strong anti-censorship stance. For instance, check out the American Library Association's code of ethics:
http://www.ala.org/alaorg/oif/ethics.html
I think that they understand the difference between letting impressionable children browse through hard-core porn and letting more mature minds have unfettered access to the information that is their birthright.
You are correct, the rules have changed for this, in the last quarter or two (maybe the end of 1999). It's much harder if not impossible for companies to account for mergers as a pooling rather than purchase.
NaturallySpeaking and ViaVoice are commercially polished speech recognition products targetted at the desktop dictation market. They are also speaker-dependent.
Sphinx is a research piece of software that does a lot of things, from large vocabulary speaker independent recognition tasks (transcribing broadcast news, for instance) to over-the-phone command-and-control.
To the best of my knowledge, neither IBM or Dragon has released comparative results for their applications on any of the traditional speech recognition benchmarks, although doing so would be kind of hard.
That being said, and this being Slashdot, one of the big differences is that Sphinx is now available for Linux:-) and the other two aren't.
It's unclear what training data, if any, is included with Sphinx-2. You need two types of training to run a speech recognizer: acoustic training, which tells the system the properties of the microphone, room, and language and/or dialect of speaker, and language model training, which tells the system what words are likely to be recognized.
I've posted a question on SourceForge about what sort of data comes with this system, but without either data or the ability to re-train the system, the usefulness of the recognizer will be curtailed. If CMU has suppplied English microphone-bandwidth acoustics, forget about german over-the-phone recognition.
As to patents, well, I wouldn't worry too much about that. The speech community has been openly publishing most of its results throught the DARPA programs for years. The body of prior art here is pretty high, and anyone claiming a patent would have an uphill battle. Also, Sphinx-2 is NOT CMU's latest and greatest, so that would work in favor of the open-source community.
The President of a company is usually responsible for the day-to-day running of a company -- for instance, are we selling enough product, do we have the right people in the right places to produce it, etc, etc.
The CEO is responsible for the LONG TERM management of a company -- issues like: does the company as a whole have enough funding? What is the appropriate positioning in the marketplace? Can we forge alliances with other companies, or buy them outright? What markets should we be entering or leaving?
In a way, the CEO is the actual head of the company.
The board is another story. The board functions as an overseer. Typically, the board does NOT set policy, make rules or even high-level decisions. This, however, varies widely from company to company, but in general, they can't -- the typical company has board meetings once or twice a quarter at most. Their job is to act as advisor to the CEO (who almost almost ALWAYS has a seat on the board, and is usually chairman), and as a brake on them. They are not employees of the company -- frequently they ARE of other companies, or are executives of other companies, and they are also the elected representatives of the shareholders. Usually a fraction, or all of the board members are re-elected each year. As a last resort, they usually have the ability to remove the CEO, as they did at Compaq last year. oh, they one area they usually do rule over are issues like executive compensation.
If it had gone to court, and Etoy WON, it probably would have been a precident-setting decision. Now, of course, we will probably have to go through this AGAIN with the next bone-head that tries to sue over similar names.
The windows popping out of the Hancock tower were NOT due to the movement of the building, if I remember correctly. They were caused by a MUCH more subtle cause -- a design flaw in the double-paned construction of the windows.
Check out the following link:
http://www.sgh.com/hancock.htm
...for a report on the failure. A summary of this appeared in the Boston Globe a few years back. I would have thought that the IgNobel people (at MIT no less) would have known abou this.
I believe it is true, however, that the sway dampener in the Hancock was put in after its construction -- but I could be wrong, and can't find verification on the web.
Wow, you mean Hollywood made a movie out of a book and ACTUALLY KEPT THE TITLE THE SAME??? I'm thuderstruck! The last time they did that was Gone With the Wind, right?:_)
This "odd lot" stuff is a myth. There is no additional cost to buying/selling 100 shares, although commissions might kill you. This has been true for like ~20 years now, since deregulation of the securities industry.
The issue of small investors, however, is very real. If you want to invest $2,000, and the share price is $3,000, you can't even get in the game. If the share price is $1,200, you can only buy *1* share, and you have $800 you can't do anything with. I've made these numbers up, but when you're talking share prices in the > $200 range, the effect is an issue. Smaller share prices let you fit the money you have to invest to a number of shares with greater ease.
And for an extreme example, check out Berkshire Hathaway, which has never split its stock. The ticker is BRK.A, and it's trading at $54,000 a share today. They issued special shares, BRK.B, to deal with this issue, which trade at exactly 1/30th of the price.
Ok, the one thing the Ebert article has done for me is make me want to check out a digital showing myself. Are there any cinemas in the Boston area showing TS2 or Bicentennial Man in digital? Are there any web listings of cinemas that have digital facilities?
