Napster will get screwed, the press will report a "victory for piracy", everyone will ignore the press and use alternate resources.
Heck, why ignore the press? My local Fox station ran a story on their 10o'clock news on Napster being shut down and they actually got most of those as alternatives. They took the time to display the links, and had the person actually say each link, giving plenty of time to write it down. I found that rather funny - here they are, listing five alternatives to Napster. They got Gnutella and Freenet, along with three others I don't remember. So, maybe the media response won't be what you seem to think it will be...
Yeah, but it's nice to have something to test against.
Suppose a security flaw is found, but no code is made publicly available. So the vendor has the developers create a patch. The developers create a simple test case, make sure that patch works against it, and ship it. Suppose now that the test case they used was inadaquate. Maybe it didn't fully use the exploit, or maybe there was a bug in it causing it to only use the exploit in one fassion.
After a few months, all of a sudden, some sys-admin finds that his "patched" system has just been cracked when a black-hat created a program - that properly implemented the exploit.
Coding demonstrations of how the exploit works can be very benificial anyway, because presumably these have been tested and debugged to fully demonstrate the flaw. It allows the developers to be completely certain that the patch actually completely and fully patches the system. Besides, if most of the script kiddies are going to use that version anyway, it cuts down on the chances of a blackhat writing a new one that gets around the fix. (After all, why bother reinventing the wheel?) It also means that there are fewer potential methods of using an exploit, so even if the patch is inadaquate, it'll work against the most common script.
It's also nice to be able to look at code and see exactly how the exploit works. Fixing a problem when you can test it against a known working exploit is much easier than needing to first write a test program to do the exploit and then fixing the machine against that test.
The real problem with having a test case is when a company takes two months to release the patch while countless sys-admins are finding their servers are getting broken into. Then again, maybe if there was no demo, the sys-admins would get some time off before some black-hat coded a version anyway.
Trust me, there are people out there who would program an exploit they discovered off a security advisory just to see if it works. They want to show off those new m4d h4x0r 5k177z they learned in their CS class, and creating and releasing a program to exploit a known (but demoless) security flaw is one way to do that. All no public demo would do is buy some time. And I don't think it would even be much time.
Most main company developers have a subscription to MSDN, and MSDN contains full versions of most Microsoft development software. (IE, OSes, development tools, Office, etc.). These developers basically get these tools for free - since the company they work for is the one paying the $2000+ subscription fee. If they find that C# works for them, then they'll use it. If not, they won't. If companies find that they have reasons to develope in C#, then it'll take off, just like any other language. Actually, given the fact that VBScript is being used, I'd say any MS language has an unfair advantage...:)
(Note to moderators: yes, this is offtopic to the article - but it's ontopic in the thread. Read posts' threads before marking posts as "offtopic.")
Your name is Scott, huh? So you wouldn't happen to be Scott Sellers, right? Sorry, conspiracy theory... After all, you do seem to be rather rabid about 3dfx...
(Which is OK with me, I personally like 3dfx (right now) better anyway, for most of the reasons you've stated. I dunno, may change depending on what they do in the future... (After all, I thought Windows 3.1 was pretty cool and didn't care to see MS go away until after Win95...) Although I'd still like to get my hands on a Voodoo 5:))
Or just use the line-in port on your soundcard and start recording songs off the radio and MP3ing them! I've done similar things to play tapes onto MP3s. (I've actually got a set of MP3s that where orginally vinal, copied onto tape, and then copied onto MP3. Sound quality's quite nice for 3rd generation audio.)
All you need to do is go to Radio Shack or your local electronics store and look for a stereo to headphone convertor, whatever they're called. I found one for about $4. (Used to connect PlayStation through computer's 5.1 surround sound as opposed to TV's crappy stereo speakers - you know you're a geek when your computer sounds better than your TV or stereo...) Recording off of this is simple, although you'll need some disk space if you can't compress on-the-fly. (It's about a 10 MB/min, actually a little less (about 8.75MB/min if my calcs are right), for CD quality audio. Plan on 10 MB/min though so that you can encode to MP3.)
This can all be done with a simple Walkman too if you want to hook it up - just get a cable to plug into the Line In/Microphone port. Again, Radio Shack or your local electronic store.
(Yay, Mozilla build 2000072608 will allow me to post again! The latest nightlies kept on crashing on POSTs...)
