Bridges are built out of lumber and metal. They are built upon concrete and earth. They are bombarded by wind and precipitation. They are driven upon by cars and trucks.
These physical objects are governed by physical laws. Given the properties of a car, I can tell how much stress it will put on a bridge. Given a metal, I determine how resilient it is and extrapolate that result. I can estimate, with reasonable accuracy, the stability of the soil upon which I build my bridge. An I-beam is an I-beam is and I-beam.
This is not the case with software components. Use this object with that IPC mechanism, you segfault. Call this function and fail to catch that error, you hang. Software components interact with each other in weird, unpredictable ways. If a bird lands on one specific part of your bridge, does it crumble to the ground? Of course not. But I have a function in the project I'm working on which will crash if the no-op statement
cout It could also be argued that the "guild"
stevew> already exists - it's called ACM.
It has no punitive ability. No employer will care if you get kicked out of the ACM. It's more like a club, a place folks go to share information and hand out awards.
Bridges are built out of lumber and metal. They are built upon concrete and earth. They are bombarded by wind and precipitation. They are driven upon by cars and trucks.
These physical objects are governed by physical laws. Given the properties of a car, I can tell how much stress it will put on a bridge. Given a metal, I determine how resilient it is and extrapolate that result. I can estimate, with reasonable accuracy, the stability of the soil upon which I build my bridge. An I-beam is an I-beam is and I-beam.
This is not the case with software components. Use this object with that IPC mechanism, you segfault. Call this function and fail to catch that error, you hang. Software components interact with each other in weird, unpredictable ways. If a bird lands on one specific part of your bridge, does it crumble to the ground? Of course not. But I have a function in the project I'm working on which will crash if the no-op statement
cout It could also be argued that the "guild" stevew> already exists - it's called ACM.
It has no punitive ability. No employer will care if you get kicked out of the ACM. It's more like a club, a place folks go to share information and hand out awards.
cryptogeek> So how is the situation _better_ than cryptogeek> this? First, these advances only apply cryptogeek> to public-key encryption, not cryptogeek> secret-key encryption schemes like cryptogeek> DES.
Isn't all encryption using a fixed-length key solvable by a non-deterministic machine in polynomial time? If a password was a maximum of x characters with y characters in its alphabet, wouldn't a non-deterministic machine just have to spawn off y^x copies of itself and try each combination in parallel? Or am I missing something here?
They factored in support costs, but they based it on the current support costs. Current support costs are high, because much of the hard-core support must come from microsoft and is of high price and low quality. With competition, of course, the Baby Bills would have to bother putting together decent support, or be eaten up by the other guys.
Well, I'll give it a shot: automatic, a slang term meaning a firearm that does not need manual reloading, is derived from the same root as automaton, meaning, uhhhh, well, meaning "bot." (When speaking of turing-complete automata, not the finite automata that can only grok regular expressions.)
Actually, reading my C-Tree manuals right now, I wished I was reading the work of a dead guy. /* Just kidding, friendly Faircom attorneys.:) */
True, "bot" and "robot" are different terms, and describe different things. But the term "bot" was derived by shortening "robot." (Unless it was derived by shortenting "botulism," describing how a coder feel when his bot runs into a corner and turns left infinitely.)
Hunt the Wumpus is often used in AI classes (as it was in mine) as a context for coding an agent that operated without human input. That is, we coded a wumpus hunter that worked without any human input.
They've tried giving away copies of UnixWare (or was it OpenServer) free to home users as a way to combat Linux's popularity with hobbyists. It hasn't worked, of course. Opening up their code probably wouldn't save them either, although it would really help out the free software community.
The "SCO Linux" idea is the best one I've heard for SCO to save itself. They would have to do it quick, before all their customers finish porting their stuff to an existing distro.
SCO includes them with some (or perhaps all) distributions under the name "Skunkware" (no affiliation with the NASA project). The skunkware stuff comes on a separate CD and must be installed at the user's inclination -- it is not integrated into the initial installation.
Back in the early 1900s, another time when monopolies were powerful, the ultrarich made arguments against charity, saying that it undermined the capitalist system. They believed instead in "social darwinism," where the smart and beautiful make money and live to procreate, and where the stupid and ugly died of starvation before producing young.
It's interesting that the monopolists here are using a similar argument, that the computer industry will be undermined by free software, that the only good way for software to be produced is with monetary incentives. There was even some antiAmericanism implied, by labeling free software as anti-capitalist.
But it's the same bullshit now as it was 100 years ago. They're trying to use pseudologic to assert that their monopoly is justified, that any other model is unscientific and evil.
I remember some DFC-type parodies in MAD Magazine back in the early 90's. In the subsequent issue they had a letter from Keane complimenting them on the quality of their satire.
