Relying on face recognition a bad idea. Certain segments of the population have a condition called "prosopagnosia" in which victims are unable to recognize faces, even familiar ones like their mother's or even their own. A similar condition is described in the famous book "The Man Who Mistook His Wife for a Hat". Here the researcher describes the more general condition of object agnosia which is the inability to recognize any type of object. Presumably those with object agnosia would fail the inkblot password scheme.
Note that prosopagnosia is not a subset of object agnosia; some with one do not suffer the other (which is the cause of much controversy as to their origins, but that's getting off topic).
Just a note to people who still want to have fun hacking away: TiVos run a custom Linux kernel on a PowerPC board. Those lucky enought with a Series 1 TiVo can hack it the kernel to do stuff like providing a bash prompt or run a web server. Those with a Series 2 with Home Media Option (HMO) can write all sorts of applets to their hearts contents; see www.tivo.com/developer to download the API.
To the original poster: Is it really worth it to build your own system if you reside within the TiVo market? Have you considered things like: hardware costs (a fast processor, video capture board, lots of RAM, motherboard, case), software (time to get the kernel + driver working, time to cobble together a UI), and other intangibles (getting a remote to work, fan noise, getting timely scheduling information)?
Kind of off-topic, but not really, no, you cannot deduct time spent on OSS projects. Under the section "Contributions You May Not Deduct" on page A-4, 2002 1040 instructions, you cannot deduct "Value of your time or services".
So if you're looking to itemize deductions this year, give money to Xiph et al.
Disclaimer: IANALTP (I am not a licensed tax preparer)
You realize that what you are suggesting is very naive? There's a whole class of computing at the embedded system layer. I don't know about you, but I don't want the defibrillator keeping me alive suddenly pause while it's garbage collecting some values. Or in terms of "general computing", the software for an air-traffic controllers needs to make millions (if not billions) of calculations on a radar beam to decide the position and velocity, if any, of an incoming plane. Computers are fast, but memory is slow The OS is going to be spending time allocating memory and bounds checking each radar ping; meanwhile planes will be crashing.
Eventually at some level code needs static buffers. Well-designed programs along with proper code validation techniques ensure a minimal number of errors. Java/C#/language-of-the-month can help software engineering, but by no means are they a panacea.
I had one student who was failing every single assignment, but then magically turned in a program which ran flawlessly, was multithreaded, and had network support. Needless to say, a few seconds on Google turned up the original -- and uncited -- source code.
Background info: I am currently taking Digital Video Special Effects, an oh-so-cool senior-level CS class. Our projects require us to use video editing equipment. Others have already discussed the hardware side; I just want to mention some programs freely available for Linux.
First of all, there is Broadcast 2000, a GPL non-linear editor. For your video capturing needs, try dvgrab (assuming that you've got a IEEE-1394 compliant capture card). And as a cheap plug for my own program, I am the author of gvplay, a simple Gnome/GTK video player. I wrote gvplay to help render my special effect (object replacement through tracking and edge detection).
While you make several good points, let me expand on some of them.
(Background: Born & raised in Pittsburgh, now an undergrad at Georgia Tech in Atlanta.)
As I compare the differences between two roughly equal-sized cities, Atlanta and Pittsburgh, I note that while the former is growing leaps and bounds, the latter is shrinking away. I attribute this to several factors:
- Taxes. When I compare the amount I pay in taxes to good ol' Pittsburgh and Pennsylvania, I look with to my friends paying the much smaller Atlanta and Georgia taxes. For example, Pennsylvania imposes a state-wide sales tax and income tax, while Georgia has just an income tax -- and one lower than Pennsylvania. Georgian are mainly taxed by land, but who cares if one is just renting an apartment?