To someone who can't read music, a Bach score is as indecipherable as a C program is to someone who doesn't know C.
And to extend, others who cannot read the source (music or C) can still see the beauty when the source is run, either on a computer, or played by an orchestra. In other words, you don't need to be a programmer to understand the beauty of (for instance) a well-written and nicely designed internet browser. However, programmers might notice some subtilties of design and neat little quirks. Likewise, a musician might notice interesting little passages and cute little tricks with harmonics that the average listener might miss.
Besides, just like an artist feels a sort of creative rush when they create something beautiful, and I feel that rush of creating something useful with a computer program that finally works sans (major) bugs. The same feeling of accomplishment that people feel after completeing anything. Given that people like to code and give it away for others to look at, and that terms such as spagetti code are used to describe design flaws and such, I'd say that there's a lot of expression in code. Same as there is expersion is architecture, and in inventing. The way a complicated circuit works may not be interesting to non-electrical engineers, but there is still expression in the methods of using it. The choices made in creating anything show expression on the part of the creator.
The mere existance of Open Source and the existance of an Open Source community seems to suggest that many people do find expression through code. After all, why bother creating code if won't get paid? Well, for the communities appreciation, of course. That seems to suggest that some people find code to be an expression. I don't think Open Source could exist should code not be a method of expression for people.
He's a troll - look at his posting history. He's one of the better trolls, granted, but a troll non the less. Try looking for "NPO Technologies" on the web (the IT firm he claims to work for). Surely they'd have some form of web presence, they're an IT firm, right? Nearly every company has a web site, and I'd be very surprised to find that an IT consulting firm would not have one. Yet searching AltaVista for either npo technologies or n.p.o. technologies returns nothing. It doesn't seem to exist. I'd expect there to be some web page about them somewhere, but there isn't... he's a troll. Albeit a good one.
...Ah, but what about fiddling with those genes post-conception? That way, the traditional (In. Out. Repeat as necessary.) method can still be used to impregnate, and the parents can play with genes to their hearts content for the first two-three months of the pregnancy. (Maybe with improved genetic theorapy, you could play with peoples genes far after that, maybe even with full-grown adults. Would be useful as a cure to various genetic illnesses...)
So the new process would be...
(The female, hopefully) Get pregnant.
Go to genetic councelor.
Determine changes.
Make changes (preferably not killing baby in process).
Wait 7-10 months, stirring occasionally.
Presto, designer-baby.
'course, if it was possible to change your own genes, hmm... "Maybe I'd like green eyes today, and I think that I'd like to grow three inches."
To clarify: By developer I mean the people making the program (code, binary, whatever) with the license. By end-user, I mean the people receiving the program.
As someone writing code, I'd rather release it under the GPL since that gives me the freedom to see all changes made to my code. On the other hand, as someone using other code, I'd prefer a BSD-style license since that maximizes my freedom to use it.
Yeah, I think the original poster reversed the licenses. BSD-style license are more free to the end-user/public domain while the GPL is more free to the developer(s). The main difference is that with a BSD style license, one can do almost anything with the source - it's almost like not taking the copyright at all. The GPL on the other hand restricts the usage of the code, allowing the developer more control, and therefore more freedom to do what he/she pleases with the source.
D'oh! Wrong page. The actual page was 713. Woulda gotten it right hadn't Mozilla crashed, forcing me into Netscape 4 where I just took the first reference to "slashdot" it found.
Although right after the real quote, there's an interesting typo...
21 THE COURT: I am ruling on relevance. There has been
22 no clam here.
I guess they don't like seafood...
(Also, to be complete, Slashdot gets mentioned once more on page 720.)
Even completely botched Motif app running over network is way more responsive than Swing running locally
...and Swing can only be run remotely over X. I was mainly trying to point out that there's no need to get rid of the useful parts of the current X desktops (pluggable UI's) if someone migrated over to another system. Personally, I'd like to see a GUI system that supports remote sessions where all the widgets were done server-side (ie, where the monitor/inputs devices/etc. are). This means that text fields and the like would all handle input server-side, while the processing of the input was done client-side (across the link).
(PS, what did you do the moderators? Based on your posting history, they're out to get you.)