While the MAD parodies were of a much higher quality than the Dilbert-Hole, Scott Adams should have done the same thing. Ignoring them, or even complimenting them, just deflates their "cutting-edge"-ness.
I don't know LaTex, but rule-of-thumbs for turing-completeness are (that I can think of):
-- some mechanism for infinite storage (a stack, a natural number with no upper bound, a malloc equivalent, and arbitrarily big string, etc.)
-- some mechanism for decision-making or nondeterminism (an if statement, conditionals, etc.)
-- some mechanism for infinite recursion or infinite looping
-- some sort of "thread of control" including a start point and and end point
Is that everything? I remember discussing this with a CS prof back in college, and those were some of the criterea we brainstormed. An easy proof (for LaTeX) might be to try and code a deterministic semi-Thue rewrite system for it (See the "Handbook of Theoretical Computer Science," volume B chapter 6), which can basically take the form of a recursive search-and-replace function with several search/replace pairs.
Kick ass! Here I thought I had to write robust, elegant code to be productive. Now a CNN reporter (I didn't see the name -- was it Knuth? I heard he's working for them now) tell's me it's all about the number of lines of code. Thank god! Time to double-space my comments.
That is, he's not against the way that MP3's allow wide distribution of music as small files. He just doesn't like the sound quality of MP3 files, and thinks this will cause another file format to someday supercede it. He also bitched about the Rio, not because of the RIAA's nonsense arguments about the free market collapsing because of unenforced copyright tariffs, but because he doesn't think it's ergonomic enough. This strikes me as a legitimate beef -- hell, I hate my mouse and would rather have a trackball. I hate the GIF file format because members of a project I worked on had trouble with coding LZW compression; that doesn't mean I'm the sworn enemy of people who publish freely using GIFs.
I mean, how many people -- newbies or real-life web designers -- actually examine the source of every site they visit? The only people I can think of who would look there unprovoked would be people wanting to copy the layout. Not likely.
<!-------Attention Open Source Enthusiasts!--------------------
A special message for Open Source Software enthusiasts: We are very interested in developing content that takes advantage of Open Source software such as Linux. For starters, we are conducting a Gore 2000 Linux Screen Saver design competition among volunteers to see who can build the best Gore 2000 Linux screen savers. If you would like to submit a Gore 2000 Linux screen saver for consideration, or if you have ideas for this project, please send an e-mail to:
opensource@algore2000.com
Watch this space for more Open Source ideas and volunteer opportunities.
How's a computer going to understand the thought, "I want to look at my AOL stock quotes" when IT CAN'T UNDERSTAND THE ASCII STRING "I want to look at my AOL stock quotes"? Bill Gates should have stayed in college.
Assuming KOffice is open (I assume it is; I haven't seen its licensing), it is likely that exploitable holes in macro security will be fixed in a timely manner. The only way macro viruses will be eliminated is if Microsoft redesigns Office BASIC to be safe, and the only way this will happen is if a devestating-enough virus makes MS customers complain enough that Redmond does something.
As other people have posted, both Emacs and vi had scripting virus-type bugs that have since been fixed. In an open-source environment, security problems can be fixed before they become devastating. In closed-source environment, security problems are fixed only after they become devastating.
If the Melissa virus (which is really pretty mild as far as viruses go) causes MS to close up security holes before geniunely bad viruses are released, then the Melissa virus will have been A Good Thing.
Flames and viruses both serve some good purposes
on
The Melissa Syndrome
·
· Score: 1
Flames and viruses may both come across as hostility, but they share similar positive qualities. They're blunt ways to point out weakness in an argument or system. A wise Win98/Outlook/Word user will look at the Melissa virus and say, "I'm going to move to a system where the makers give enough of a shit about the users to fix the software holes that allow viruses." People have known about the threat of scripting viruses for years, but only when massive damage is done do mainstream folks wake up and pay attention.
So, will it weed out us obsessive/. addicts, seeking only a middle ground of recreational/. users? Or is this only to weed out automated attempts at grabbing moderator status?
The X11 license and BSD license were around before the GNU GPL, and RMS rejected them as being too open. They allowed derivative software to become non-free. For that matter, "public-domain" open source software did the same thing. How is the JCL any different from these? The GPL was partly a response to these overly-open licenses, and less "viral" types of licenses have been concocted after the GPL (e.g., the artistic license). But from cursory glances at sites like freshmeat, it looks like the GPL is the most successful of any of these "free" licenses.
Why don't companies use some of the empty space in mice to store batteries? Granted, the batteries wouldn't be that large, but some power is better than none in a pinch.
Slashdot Mirror
stevew> The bridge metaphor does apply!
Bridges are built out of lumber and metal. They are built upon concrete and earth. They are bombarded by wind and precipitation. They are driven upon by cars and trucks.
