You know, it seems for every time something beneficial is found in a food, someone else discovers something else in it that causes cancer. And vice versa.
So you know what I recommend? Use some common sense. Eat a reasonable, balanced diet. Enjoy your favorite foods, but don't overindulge - everything in moderation.
You gotta go sometime, so you might as well eat what you enjoy eating. It appears the "experts" disagree and can't really tells us for sure one way or the other what's really good or bad for us, anyway. By the time you get done going through your tally sheet of what you can and can't have, what you should and shouldn't consume, you've probably added enough stress to your day that you'll keel over from a heart attack or develop an ulcer!
Absolutely. MS could probably still make money off this version if they just Carbonized it. Doing so might possibly bring it up to date with many modern expectations, such as Unicode support, too, though I am not a Mac programmer so haven't a clue what the Carbon API does for you.
Hey, as soon as I can get boost-libs to build, I'll have A1 running on NetBSD/mac68k! There was one conflict specific to 68k IIRC (a variable name collision), but once I fixed that I still had a bunch of issues that seemed more generic. I don't know if I have the skill or the time to fix them all.
Regarding Irix, I know very little about the platform. I'm not sure I even have the GNU toolchain working at present. I'll leave it up to the Nekochan guys to figure out.
Unfortunately, the networking doesn't seem to work on the Irix build. I have an O2 I was going to try to use for LAN parties, but no love. I suppose someone could play solo on it as a warmup or something.
This is a very good point. If the apps are using Qt, then the GUI should be Qt also, when low-end hardware is a consideration.
Since Edubuntu seems to be just a set of packages and themes installed on Ubuntu, I was going to suggest installing a more lightweight desktop, also. Xfce or EDE will use fewer resources than either KDE or Gnome will. (I know there is a "xubuntu-desktop" package for Xfce, but I don't know about EDE support in Ubuntu.) But if those edutainment apps are the ones that will primarily be used, it may still make sense from the resource usage standpoint to go with KDE.
I am not a "gamer" - I don't buy lots of games, I don't overclock, I don't need the latest and greatest video card every two months. That said, here are the games I've enjoyed over the past 20 years.
Apple// games
LodeRunner
Conan
Eamon
Classic Macintosh games
Maelstrom
Warlords
SimCity 2000
Marathon Trilogy
MacOS X
Aleph One - yes, thanks to open source, I am still playing many third-party Marathon scenarios
There are a few other time wasters that I did enjoy, but these are the ones that kept me coming back over and over.
CSS3 compliant browser: a[href$=".pdf"]:after {content: "[PDF]"; font-size:smaller} in your user style sheet. Modify as needed for other types of "annoying" links.
There are several very simple things I'd like to see added to del.icio.us, which I'd do myself if I could: sorting (other than by date added), and pre- and post-title "notation" text (as opposed to a single comment field). I was going to write my own bookmark manager when I discovered del.icio.us, which does everything I wanted (and more neatly) except for these two very simple things. I've looked at the OSS alternatives, but they don't seem as advanced nor very active. FYI, there's a project at SF called "scrumptious" too, which seems about the same thing.
The problem is the voting system. Plurality voting leads to Duverger's Law, i.e. a two-party system. Implement Condorcet voting and you end the Duopoly. The problem is, the Repubs and Dems aren't going to be interested in changing the system, because they realize that it contributes to keeping them in power. Therefore, we have to get at least a few third-party voices in using the current system in order to get the issue in the eye of the public. Someone's got to be the first to agitate for change, or it will never happen.
On the issue of voting reform, don't be misled by the smokescreen that is Instant Runoff. Read up on different voting systems. You'll find that IRV is even worse than plurality voting. It's simple, yes, but it has grievous errors that yield bad results. Worst of all, IMO, is it gives the impression that minor parties are getting a fair shake - so they'd stop advocating for fairer voting systems. A few nations have implemented IRV, but they are still essentially two-party governments. I find it hard to believe that, given a real choice, people would still align themselves into a mere two camps when there are obviously a multitude of issues with a multitude of opinions to choose from. If the math weren't enough to prove IRV is bad, that should be.