Right now, they aren't even space-effective. The current working fuel-cell car are basically a drivers area, with the rest of the area normally reserved for passengers taken up by fuel-cells. They won't really be feasible for another 20-30 years optimistically, quite probably longer. I know that some work related to fuel cells is being done at WPI, but all they have about it is this short piece. I went looking for more details, but the "Center for Fuel Cell Studies" link (located here) is greyed out. (As is the fuel cell laboratory link...)
Very good, you corrected the "feal" mistake. Unfortunately, the (much more obvious, I think) mistake of ppointed is still there.
infodragon ppointed [sic] us to a Penguinista article which does a very good job of addressing the issue surrounding the recent Tucows and ZD-Net complaining that Linux reviews are bought instead of earned. It says pretty much exactly what I feel about the whole thing.
Good to know some mistakes are fixed, though. (what I feel used to read what I feal). Maybe he'll fix the ppointed mistake, so the parent comment really won't make an sense.
X has some...unpleastness, but it's also incredibly flexible and useful. Remote display absolutely ROCKS. Window manager independence makes be drool. Widget choice makes me horny. Etc.
There's nothing saying that you needn't have plugable window/widget managers for a new non-X system. And keep in mind, X has it's own widgets, which nobody uses.
For a look at a plugable, cross-platform UI, look at Swing from Sun for Java. (Sometimes refered to as the Java Foundation Classes, fortunately it's nicer than the Microsoft Foundation Classes where it seems they stole their name from...) It's got pluggable look-and-feels (so it's basically skinnable - although none of the current l&f's seem to be particularly skinnable user-wise). There are many things I don't like about X, and most of them have to do with the fact that the pluggable windows/widget managers happen to be done client-side when they really should be done on my server... (Try running a pic-based GTK+ theme, using a pic based Sawmill scheme, over a 56.6K modem connected at 46.6Kbs - then see if you like pic-based schemes...)
I know what you mean - last year I was asked to design a web site, which is based on XML technologies. Originally, the documents were going to be written in XML and use XSL (yes, XSL, not XSLT, they are different), but I ended up going with plain HTML. (Actually, XHTML.) Going over to the W3C, I downloaded the CSS spec and started writing CSS compliant pages - that didn't render under IE5 OR Netscape. Correcting for IE, I broke Netscape - correcting for Netscape, I broke IE.
I've just recently updated my web page (http://www.wpi.edu/~dpotter/,/. it and die as I'll lose my account) to support both IE, Netscape 4, and Mozilla through use of Javascript. Disable Javascript, though, and it won't work. I have to use Javascript to detect the browser version and then route around incompatibilities in both browsers. (Look at the.css file and you'll notice that I've copied a basic style rule to many elements because Netscape 4 doesn't follow the inheritance rule properly - very annoying.)
At one point, I was ready to throw in the towel with the XML web page and just say "IE only" but I can't because the company I work for uses Netscape 4.73 as their standard browser! (And, BTW, so does the armed forces:).) It's now being designed by someone else though (probably a good thing), and I think he's just using very basic HTML.
At one point, I was ready to write my own browser to properly support CSS. In fact, I am writing my own parser/renderer to support CSS2 - it's not available anywhere yet, but it's in Java (for now, maybe in C++ later), and when it's ready (it's really, REALLY, in early stages right now), I'll release it under the GPL (sorry BSD fans:)). It's very, VERY annoying to try and create a web page that utilizes the features specified in standards at least two years old, just to find that nobody actually supports them.
On that note, Mozilla's HTML/CSS support looks very, very nice, and I can't wait for Mozilla to be ready for prime-time. Keep up the good work, Mozilla! Submit bugs to help them!
...but my plan was to instead have a long rant (maybe by Katz? Nah, he's on the other side of the issue...) on copyright infringment... Get a really poor encoder too so that the encoded song would larger than average. Maybe encode the rant at 256kbs or something...
Well, all you Napster users out there can be glad I don't have an account.
Yeah, it would seem that someone left off a closing quote when writing their HTML. Try purchasing the book at ThinkGeek as I'd expect it'll work better than a large Katz article... And someone on the editing staff, add the " to the end. Honestly, <a href="http://www.thinkgeek.com>ThinkGeek</a> ? Such poor HTML skills.:)
Once microsoft buys microsoft.com there is no reason for microsoft.org or microsoft.net to be around...