- Education. And by this, I don't mean college-level, for both Atlanta and Pittsburgh are equals. I mean the primary and secondary educations. Having attended the public school system, I must say that I was rarely, if ever, encouraged to learn more about technology. Rather, the teachers were of the old-school (pun attended!) variety. Among other things, they truly believed it to be easier to deny students access to web sites like Rootshell rather than to take measures to tighten security. Security through obsecurity, anyone? I don't know how Atlanta's schools measure up, but I doubt they can be much worse than Pittsburgh's.
- On a related note, priorities. Georgia uses its lottery revenue to fund the HOPE scholarship, which gives students a free ride through college as long as a 3.0 GPA is maintained. Many families in surrounding states move to Georgia whenever their children are about to graduate from high school. Those families would have parents in the 50s or so, which means that they are near the top of their income bracket. And what about Pennsylvania? Their lottery goes towards Senior Citizens. I may be prejudiced by this, but I doubt grandparents are quick to start up new businesses, especially technological-oriented ones.
- Transportation. Marta (Atlanta) kick's the T's (Pittsburgh) butt.
- Nightlife. Although I don't drink, I do know that Atlanta's Buckhead region is a whole lot more exciting than Pittsburgh's South Side.
What's the difference here folks? Is it because it's "just a game"
I would say so. At least to me, I make a distinction between entertainment software and operational software. For the latter category, I prefer to use the best choice available (which usually ends up being open source, but that's a different thread). When I'm working, the software had better work; who cares about stupid add-ons such as the Microsoft paperclip? On the other hand, when I play a computre game, I want all of the bells & whistles -- sound, CD music, textured backgrounds, etc. I'm sorry, but none of the open source games I've seen come even close to that of Railroad Tycoon II.
Why can't they say "well, uh there were 153,893 windows versions bought, and then 53,000 people turned in their coupons for the linux version... So we sold 100,893 windows versions."
Or am I missing something?
I think so. I believe the point that Carmack is trying to make is to send a message to retailers about the Linux market.
Most software houses don't sell their works straight to the public. Instead, they publish them and then sell to the retailers (e.g., CompUSA). If the retailers don't demand Linux-based games, then the publishers don't sell any. Thus, they won't bother writing any Linux games.
Take a look at two suggestions I've seen:
Sell just a PC version, but let the user download the binary for his OS. This logic is severely flawed. Ok, so id knows how many Linux/Mac clients exist. Whee. They then publish their server logs. Whee. Other software companies check out id's page and sees some statistics. How do they know that id did not make up those numbers? In other words, other companies can not verify, for themselves, the Linux market.
Distribute a hybrid CD. For the registration card, allow the user to choose his OS. This is flawed like above. Again, id could publish the values; yet the numbers can not be emperically verified.
Suppose that id continues as plans. Linux users buy Linux versions of Q3; Mac and PC users do likewise. Now, each retailer can verify for themselves the demand for the different OSs. Knowing those values, they can then demand to the software houses what operating systems they (the publishers) should support.
If the demand for Linux is non-existant, then the retailers won't buy any Linux games from the publishers. Even if the publishers want to support Linux, they might choose to not do so if nobody is going to buy Linux.
But if Q3, Linux version sells extremely well, the retailers are sure to notice. They then pass on this demand to the publishers.
And IDC's House points out CE is still a young product. "It hasn't reached its third release, and the third release is the charm for Microsoft operating systems," she says, referring to the success of Windows 3.0 and NT 3.0.
Interestingly enough, the first public release of Win NT was 3.0. If I recall correctly, Win 3.0 beta was written to be uncompatible with a non-Microsoft DOS (i.e., DR-DOS). This was also around the time of them dropping the joint MS-IBM OS/2 project. To me, saying that Microsoft doesn't get anything "right until version 3" is rather misleading. This implies that their third release was successful due to a superior product, not marketing or strong-arm tactics.
He was even able to prove that his solution was optimal
Not quite. The Huffman algorithm has been proven to produce optimal prefix code, which are codes such that no codeword is also a prefix of another codeword. According to my algorithms book, Huffman compression saves 20% to 90%.