Re: SQL -- it can be any SQL server, really. However, I will add that we are somewhat in bed with Microsoft on the visualization end, simply because IE5 does XML quite well (note to Mozilla people: get with the program).
Re: Open source. Unfortunately, not up to me. Much of the technology is "open source" in the sense that papers have been published about it (not what you were looking for, I know), but we've already licensed some of the core technology to another company, and being a phone company (GTE) we consider the speech rec somewhat of a competitive advantage (wipe those Echelon thoughts out of your mind! We use it for call center and directory assistance automation! Sheesh:-)
As I posted probably about 6 months ago in a thread about speech recognition, there are some significant issues with open-sourcing beyond the recognizer code. The learning processes behind the recognition are based on a considerable amount of data for which licensing is an issue, such as CNN broadcasts. In fact, we use over 100 hours of broadcast news audio to train the system, and several million words of text for the language model. This comes to us through the Linguistic Data Consortium at the University of Pennsylvania (http://www.ldc.upenn.edu). This is an academic group set up to maintain these common train-and-test databases for researchers, and there's a fairly sizeable fee to join. They handle the intellectual property issues with the training data.
And, unfortunately, without the training data, it's kind of hard to use the system. At least, if you want to use it on something it's not already trained on (in our case: north american broadcast news).
Unfortunately, one of the downsides is that we don't have a whizzy on-line demo for people to play with. I suspect that will change in a year, but until now you'll have to make-do with the screen shots.
HOWEVER, I do want to add that the system does run on standard, ordinary PC hardware. The indexer currently runs in 3x realtime (so a half-hour wavefile takes 1.5 hours to index) on a P3-500 running RedHat 6.0 with 512mb of memory. It deposits its product in an SQL-Server database on an equivalent PC running NT (no jeers, please). So the analysis and querying/browsing are decoupled.
We plan on having this down to realtime this year, both through algorithmic improvements and some additional hardware.
Yes, there are quite a few places that do that (CMU has a system called "InforMedia" that does), but there are a few problems:
a) Close-captioning is not an exact representation of what was said. Quite often it is a paraphrase (but speech recognition is errorful, of course);
b) There are many many many useful sources of audio that don't have closed captioning. A meeting, for instance. Or a foreign broadcast (closed-captioning is much more prevalent in the US than in other nations).
It may be rather crass, but for the last few years when my mom has asked me what I wanted, I hand her a book list! (Since she's an ex-librarian, books have always been a prominent gift, but I made it explicit and said that that's more or less all that I want).
I always have a list ready, since I'm constantly jotting down notes about books I want.
This year, I did the logical thing: I transformed this into an Amazon.com wish list. That way I can "hand" her the list electronically, and she can click-to-buy.
o Xfree86. It's just a name. You can run Xfree86 on anything if you *compile* it on it...
o "...written for the VM which hides the actual hardware from the programmer"... Not really a true statement: the virtual machine that it presents is a REPRESENTATION of the ACTUAL underlying hardware, i.e. you program it EXACTLY the same way you program the underlying hardware.
While mainframes may not have "conventional" (i.e. one chip) CPUs, architecturally they are virtually identical, as are most commercial machines (e.g. memory, registers, PSW, etc)
Slashdot Mirror
Hmm... Here's how you install Win98:
:-)
o Boot off Win98 floppy.
o Answer "yes" when it asks you if you really want to re-install over the existing system
o Sit there for 20 minutes while it installs.
o Click on the "connect to internet" icon when install is done to reconfigure for your ISP
o Go to Start->Windows Update to get any patches from the web-based interface.
No fdisking. No formatting. yea, a few reboots, but the system does them itself if you aren't there.
BTW, just for comparison, I'd like to see the equivalent Red Hat install path INCLUDING PATCHING THE KERNEL TO RUN USB
Did you actually READ the other two articles he was referring to? He did exactly what you suggested, installed and used RedHat.
And as to your points about what the "average" user wants -- what you "believe" is probably not what is (hey, I don't mean this as an insult, it's true for most people). It's a good point -- there are 20 MILLION aol users. More than any Linux users I can count. For some people, it is the SOLE REASON THEY HAVE PURCHASED A COMPUTER. Scary, but true. I wouln't be so dismissive.
Interesting, I had a different reading of the article. I think he was saying that if a community of users is telling you that Microsoft (or Apple, or Sun, etc), is better because of x, maybe you should think a little bit about that and see if there is any truth to it -- NOT just write them off as a bunch of ignorant neanderthals. They might be telling you something VERY important, or giving you a clue as to the next important feature that Linux should have.
And before you cut me off at the legs, think about how not too long ago (like a year) writers like this would complain about how hard it was to install Linux, and that Windows installation was easier. Now the reviews are getting more positive in that respect -- we have simpler installers and pre-installed computers.