Sure there is:
microsoft.com - all of Microsoft's buisness related information, including stock-holder reports, product information, and links to ways to purchase MS products.
microsoft.org - Microsoft User Groups (yes, they exist - my (college, so we're immature) LUG had fun flaming the local NTUG around the Win2K launch, but that's another story). A site dedicated to the various MS UG's out there.
microsoft.net - Just another way of accessing MSN. Or maybe a way in for Microsoft employees to access the Microsoft intranet.
While more TLDs might make it more expensive to snarf up all the domain names, all it means is that a company has to be more creative in their excuses as to why they need the name. I wouldn't expect companies to give up their practice of eating up all the domain names that they can, anyway. All that would do is make them more creative in the ways they do it. (Force Microsoft to be innovative in taking *microsoft*.*:))
That, or you could simply have subsequent votes on the same token CHANGE the vote for that token.
That's solving the wrong problem - while maybe it'd be nice to change your vote, it still doesn't change the fact that you'd be traceable because your "token" still is marked. And even if you do have checksums preventing J0hn Sc1p7 k1dd13 from creating phony tokens, once they validate and get a token, there's nothing to stop them from attempting to use it twice. The solution might be "don't associate John Doe with his token" but there's still a problem with that - it still might be possible to trace you to your ballot.
The original idea was "you validate, get a token. You go over to a different server, and vote." The problem was, then that token supposedly evaporated (I guess.). To make it so that I can't vote twice by figuring out what the token looks like means that the tokens need to be marked, so that the server can see if they've already been used. Even if in the database of registered voters there is no indication of what token they received, I'll bet it would be possible to look at database transaction logs for even just server transaction logs and figure it out.
Even assuming there's NO WAY server-side to figure out who voted for who, it still might be possible to figure out which token a certain person received and then via packet sniffing and cracked encryption figure out who voted for who. And in ellections, there's much more of a chance of people interested in doing that, so security through obscurity is even less of a solution here than it is anywhere else.
The token would have to have identifying marks. If it's an E-token, what's to prevent it from being copied? So John Doe validates, gets the token, and then starts copying the token and starts voting with his duplicated tokens for a write-in canidate who happens to be... John Doe.
Solution? Each token is unique and can only be used once. But then...
TRANSACTION LOG FOR VOTE SERVER User authenicated as John Doe, given token #AA-01431940-F294. Token #AA-01431940-F294 was used to vote for canidate Jane Doe. ATTEMPT TO USE TOKEN #AA-01431940-F294 AFTER INITAL VOTE - subsequent vote ignored.
Don't forget, even separating the two servers (one validates, one votes) doesn't work, because then all someone needs to do to track who voted for whom is look at the two logs. The only REAL solution would be not to keep logs, but I doubt a sysadmin would REALLY be willing not to keep detailed logs of attempts to access the system... (Hmm, look, we've got 3000 attempts to use token #1F-00031337-1337 from IP 221.142.391.257 - maybe it's time to block that IP...) Plus I'd expect that various law enforcement agents would want the logs to attempt to trace possible voter fraud. (One person attempting to validate as a different person, or people attempting to duplicate "tokens.")
We are talking about selling the radio in the car. The radio isn't the car, and doesn't make it run. It makes it nicer...
Uh, that's more like selling the engine out of a car and not the radio. Selling the radio would be like selling off the computer speakers. A computer will not run unless it has (at least one) OS on it. If I decide to buy a brand new Dell computer and uninstall Windows from it, I have a nice, non-bootable piece of hardware. I'd have to install a new OS in it's place to use it. (Whether it be FreeBSD, MacOS, whatever.)
(A better car analogy might be something along the lines of the steering wheel, or the frame, or something, but there's few things in life which really make a good "hardware vs. software vs. OS" analogy...)
I can't imagine that the number of people buying "naked PCs" is a larger segment of the market though. Most people go down to the local "computer supercenter" and buy a PC on credit. Usually preinstalled. Very FEW people I know go to Mom 'n' Pop's PC Place and buy a PC. Even if they did, most come preinstalled with an OS. (Sometimes, even an OS of your choice.) Joe Six-Pack just goes to the nearest computer store he knows and orders the model the salesman tells him to.
Besides, most of the people who would buy PC's without an OS are in the same segment of people likely to use Linux/*BSD/BeOS/etc. Of course, included in this group are the w4r3z d00ds likely to pirate Windows anyway. (BTW, I know some people who like warez who like Linux 'cause it's free as in beer, so there is overlapping, too.)