This was true in the past, but times have changed. There is a new franchise in California, "Pac Hut", that has been silently buying up Pizza Huts off of Pepisico. What tipped me off was that one of the restaurants served Coke products.
Beilinson also promises that you'll eventually be able to manage your Windows and Unix directories from one terminal, a significant improvement for hard-pressed network administrators.
That's funny, I've been able to manage all of my files from one terminal. I guess I just take that for granted.
All right, rant time: If they're trying to make things easier for "hard-pressed" admins, then why does Microsoft insist on using a GUI? Why aren't there any scripting languages? Don't scripts make things *easier* to manage then repetitive point & clicking? And no, batch files do not count as real scripts.
That is *not* the point of the trial. It is legal in the U.S. for a company to hold a monopoly, it is illegal to use that monopoly to exploit other markets. The government's case is that Microsoft used their OS monopoly to take over the browser market.
The anti-Microsoft sentiment is rather high at Georgia Tech. A few years ago, the administrators came up with a computer ownership policy, requiring all incoming freshman to own a computer with a certain suite of software. No big surprise, they wanted everyone to run Windows 95. After pressure from the students citing free choice, the revised (and current) policy allows for both Windows and MacOS. (The Linux user group is in the process of compiling a compatible package for Linux users.)
I guess the situation would be vastly different in a liberal arts school, where the students wouldn't know nor care about their operating system. However, I find this to be a paradox. They wouldn't (for example) buy a car if it kept stalling every hour. Yet, they accept computer instability as a fact of life. One would think that, given their education in humanities and social sciences, that they wouldn't tolerate such a product.
I find the situation much different at Georgia Tech. Here, a good portion of the Computer Science and Computer Engineering population carry Pilots and/or laptop computers with them.
A nifty module for the Visor would be a cellular ethernet connection. That would definitely make it easier for students to check their e-mail on route to class.
And finally, the obligatory question: Will it run Linux?
The one question I have is does the V-chip also block commericals? Commericals for, say, South Park are nearly as bad as the show itself.
But at this point we get into the whole security-through-obsecurity debate. Does blocking out all of the "offensive" material prevent children learning about it?
Slashdot Mirror
Relying on face recognition a bad idea. Certain segments of the population have a condition called "prosopagnosia" in which victims are unable to recognize faces, even familiar ones like their mother's or even their own. A similar condition is described in the famous book "The Man Who Mistook His Wife for a Hat". Here the researcher describes the more general condition of object agnosia which is the inability to recognize any type of object. Presumably those with object agnosia would fail the inkblot password scheme.
Note that prosopagnosia is not a subset of object agnosia; some with one do not suffer the other (which is the cause of much controversy as to their origins, but that's getting off topic).
Just a note to people who still want to have fun hacking away: TiVos run a custom Linux kernel on a PowerPC board. Those lucky enought with a Series 1 TiVo can hack it the kernel to do stuff like providing a bash prompt or run a web server. Those with a Series 2 with Home Media Option (HMO) can write all sorts of applets to their hearts contents; see www.tivo.com/developer to download the API.
To the original poster: Is it really worth it to build your own system if you reside within the TiVo market? Have you considered things like: hardware costs (a fast processor, video capture board, lots of RAM, motherboard, case), software (time to get the kernel + driver working, time to cobble together a UI), and other intangibles (getting a remote to work, fan noise, getting timely scheduling information)?
Kind of off-topic, but not really, no, you cannot deduct time spent on OSS projects. Under the section "Contributions You May Not Deduct" on page A-4, 2002 1040 instructions, you cannot deduct "Value of your time or services".
So if you're looking to itemize deductions this year, give money to Xiph et al.
Disclaimer: IANALTP (I am not a licensed tax preparer)
You realize that what you are suggesting is very naive? There's a whole class of computing at the embedded system layer. I don't know about you, but I don't want the defibrillator keeping me alive suddenly pause while it's garbage collecting some values. Or in terms of "general computing", the software for an air-traffic controllers needs to make millions (if not billions) of calculations on a radar beam to decide the position and velocity, if any, of an incoming plane. Computers are fast, but memory is slow The OS is going to be spending time allocating memory and bounds checking each radar ping; meanwhile planes will be crashing.