These physical objects are governed by physical laws. Given the properties of a car, I can tell how much stress it will put on a bridge. Given a metal, I determine how resilient it is and extrapolate that result. I can estimate, with reasonable accuracy, the stability of the soil upon which I build my bridge. An I-beam is an I-beam is and I-beam.
This is not the case with software components. Use this object with that IPC mechanism, you segfault. Call this function and fail to catch that error, you hang. Software components interact with each other in weird, unpredictable ways. If a bird lands on one specific part of your bridge, does it crumble to the ground? Of course not. But I have a function in the project I'm working on which will crash if the no-op statement
cout It could also be argued that the "guild"
stevew> already exists - it's called ACM.
It has no punitive ability. No employer will care if you get kicked out of the ACM. It's more like a club, a place folks go to share information and hand out awards.
......
stevew> The bridge metaphor does apply!
Bridges are built out of lumber and metal. They are built upon concrete and earth. They are bombarded by wind and precipitation. They are driven upon by cars and trucks.
These physical objects are governed by physical laws. Given the properties of a car, I can tell how much stress it will put on a bridge. Given a metal, I determine how resilient it is and extrapolate that result. I can estimate, with reasonable accuracy, the stability of the soil upon which I build my bridge. An I-beam is an I-beam is and I-beam.
This is not the case with software components. Use this object with that IPC mechanism, you segfault. Call this function and fail to catch that error, you hang. Software components interact with each other in weird, unpredictable ways. If a bird lands on one specific part of your bridge, does it crumble to the ground? Of course not. But I have a function in the project I'm working on which will crash if the no-op statement
cout It could also be argued that the "guild"
stevew> already exists - it's called ACM.
It has no punitive ability. No employer will care if you get kicked out of the ACM. It's more like a club, a place folks go to share information and hand out awards.
cryptogeek> So how is the situation _better_ than
cryptogeek> this? First, these advances only apply
cryptogeek> to public-key encryption, not
cryptogeek> secret-key encryption schemes like
cryptogeek> DES.
Isn't all encryption using a fixed-length
key solvable by a non-deterministic machine
in polynomial time? If a password was a maximum
of x characters with y characters in its alphabet,
wouldn't a non-deterministic machine just have
to spawn off y^x copies of itself and try each
combination in parallel? Or am I missing
something here?
They factored in support costs, but they based it on the current support costs. Current support costs are high, because much of the hard-core support must come from microsoft and is of high price and low quality. With competition, of course, the Baby Bills would have to bother putting together decent support, or be eaten up by the other guys.
Well, I'll give it a shot: automatic, a slang term meaning a firearm that does not need manual reloading, is derived from the same root as automaton, meaning, uhhhh, well, meaning "bot." (When speaking of turing-complete automata, not the finite automata that can only grok regular expressions.)
:) */
Actually, reading my C-Tree manuals right now, I wished I was reading the work of a dead guy.
/* Just kidding, friendly Faircom attorneys.
True, "bot" and "robot" are different terms, and describe different things. But the term "bot" was derived by shortening "robot." (Unless it was derived by shortenting "botulism," describing how a coder feel when his bot runs into a corner and turns left infinitely.)
Hunt the Wumpus is often used in AI classes (as it was in mine) as a context for coding an agent that operated without human input. That is, we coded a wumpus hunter that worked without any human input.
They've tried giving away copies of UnixWare (or was it OpenServer) free to home users as a way to combat Linux's popularity with hobbyists. It hasn't worked, of course. Opening up their code probably wouldn't save them either, although it would really help out the free software community.
The "SCO Linux" idea is the best one I've heard for SCO to save itself. They would have to do it quick, before all their customers finish porting their stuff to an existing distro.
I'm typing this in Netscape 4.05, running Linux/PPC 4.1. Or am I misunderstaning something about your post?
I have been tempted to get YDL, or TurboLinux, just to test out the distros. How do you like YDL? I've seen no reviews of it.
SCO includes them with some (or perhaps all) distributions under the name "Skunkware" (no affiliation with the NASA project). The skunkware stuff comes on a separate CD and must be installed at the user's inclination -- it is not integrated into the initial installation.
Back in the early 1900s, another time when monopolies were powerful, the ultrarich made arguments against charity, saying that it undermined the capitalist system. They believed instead in "social darwinism," where the smart and beautiful make money and live to procreate, and where the stupid and ugly died of starvation before producing young.
It's interesting that the monopolists here are using a similar argument, that the computer industry will be undermined by free software, that the only good way for software to be produced is with monetary incentives. There was even some antiAmericanism implied, by labeling free software as anti-capitalist.
But it's the same bullshit now as it was 100 years ago. They're trying to use pseudologic to assert that their monopoly is justified, that any other model is unscientific and evil.
I remember some DFC-type parodies in MAD Magazine back in the early 90's. In the subsequent issue they had a letter from Keane complimenting them on the quality of their satire.