Slashdot Mirror
Heck, why ignore the press? My local Fox station ran a story on their 10o'clock news on Napster being shut down and they actually got most of those as alternatives. They took the time to display the links, and had the person actually say each link, giving plenty of time to write it down. I found that rather funny - here they are, listing five alternatives to Napster. They got Gnutella and Freenet, along with three others I don't remember. So, maybe the media response won't be what you seem to think it will be...
Suppose a security flaw is found, but no code is made publicly available. So the vendor has the developers create a patch. The developers create a simple test case, make sure that patch works against it, and ship it. Suppose now that the test case they used was inadaquate. Maybe it didn't fully use the exploit, or maybe there was a bug in it causing it to only use the exploit in one fassion.
After a few months, all of a sudden, some sys-admin finds that his "patched" system has just been cracked when a black-hat created a program - that properly implemented the exploit.
Coding demonstrations of how the exploit works can be very benificial anyway, because presumably these have been tested and debugged to fully demonstrate the flaw. It allows the developers to be completely certain that the patch actually completely and fully patches the system. Besides, if most of the script kiddies are going to use that version anyway, it cuts down on the chances of a blackhat writing a new one that gets around the fix. (After all, why bother reinventing the wheel?) It also means that there are fewer potential methods of using an exploit, so even if the patch is inadaquate, it'll work against the most common script.
It's also nice to be able to look at code and see exactly how the exploit works. Fixing a problem when you can test it against a known working exploit is much easier than needing to first write a test program to do the exploit and then fixing the machine against that test.
The real problem with having a test case is when a company takes two months to release the patch while countless sys-admins are finding their servers are getting broken into. Then again, maybe if there was no demo, the sys-admins would get some time off before some black-hat coded a version anyway.
Trust me, there are people out there who would program an exploit they discovered off a security advisory just to see if it works. They want to show off those new m4d h4x0r 5k177z they learned in their CS class, and creating and releasing a program to exploit a known (but demoless) security flaw is one way to do that. All no public demo would do is buy some time. And I don't think it would even be much time.
(Note to moderators: yes, this is offtopic to the article - but it's ontopic in the thread. Read posts' threads before marking posts as "offtopic.")
(Which is OK with me, I personally like 3dfx (right now) better anyway, for most of the reasons you've stated. I dunno, may change depending on what they do in the future... (After all, I thought Windows 3.1 was pretty cool and didn't care to see MS go away until after Win95...) Although I'd still like to get my hands on a Voodoo 5 :))
All you need to do is go to Radio Shack or your local electronics store and look for a stereo to headphone convertor, whatever they're called. I found one for about $4. (Used to connect PlayStation through computer's 5.1 surround sound as opposed to TV's crappy stereo speakers - you know you're a geek when your computer sounds better than your TV or stereo...) Recording off of this is simple, although you'll need some disk space if you can't compress on-the-fly. (It's about a 10 MB/min, actually a little less (about 8.75MB/min if my calcs are right), for CD quality audio. Plan on 10 MB/min though so that you can encode to MP3.)
This can all be done with a simple Walkman too if you want to hook it up - just get a cable to plug into the Line In/Microphone port. Again, Radio Shack or your local electronic store.
(Yay, Mozilla build 2000072608 will allow me to post again! The latest nightlies kept on crashing on POSTs...)
And to extend, others who cannot read the source (music or C) can still see the beauty when the source is run, either on a computer, or played by an orchestra. In other words, you don't need to be a programmer to understand the beauty of (for instance) a well-written and nicely designed internet browser. However, programmers might notice some subtilties of design and neat little quirks. Likewise, a musician might notice interesting little passages and cute little tricks with harmonics that the average listener might miss.
Besides, just like an artist feels a sort of creative rush when they create something beautiful, and I feel that rush of creating something useful with a computer program that finally works sans (major) bugs. The same feeling of accomplishment that people feel after completeing anything. Given that people like to code and give it away for others to look at, and that terms such as spagetti code are used to describe design flaws and such, I'd say that there's a lot of expression in code. Same as there is expersion is architecture, and in inventing. The way a complicated circuit works may not be interesting to non-electrical engineers, but there is still expression in the methods of using it. The choices made in creating anything show expression on the part of the creator.
The mere existance of Open Source and the existance of an Open Source community seems to suggest that many people do find expression through code. After all, why bother creating code if won't get paid? Well, for the communities appreciation, of course. That seems to suggest that some people find code to be an expression. I don't think Open Source could exist should code not be a method of expression for people.