Eventually at some level code needs static buffers. Well-designed programs along with proper code validation techniques ensure a minimal number of errors. Java/C#/language-of-the-month can help software engineering, but by no means are they a panacea.
Hey Gary. What's up?
I had one student who was failing every single assignment, but then magically turned in a program which ran flawlessly, was multithreaded, and had network support. Needless to say, a few seconds on Google turned up the original -- and uncited -- source code.
What a dummy.
First of all, there is Broadcast 2000, a GPL non-linear editor. For your video capturing needs, try dvgrab (assuming that you've got a IEEE-1394 compliant capture card). And as a cheap plug for my own program, I am the author of gvplay, a simple Gnome/GTK video player. I wrote gvplay to help render my special effect (object replacement through tracking and edge detection).
While you make several good points, let me expand on some of them.
(Background: Born & raised in Pittsburgh, now an undergrad at Georgia Tech in Atlanta.)
As I compare the differences between two roughly equal-sized cities, Atlanta and Pittsburgh, I note that while the former is growing leaps and bounds, the latter is shrinking away. I attribute this to several factors:
- Taxes. When I compare the amount I pay in taxes to good ol' Pittsburgh and Pennsylvania, I look with to my friends paying the much smaller Atlanta and Georgia taxes. For example, Pennsylvania imposes a state-wide sales tax and income tax, while Georgia has just an income tax -- and one lower than Pennsylvania. Georgian are mainly taxed by land, but who cares if one is just renting an apartment?
- Education. And by this, I don't mean college-level, for both Atlanta and Pittsburgh are equals. I mean the primary and secondary educations. Having attended the public school system, I must say that I was rarely, if ever, encouraged to learn more about technology. Rather, the teachers were of the old-school (pun attended!) variety. Among other things, they truly believed it to be easier to deny students access to web sites like Rootshell rather than to take measures to tighten security. Security through obsecurity, anyone? I don't know how Atlanta's schools measure up, but I doubt they can be much worse than Pittsburgh's.
- On a related note, priorities. Georgia uses its lottery revenue to fund the HOPE scholarship, which gives students a free ride through college as long as a 3.0 GPA is maintained. Many families in surrounding states move to Georgia whenever their children are about to graduate from high school. Those families would have parents in the 50s or so, which means that they are near the top of their income bracket. And what about Pennsylvania? Their lottery goes towards Senior Citizens. I may be prejudiced by this, but I doubt grandparents are quick to start up new businesses, especially technological-oriented ones.
- Transportation. Marta (Atlanta) kick's the T's (Pittsburgh) butt.
- Nightlife. Although I don't drink, I do know that Atlanta's Buckhead region is a whole lot more exciting than Pittsburgh's South Side.
I would say so. At least to me, I make a distinction between entertainment software and operational software. For the latter category, I prefer to use the best choice available (which usually ends up being open source, but that's a different thread). When I'm working, the software had better work; who cares about stupid add-ons such as the Microsoft paperclip? On the other hand, when I play a computre game, I want all of the bells & whistles -- sound, CD music, textured backgrounds, etc. I'm sorry, but none of the open source games I've seen come even close to that of Railroad Tycoon II.
I think so. I believe the point that Carmack is trying to make is to send a message to retailers about the Linux market.
Most software houses don't sell their works straight to the public. Instead, they publish them and then sell to the retailers (e.g., CompUSA). If the retailers don't demand Linux-based games, then the publishers don't sell any. Thus, they won't bother writing any Linux games.
Take a look at two suggestions I've seen:
Sell just a PC version, but let the user download the binary for his OS.
This logic is severely flawed. Ok, so id knows how many Linux/Mac clients exist. Whee. They then publish their server logs. Whee. Other software companies check out id's page and sees some statistics. How do they know that id did not make up those numbers? In other words, other companies can not verify, for themselves, the Linux market.