On the issue of "throwing your vote away" - that's what you do when you don't vote for what you really believe. If you don't vote your conscience, who's really losing out? I won't sacrifice that just to have the lesser of two evils in office for a couple years. Stand behind your ideals.
Duverger's Law (two-party system) is caused by plurality voting. We could end the Duopoly if we just instituted Condorcet voting. Politics is not one-dimensional, yet so many people think of everything being neatly divided into two camps. I just don't understand that. We need a voting system that lets us honestly choose who we really want, rather than voting strategically based on "winnability" considerations. Condorcet voting would force politicians to have a clear, distinct message, and then stick to that platform.
Privatized sidewalks means that you wouldn't have the sidewalks if you hadn't built them yourself, because you wouldn't have been taxed for them. And then since they are yours, and on your property, you could probably charge people for passage.
Plurality voting leads to Duverger's Law - a two-party system. You're right, having a Duopoly in power means not really having any choices at all. Both drift to the middle to gain voters.
What we need is voting reform (not campaign finance reform). We need Condorcet voting. With the electorate freed up to make choices based on platforms and ideas rather than "winnability" we'll finally see honest competition and debate of issues. We need more freedom (voting reform) not less freedom (campaign finance reform).
Since the Duopoly is not going to change the system that benefits them, it's important to start voting for third parties now. Having even a few in office will allow the issue to be raised and put in the public eye. The Libertarian and Green parties endorse IRV, which is a big mistake as it's about the only well-known voting system that is actually worse than we currently have. But at least they address the issue of voting reform.
Most people don't think of the mundane issues like the process of voting, but it's very important. It has shaped the political landscape in the USA for over a hundred years.
The problem isn't the money. I should be free to spend my money to support whatever cause I want - free expression.
The problem is the voting system. Plurality voting causes two-party systems, and a choice between two parties is not much of a choice at all.
Two parties means both drift to the center and have no meaningful differences. It's a Duopoly rule. Politics is not a one-dimensional, left/right spectrum, so why do we think two parties can adequately express all points of view? Be an advocate for Condorcet Voting.
I believe that with a voting system in place that rewards sincerity rather than encouraging "strategic thinking", politicians would be obliged to have a clear, distinct message and then stick to it.
Think about it - you make all candidates free to run a political race, rather than crippling those that aren't "major parties". You make all people free to donate whatever they want to getting their message out, which they have the right to do, and since there are more candidates to spread the money amongst you still don't have the huge warchests and media blitzes like you do now. The more freedom you have in the system, the better. Clamping down and making more regulations doesn't help.
Two parties means both drift to the center and have no meaningful differences. It's a Duopoly rule. Politics is not a one-dimensional, left/right spectrum, so why do we think two parties can adequately express all points of view? Be an advocate for Condorcet Voting.
Nations with a multi-party system often do not have a very good voting method behind it to make their selections with. For example, it's often been demonstrated that Instant Runoff Voting, though giving the appearance that third parties are given a fair shot, still leads to a two-party system. I find it extremely hard to believe that, with the multitude of political issues, most people can still fall into one of two camps. This leads me to conclude that IRV is flawed - and the math backs me up. Until a third party garners enough support to actually rival a major party, it will remain on the fringe. When it has that support, it suddenly unseats one of the major parties. It has to do with the whole concept of "elimination rounds" in any runoff system - until you have enough "primary" support to get to the final round, you don't really matter. Condorcet methods neatly solve this by realizing that your secondary-vs-tertiary preference is just as important as your primary-vs-secondary preference, and your primary-vs-tertiary preference, so it takes them all into account at the same simultaneously instead of sequentially.
I believe that with a voting system in place that rewards sincerity, politicians would be obliged to have a clear message and then stick to it.
Two parties means both drift to the center and have no meaningful differences. It's a Duopoly rule. Politics is not a one-dimensional, left/right spectrum, so why do we think two parties can adequately express all points of view? Be an advocate for Condorcet Voting.
You know that many (if not all) sections have been slashboxes for a long, long time, right? That's how I used to keep up on BSD, Science, and YRO stuff, anyway. Just activate those slashboxes and there you go.
First off, let me say I welcome the increased customization of the home page. But I think it could be done in a different way that would be easier to extend in the future, including doing things like you suggest.