Oh come on, those aren't even the same things.
:-)]
First, I doubt too many libraries subscribe to Hustler. Some may subscribe to Playboy, though -- it is still considered a rather prestigious publication for fiction writing. [don't laugh here, you KNOW what I mean...
But to address your main point, there is a HUGE difference between relying on a human to make those decisions and blindly turning over the reigns to a computer program written by a bunch of people you don't know.
As to political motivations, well, as I said before, librarians are human. It takes an awfully strong person to stand up to the person who writes your paycheck. But I have found that most librarians are highly ethical people with a strong anti-censorship stance. For instance, check out the American Library Association's code of ethics:
http://www.ala.org/alaorg/oif/ethics.html
I think that they understand the difference between letting impressionable children browse through hard-core porn and letting more mature minds have unfettered access to the information that is their birthright.
You are correct, the rules have changed for this, in the last quarter or two (maybe the end of 1999). It's much harder if not impossible for companies to account for mergers as a pooling rather than purchase.
Depends on what you mean by "compares" :-)
:-) and the other two aren't.
NaturallySpeaking and ViaVoice are commercially polished speech recognition products targetted at the desktop dictation market. They are also speaker-dependent.
Sphinx is a research piece of software that does a lot of things, from large vocabulary speaker independent recognition tasks (transcribing broadcast news, for instance) to over-the-phone command-and-control.
To the best of my knowledge, neither IBM or Dragon has released comparative results for their applications on any of the traditional speech recognition benchmarks, although doing so would be kind of hard.
That being said, and this being Slashdot, one of the big differences is that Sphinx is now available for Linux
Two notes --
It's unclear what training data, if any, is included with Sphinx-2. You need two types of training to run a speech recognizer: acoustic training, which tells the system the properties of the microphone, room, and language and/or dialect of speaker, and language model training, which tells the system what words are likely to be recognized.
I've posted a question on SourceForge about what sort of data comes with this system, but without either data or the ability to re-train the system, the usefulness of the recognizer will be curtailed. If CMU has suppplied English microphone-bandwidth acoustics, forget about german over-the-phone recognition.
As to patents, well, I wouldn't worry too much about that. The speech community has been openly publishing most of its results throught the DARPA programs for years. The body of prior art here is pretty high, and anyone claiming a patent would have an uphill battle. Also, Sphinx-2 is NOT CMU's latest and greatest, so that would work in favor of the open-source community.
It seems that you managed to enjoy the game without grasping the entire point :-) :-)
I think that these interviews are one of the *best* features of Slashdot.
The President of a company is usually responsible for the day-to-day running of a company -- for instance, are we selling enough product, do we have the right people in the right places to produce it, etc, etc.
The CEO is responsible for the LONG TERM management of a company -- issues like: does the company as a whole have enough funding? What is the appropriate positioning in the marketplace? Can we forge alliances with other companies, or buy them outright? What markets should we be entering or leaving?
In a way, the CEO is the actual head of the company.
The board is another story. The board functions as an overseer. Typically, the board does NOT set policy, make rules or even high-level decisions. This, however, varies widely from company to company, but in general, they can't -- the typical company has board meetings once or twice a quarter at most. Their job is to act as advisor to the CEO (who almost almost ALWAYS has a seat on the board, and is usually chairman), and as a brake on them. They are not employees of the company -- frequently they ARE of other companies, or are executives of other companies, and they are also the elected representatives of the shareholders. Usually a fraction, or all of the board members are re-elected each year. As a last resort, they usually have the ability to remove the CEO, as they did at Compaq last year. oh, they one area they usually do rule over are issues like executive compensation.
...but not for the obvious reason.
If it had gone to court, and Etoy WON, it probably would have been a precident-setting decision. Now, of course, we will probably have to go through this AGAIN with the next bone-head that tries to sue over similar names.
The windows popping out of the Hancock tower were NOT due to the movement of the building, if I remember correctly. They were caused by a MUCH more subtle cause -- a design flaw in the double-paned construction of the windows.
Check out the following link:
http://www.sgh.com/hancock.htm
...for a report on the failure. A summary of this appeared in the Boston Globe a few years back. I would have thought that the IgNobel people (at MIT no less) would have known abou this.
I believe it is true, however, that the sway dampener in the Hancock was put in after its construction -- but I could be wrong, and can't find verification on the web.
Wow, you mean Hollywood made a movie out of a book and ACTUALLY KEPT THE TITLE THE SAME??? I'm thuderstruck! The last time they did that was Gone With the Wind, right? :_)
Which Asimov short story was this based on?