While the MAD parodies were of a much higher quality than the Dilbert-Hole, Scott Adams should have done the same thing. Ignoring them, or even complimenting them, just deflates their "cutting-edge"-ness.
I don't know LaTex, but rule-of-thumbs for turing-completeness are (that I can think of):
-- some mechanism for infinite storage (a stack, a natural number with no upper bound, a malloc equivalent, and arbitrarily big string, etc.)
-- some mechanism for decision-making or nondeterminism (an if statement, conditionals, etc.)
-- some mechanism for infinite recursion or infinite looping
-- some sort of "thread of control" including a start point and and end point
Is that everything? I remember discussing this with a CS prof back in college, and those were some of the criterea we brainstormed. An easy proof (for LaTeX) might be to try and code a deterministic semi-Thue rewrite system for it (See the "Handbook of Theoretical Computer Science," volume B chapter 6), which can basically take the form of a recursive search-and-replace function with several search/replace pairs.
Hope this was remotely helpful. Good luck!
Kick ass! Here I thought I had to write robust, elegant code to be productive. Now a CNN reporter (I didn't see the name -- was it Knuth? I heard he's working for them now) tell's me it's all about the number of lines of code. Thank god! Time to double-space my comments.
That is, he's not against the way that MP3's allow wide distribution of music as small files. He just doesn't like the sound quality of MP3 files, and thinks this will cause another file format to someday supercede it. He also bitched about the Rio, not because of the RIAA's nonsense arguments about the free market collapsing because of unenforced copyright tariffs, but because he doesn't think it's ergonomic enough. This strikes me as a legitimate beef -- hell, I hate my mouse and would rather have a trackball. I hate the GIF file format because members of a project I worked on had trouble with coding LZW compression; that doesn't mean I'm the sworn enemy of people who publish freely using GIFs.
I mean, how many people -- newbies or real-life web designers -- actually examine the source of every site they visit? The only people I can think of who would look there unprovoked would be people wanting to copy the layout. Not likely.
<!-------Attention Open Source Enthusiasts!--------------------
A special message for Open Source Software enthusiasts: We are
very interested in developing content that takes advantage of Open
Source software such as Linux. For starters, we are conducting a
Gore 2000 Linux Screen Saver design competition among volunteers
to see who can build the best Gore 2000 Linux screen savers. If
you would like to submit a Gore 2000 Linux screen saver for
consideration, or if you have ideas for this project, please
send an e-mail to:
opensource@algore2000.com
Watch this space for more Open Source ideas and
volunteer opportunities.
-------------------------------------------------- ------------>
More April fool's crap? I thought the point of having editors was to *increase* the SNR.
How's a computer going to understand the thought, "I want to look at my AOL stock quotes" when IT CAN'T UNDERSTAND THE ASCII STRING "I want to look at my AOL stock quotes"? Bill Gates should have stayed in college.
Assuming KOffice is open (I assume it is; I haven't seen its licensing), it is likely that exploitable holes in macro security will be fixed in a timely manner. The only way macro viruses will be eliminated is if Microsoft redesigns Office BASIC to be safe, and the only way this will happen is if a devestating-enough virus makes MS customers complain enough that Redmond does something.
As other people have posted, both Emacs and vi had scripting virus-type bugs that have since been fixed. In an open-source environment, security problems can be fixed before they become devastating. In closed-source environment, security problems are fixed only after they become devastating.
If the Melissa virus (which is really pretty mild as far as viruses go) causes MS to close up security holes before geniunely bad viruses are released, then the Melissa virus will have been A Good Thing.
Flames and viruses may both come across as hostility, but they share similar positive qualities. They're blunt ways to point out weakness in an argument or system. A wise Win98/Outlook/Word user will look at the Melissa virus and say, "I'm going to move to a system where the makers give enough of a shit about the users to fix the software holes that allow viruses." People have known about the threat of scripting viruses for years, but only when massive damage is done do mainstream folks wake up and pay attention.
So, will it weed out us obsessive /. addicts, seeking only a middle ground of recreational /. users? Or is this only to weed out automated attempts at grabbing moderator status?
The X11 license and BSD license were around before the GNU GPL, and RMS rejected them as being too open. They allowed derivative software to become non-free. For that matter, "public-domain" open source software did the same thing. How is the JCL any different from these? The GPL was partly a response to these overly-open licenses, and less "viral" types of licenses have been concocted after the GPL (e.g., the artistic license). But from cursory glances at sites like freshmeat, it looks like the GPL is the most successful of any of these "free" licenses.
(Correct me if I'm wrong.)
Why don't companies use some of the empty space in mice to store batteries? Granted, the batteries wouldn't be that large, but some power is better than none in a pinch.
I can't believe the most intelligent post on this article has come from Meept.