Distribute a hybrid CD. For the registration card, allow the user to choose his OS.
This is flawed like above. Again, id could publish the values; yet the numbers can not be emperically verified.
Suppose that id continues as plans. Linux users buy Linux versions of Q3; Mac and PC users do likewise. Now, each retailer can verify for themselves the demand for the different OSs. Knowing those values, they can then demand to the software houses what operating systems they (the publishers) should support.
If the demand for Linux is non-existant, then the retailers won't buy any Linux games from the publishers. Even if the publishers want to support Linux, they might choose to not do so if nobody is going to buy Linux.
But if Q3, Linux version sells extremely well, the retailers are sure to notice. They then pass on this demand to the publishers.
There is a CNNfn just waiting to be slashdotted.
Also, is it just me, or are the ones defending Microsoft all Anonymous Cowards?
Some other pictures of IO are at http://galileo.jpl.nasa.gov/ima ges/io/ioimages.html.
This reminds me of a CNN article discussing WinCE's future. A quote from the article:
Interestingly enough, the first public release of Win NT was 3.0. If I recall correctly, Win 3.0 beta was written to be uncompatible with a non-Microsoft DOS (i.e., DR-DOS). This was also around the time of them dropping the joint MS-IBM OS/2 project. To me, saying that Microsoft doesn't get anything "right until version 3" is rather misleading. This implies that their third release was successful due to a superior product, not marketing or strong-arm tactics.
Not quite. The Huffman algorithm has been proven to produce optimal prefix code, which are codes such that no codeword is also a prefix of another codeword. According to my algorithms book, Huffman compression saves 20% to 90%.
> I'm pretty sure Pizza Hut is owned by Pepsi,
This was true in the past, but times have changed. There is a new franchise in California, "Pac Hut", that has been silently buying up Pizza Huts off of Pepisico. What tipped me off was that one of the restaurants served Coke products.
Beilinson also promises that you'll eventually be able to manage your Windows and Unix directories from one terminal, a significant improvement for hard-pressed network administrators.
That's funny, I've been able to manage all of my files from one terminal. I guess I just take that for granted.
All right, rant time: If they're trying to make things easier for "hard-pressed" admins, then why does Microsoft insist on using a GUI? Why aren't there any scripting languages? Don't scripts make things *easier* to manage then repetitive point & clicking? And no, batch files do not count as real scripts.
That is *not* the point of the trial. It is legal in the U.S. for a company to hold a monopoly, it is illegal to use that monopoly to exploit other markets. The government's case is that Microsoft used their OS monopoly to take over the browser market.
The anti-Microsoft sentiment is rather high at Georgia Tech. A few years ago, the administrators came up with a computer ownership policy, requiring all incoming freshman to own a computer with a certain suite of software. No big surprise, they wanted everyone to run Windows 95. After pressure from the students citing free choice, the revised (and current) policy allows for both Windows and MacOS. (The Linux user group is in the process of compiling a compatible package for Linux users.)
I guess the situation would be vastly different in a liberal arts school, where the students wouldn't know nor care about their operating system. However, I find this to be a paradox. They wouldn't (for example) buy a car if it kept stalling every hour. Yet, they accept computer instability as a fact of life. One would think that, given their education in humanities and social sciences, that they wouldn't tolerate such a product.
I find the situation much different at Georgia Tech. Here, a good portion of the Computer Science and Computer Engineering population carry Pilots and/or laptop computers with them.
A nifty module for the Visor would be a cellular ethernet connection. That would definitely make it easier for students to check their e-mail on route to class.
And finally, the obligatory question: Will it run Linux?
The one question I have is does the V-chip also block commericals? Commericals for, say, South Park are nearly as bad as the show itself.
But at this point we get into the whole security-through-obsecurity debate. Does blocking out all of the "offensive" material prevent children learning about it?