He's a troll - look at his posting history. He's one of the better trolls, granted, but a troll non the less. Try looking for "NPO Technologies" on the web (the IT firm he claims to work for). Surely they'd have some form of web presence, they're an IT firm, right? Nearly every company has a web site, and I'd be very surprised to find that an IT consulting firm would not have one. Yet searching AltaVista for either npo technologies or n.p.o. technologies returns nothing. It doesn't seem to exist. I'd expect there to be some web page about them somewhere, but there isn't... he's a troll. Albeit a good one.
So the new process would be...
'course, if it was possible to change your own genes, hmm... "Maybe I'd like green eyes today, and I think that I'd like to grow three inches."
As someone writing code, I'd rather release it under the GPL since that gives me the freedom to see all changes made to my code. On the other hand, as someone using other code, I'd prefer a BSD-style license since that maximizes my freedom to use it.
Yeah, I think the original poster reversed the licenses. BSD-style license are more free to the end-user/public domain while the GPL is more free to the developer(s). The main difference is that with a BSD style license, one can do almost anything with the source - it's almost like not taking the copyright at all. The GPL on the other hand restricts the usage of the code, allowing the developer more control, and therefore more freedom to do what he/she pleases with the source.
Although right after the real quote, there's an interesting typo...
21 THE COURT: I am ruling on relevance. There has been
22 no clam here.
I guess they don't like seafood...
(Also, to be complete, Slashdot gets mentioned once more on page 720.)
How hard is it to make that a link? Same page as a link. And the lines occur on page 652. (Insert rant about no anchors in the text here.)
Want some Penguin mints? Caffeine good. Mmm. Caffeine. Good.
...and Swing can only be run remotely over X. I was mainly trying to point out that there's no need to get rid of the useful parts of the current X desktops (pluggable UI's) if someone migrated over to another system. Personally, I'd like to see a GUI system that supports remote sessions where all the widgets were done server-side (ie, where the monitor/inputs devices/etc. are). This means that text fields and the like would all handle input server-side, while the processing of the input was done client-side (across the link).
(PS, what did you do the moderators? Based on your posting history, they're out to get you.)
Right now, they aren't even space-effective. The current working fuel-cell car are basically a drivers area, with the rest of the area normally reserved for passengers taken up by fuel-cells. They won't really be feasible for another 20-30 years optimistically, quite probably longer. I know that some work related to fuel cells is being done at WPI, but all they have about it is this short piece. I went looking for more details, but the "Center for Fuel Cell Studies" link (located here) is greyed out. (As is the fuel cell laboratory link...)
Very good, you corrected the "feal" mistake. Unfortunately, the (much more obvious, I think) mistake of ppointed is still there.
Good to know some mistakes are fixed, though. (what I feel used to read what I feal). Maybe he'll fix the ppointed mistake, so the parent comment really won't make an sense.
There's nothing saying that you needn't have plugable window/widget managers for a new non-X system. And keep in mind, X has it's own widgets, which nobody uses.
For a look at a plugable, cross-platform UI, look at Swing from Sun for Java. (Sometimes refered to as the Java Foundation Classes, fortunately it's nicer than the Microsoft Foundation Classes where it seems they stole their name from...) It's got pluggable look-and-feels (so it's basically skinnable - although none of the current l&f's seem to be particularly skinnable user-wise). There are many things I don't like about X, and most of them have to do with the fact that the pluggable windows/widget managers happen to be done client-side when they really should be done on my server... (Try running a pic-based GTK+ theme, using a pic based Sawmill scheme, over a 56.6K modem connected at 46.6Kbs - then see if you like pic-based schemes...)
I've just recently updated my web page (http://www.wpi.edu/~dpotter/, /. it and die as I'll lose my account) to support both IE, Netscape 4, and Mozilla through use of Javascript. Disable Javascript, though, and it won't work. I have to use Javascript to detect the browser version and then route around incompatibilities in both browsers. (Look at the .css file and you'll notice that I've copied a basic style rule to many elements because Netscape 4 doesn't follow the inheritance rule properly - very annoying.)
At one point, I was ready to throw in the towel with the XML web page and just say "IE only" but I can't because the company I work for uses Netscape 4.73 as their standard browser! (And, BTW, so does the armed forces :).) It's now being designed by someone else though (probably a good thing), and I think he's just using very basic HTML.