Currently there are two separate systems that determine whether or not you see something. There's the new full/abbreviated/none system that uses the "home page" flag set by the editors in conjunction with the user's sectional preferences (which used to be a simpler yes/no by section), and there's yes/no by editor.
My idea uses a point-based scheme to decide whether or not a story appears on the home page. It incorporates all aspects of the story into one system. In the user prefs, you promote/demote stories by one point based on section, topic, editor posting it, and whatever other characteristics of the story exposed for this. NumPosts over X where you decide what X is, or Y% of comments at +4 or better where you decide what Y is, or submitter is a friend... these are all possibilities. Then the editor posting it gives it a +1 if he marks it "home page" news. Tally up the points, and if it's positive it shows up full-text. If it's zero, it shows up minimized. If negative, it won't show up on your home page (but will in its section of course).
For example, a JonKatz story might have your personal -1, and the +1 for "front page" yields a 0 - you see only the minimized version. But say you really like stories in the BSD section (+1), and Jon happens to be writing about BSD. Then you would see the full story. Assuming no user customization, the default works just like it does today: stories at +1 (home page stories) show up full-text, all others are at 0 and show up abbreviated.
I'd also like to see support for regionalization added to slash. I'd bet there are lots of slash sites that would like to post news that's specific to a certain geographic locale, but don't want to make it front page news for everyone. With regionalization and a customization system like this, Slashdot could mention local conferences (which would be a useful service to those in that area) without getting flamed by the 99% that don't live there.
I really like that fact that I can eliminate slashboxes from the right side of my page now. My slashbox column was taller than the story column. Now that it is easy to follow all stories I'm interested in at once, I get a leaner page.
Preference cycles simply reflect reality. I might prefer B to A because of his stance on education, C to B because of her opinions on foreign trade, and A to C because of his military background. It's possible, though extremely unlikely IMO, that a similar cycle will manifest itself in a collective body. The fact that the voting system reflects reality isn't a weakness; it simply goes to show that finding a concensus in a large group of people is a very difficult thing to do!
The basic definition of a Condorcet method (the winner is the guy who can beat anybody and everybody else head-to-head) is really pretty intuitive. (You don't need to start using words like "pairwise" and "matrix" to explain this!) So are the rules for disambiguating cycles. There are more than one, sure, so just pick one and go with it. I don't see how the public would be any more confused than they are with the Electoral College.
IRV doesn't even meet monotonicity for crying out loud...you can cause a candidate to fall back in the race by ranking him higher than his competitors! What good is a system like that?
With IRV, you can't sub-aggregate the votes by districts either...once you "throw away" the "runoff rounds" there's no way to sum all the districts together. (The summability criterion. Not on the chart, but another failure for IRV.) Condorcet, OTOH, gives you a very neat summary (A vs B, A vs C, A vs D, B vs C, etc) that can be tallied up at any district level. IRV would have you counting (potentially) 280 million votes at the US level (for example), Condorcet would have you combining 50 state totals.
IRV gives a nice illusion that your full preference is being recorded and makes a difference, but that's all it is — an illusion.
Say this with me: Instant Runoff is flawed, use Condorcet instead; Instant Runoff is flawed, use Condorcet instead; Instant Runoff is flawed, use Condorcet instead. IRV has serious problems which are easily found with a simple Google search. Condorcet is no harder to use than IRV, doesn't yield bizarre unintuitive results like IRV, and removes incentives to vote dishonestly.
With Opera already supported on three Unices (Linux, FreeBSD, Solaris) why isn't it available for more? Surely there couldn't be much more work involved than recompiling if things have been engineered correctly. Those of us using NetBSD, Irix, OpenBSD, and others would really like to use Opera too!
My "low end" box is a Mac SE/30 running NetBSD (of course!). It's about as low-end as you can get in the Mac world and still run Unix. (Yeah, some other models are also 16 MHz but a bit slower because of other design issues, and there is one 68020 machine.) It's the most expandable of the bottom-end Macs, which is why I chose it - mine has 80MB and a 9GB drive, and I've heard of larger.