This "odd lot" stuff is a myth. There is no additional cost to buying/selling 100 shares, although commissions might kill you. This has been true for like ~20 years now, since deregulation of the securities industry.
The issue of small investors, however, is very real. If you want to invest $2,000, and the share price is $3,000, you can't even get in the game. If the share price is $1,200, you can only buy *1* share, and you have $800 you can't do anything with. I've made these numbers up, but when you're talking share prices in the > $200 range, the effect is an issue. Smaller share prices let you fit the money you have to invest to a number of shares with greater ease.
And for an extreme example, check out Berkshire Hathaway, which has never split its stock. The ticker is BRK.A, and it's trading at $54,000 a share today. They issued special shares, BRK.B, to deal with this issue, which trade at exactly 1/30th of the price.
Ok, the one thing the Ebert article has done for me is make me want to check out a digital showing myself. Are there any cinemas in the Boston area showing TS2 or Bicentennial Man in digital? Are there any web listings of cinemas that have digital facilities?
Ohoh, I've stirred up a firestorm here :-)
:-)
Re: SQL -- it can be any SQL server, really. However, I will add that we are somewhat in bed with Microsoft on the visualization end, simply because IE5 does XML quite well (note to Mozilla people: get with the program).
Re: Open source. Unfortunately, not up to me. Much of the technology is "open source" in the sense that papers have been published about it (not what you were looking for, I know), but we've already licensed some of the core technology to another company, and being a phone company (GTE) we consider the speech rec somewhat of a competitive advantage (wipe those Echelon thoughts out of your mind! We use it for call center and directory assistance automation! Sheesh
As I posted probably about 6 months ago in a thread about speech recognition, there are some significant issues with open-sourcing beyond the recognizer code. The learning processes behind the recognition are based on a considerable amount of data for which licensing is an issue, such as CNN broadcasts. In fact, we use over 100 hours of broadcast news audio to train the system, and several million words of text for the language model. This comes to us through the Linguistic Data Consortium at the University of Pennsylvania (http://www.ldc.upenn.edu). This is an academic group set up to maintain these common train-and-test databases for researchers, and there's a fairly sizeable fee to join. They handle the intellectual property issues with the training data.
And, unfortunately, without the training data, it's kind of hard to use the system. At least, if you want to use it on something it's not already trained on (in our case: north american broadcast news).
Unfortunately, one of the downsides is that we don't have a whizzy on-line demo for people to play with. I suspect that will change in a year, but until now you'll have to make-do with the screen shots.
HOWEVER, I do want to add that the system does run on standard, ordinary PC hardware. The indexer currently runs in 3x realtime (so a half-hour wavefile takes 1.5 hours to index) on a P3-500 running RedHat 6.0 with 512mb of memory. It deposits its product in an SQL-Server database on an equivalent PC running NT (no jeers, please). So the analysis and querying/browsing are decoupled.
We plan on having this down to realtime this year, both through algorithmic improvements and some additional hardware.
Yes, there are quite a few places that do that (CMU has a system called "InforMedia" that does), but there are a few problems:
a) Close-captioning is not an exact representation of what was said. Quite often it is a paraphrase (but speech recognition is errorful, of course);
b) There are many many many useful sources of audio that don't have closed captioning. A meeting, for instance. Or a foreign broadcast (closed-captioning is much more prevalent in the US than in other nations).
...which would be...?
Might as well use this as a chance to plug my project:
e arch/extraction/roughn_ready/index.html
http://www.gte.com/AboutGTE/gto/bbnt/speech/res
...which not only tells you what words were said, but who said them, and what topics were being talked about...
This post does *NOT* substantiate the rumor. And the article is simply a link to the SAME article as two weeks ago.
C'mon, it's not that hard to look for rehashes. I can't believe that anyone has THAT much of a short attention span!
Didn't this get posted about 2 weeks ago?
It may be rather crass, but for the last few years when my mom has asked me what I wanted, I hand her a book list! (Since she's an ex-librarian, books have always been a prominent gift, but I made it explicit and said that that's more or less all that I want).
I always have a list ready, since I'm constantly jotting down notes about books I want.
This year, I did the logical thing: I transformed this into an Amazon.com wish list. That way I can "hand" her the list electronically, and she can click-to-buy.
Um, I hate to point this out, but:
o Xfree86. It's just a name. You can run Xfree86 on anything if you *compile* it on it...
o "...written for the VM which hides the actual hardware from the programmer"... Not really a true statement: the virtual machine that it presents is a REPRESENTATION of the ACTUAL underlying hardware, i.e. you program it EXACTLY the same way you program the underlying hardware.
While mainframes may not have "conventional" (i.e. one chip) CPUs, architecturally they are virtually identical, as are most commercial machines (e.g. memory, registers, PSW, etc)