At one point, I was ready to write my own browser to properly support CSS. In fact, I am writing my own parser/renderer to support CSS2 - it's not available anywhere yet, but it's in Java (for now, maybe in C++ later), and when it's ready (it's really, REALLY, in early stages right now), I'll release it under the GPL (sorry BSD fans :)). It's very, VERY annoying to try and create a web page that utilizes the features specified in standards at least two years old, just to find that nobody actually supports them.
On that note, Mozilla's HTML/CSS support looks very, very nice, and I can't wait for Mozilla to be ready for prime-time. Keep up the good work, Mozilla! Submit bugs to help them!
Well, all you Napster users out there can be glad I don't have an account.
Yeah, it would seem that someone left off a closing quote when writing their HTML. Try purchasing the book at ThinkGeek as I'd expect it'll work better than a large Katz article... And someone on the editing staff, add the " to the end. Honestly, <a href="http://www.thinkgeek.com>ThinkGeek</a> ? Such poor HTML skills. :)
Sure there is:
While more TLDs might make it more expensive to snarf up all the domain names, all it means is that a company has to be more creative in their excuses as to why they need the name. I wouldn't expect companies to give up their practice of eating up all the domain names that they can, anyway. All that would do is make them more creative in the ways they do it. (Force Microsoft to be innovative in taking *microsoft*.* :))
That's solving the wrong problem - while maybe it'd be nice to change your vote, it still doesn't change the fact that you'd be traceable because your "token" still is marked. And even if you do have checksums preventing J0hn Sc1p7 k1dd13 from creating phony tokens, once they validate and get a token, there's nothing to stop them from attempting to use it twice. The solution might be "don't associate John Doe with his token" but there's still a problem with that - it still might be possible to trace you to your ballot.
The original idea was "you validate, get a token. You go over to a different server, and vote." The problem was, then that token supposedly evaporated (I guess.). To make it so that I can't vote twice by figuring out what the token looks like means that the tokens need to be marked, so that the server can see if they've already been used. Even if in the database of registered voters there is no indication of what token they received, I'll bet it would be possible to look at database transaction logs for even just server transaction logs and figure it out.
Even assuming there's NO WAY server-side to figure out who voted for who, it still might be possible to figure out which token a certain person received and then via packet sniffing and cracked encryption figure out who voted for who. And in ellections, there's much more of a chance of people interested in doing that, so security through obscurity is even less of a solution here than it is anywhere else.
Solution? Each token is unique and can only be used once. But then...
TRANSACTION LOG FOR VOTE SERVER
User authenicated as John Doe, given token #AA-01431940-F294.
Token #AA-01431940-F294 was used to vote for canidate Jane Doe.
ATTEMPT TO USE TOKEN #AA-01431940-F294 AFTER INITAL VOTE - subsequent vote ignored.
Don't forget, even separating the two servers (one validates, one votes) doesn't work, because then all someone needs to do to track who voted for whom is look at the two logs. The only REAL solution would be not to keep logs, but I doubt a sysadmin would REALLY be willing not to keep detailed logs of attempts to access the system... (Hmm, look, we've got 3000 attempts to use token #1F-00031337-1337 from IP 221.142.391.257 - maybe it's time to block that IP...) Plus I'd expect that various law enforcement agents would want the logs to attempt to trace possible voter fraud. (One person attempting to validate as a different person, or people attempting to duplicate "tokens.")
Uh, that's more like selling the engine out of a car and not the radio. Selling the radio would be like selling off the computer speakers. A computer will not run unless it has (at least one) OS on it. If I decide to buy a brand new Dell computer and uninstall Windows from it, I have a nice, non-bootable piece of hardware. I'd have to install a new OS in it's place to use it. (Whether it be FreeBSD, MacOS, whatever.)
(A better car analogy might be something along the lines of the steering wheel, or the frame, or something, but there's few things in life which really make a good "hardware vs. software vs. OS" analogy...)
Besides, most of the people who would buy PC's without an OS are in the same segment of people likely to use Linux/*BSD/BeOS/etc. Of course, included in this group are the w4r3z d00ds likely to pirate Windows anyway. (BTW, I know some people who like warez who like Linux 'cause it's free as in beer, so there is overlapping, too.)