Why do this? Because I can. It's a rock solid little box, and performs its tasks very well.
Slashdot Mirror
You know, it seems for every time something beneficial is found in a food, someone else discovers something else in it that causes cancer. And vice versa.
So you know what I recommend? Use some common sense. Eat a reasonable, balanced diet. Enjoy your favorite foods, but don't overindulge - everything in moderation.
You gotta go sometime, so you might as well eat what you enjoy eating. It appears the "experts" disagree and can't really tells us for sure one way or the other what's really good or bad for us, anyway. By the time you get done going through your tally sheet of what you can and can't have, what you should and shouldn't consume, you've probably added enough stress to your day that you'll keel over from a heart attack or develop an ulcer!
Absolutely. MS could probably still make money off this version if they just Carbonized it. Doing so might possibly bring it up to date with many modern expectations, such as Unicode support, too, though I am not a Mac programmer so haven't a clue what the Carbon API does for you.
Hey, as soon as I can get boost-libs to build, I'll have A1 running on NetBSD/mac68k! There was one conflict specific to 68k IIRC (a variable name collision), but once I fixed that I still had a bunch of issues that seemed more generic. I don't know if I have the skill or the time to fix them all.
Regarding Irix, I know very little about the platform. I'm not sure I even have the GNU toolchain working at present. I'll leave it up to the Nekochan guys to figure out.
Unfortunately, the networking doesn't seem to work on the Irix build. I have an O2 I was going to try to use for LAN parties, but no love. I suppose someone could play solo on it as a warmup or something.
This is a very good point. If the apps are using Qt, then the GUI should be Qt also, when low-end hardware is a consideration.
Since Edubuntu seems to be just a set of packages and themes installed on Ubuntu, I was going to suggest installing a more lightweight desktop, also. Xfce or EDE will use fewer resources than either KDE or Gnome will. (I know there is a "xubuntu-desktop" package for Xfce, but I don't know about EDE support in Ubuntu.) But if those edutainment apps are the ones that will primarily be used, it may still make sense from the resource usage standpoint to go with KDE.
I am not a "gamer" - I don't buy lots of games, I don't overclock, I don't need the latest and greatest video card every two months. That said, here are the games I've enjoyed over the past 20 years.
There are a few other time wasters that I did enjoy, but these are the ones that kept me coming back over and over.
Mozilla/Firefox: TargetAlert.
CSS3 compliant browser: a[href$=".pdf"]:after {content: "[PDF]"; font-size:smaller} in your user style sheet. Modify as needed for other types of "annoying" links.
There are several very simple things I'd like to see added to del.icio.us, which I'd do myself if I could: sorting (other than by date added), and pre- and post-title "notation" text (as opposed to a single comment field). I was going to write my own bookmark manager when I discovered del.icio.us, which does everything I wanted (and more neatly) except for these two very simple things. I've looked at the OSS alternatives, but they don't seem as advanced nor very active. FYI, there's a project at SF called "scrumptious" too, which seems about the same thing.
The problem is the voting system. Plurality voting leads to Duverger's Law, i.e. a two-party system. Implement Condorcet voting and you end the Duopoly. The problem is, the Repubs and Dems aren't going to be interested in changing the system, because they realize that it contributes to keeping them in power. Therefore, we have to get at least a few third-party voices in using the current system in order to get the issue in the eye of the public. Someone's got to be the first to agitate for change, or it will never happen.
On the issue of voting reform, don't be misled by the smokescreen that is Instant Runoff. Read up on different voting systems. You'll find that IRV is even worse than plurality voting. It's simple, yes, but it has grievous errors that yield bad results. Worst of all, IMO, is it gives the impression that minor parties are getting a fair shake - so they'd stop advocating for fairer voting systems. A few nations have implemented IRV, but they are still essentially two-party governments. I find it hard to believe that, given a real choice, people would still align themselves into a mere two camps when there are obviously a multitude of issues with a multitude of opinions to choose from. If the math weren't enough to prove IRV is bad, that should be.
On the issue of "throwing your vote away" - that's what you do when you don't vote for what you really believe. If you don't vote your conscience, who's really losing out? I won't sacrifice that just to have the lesser of two evils in office for a couple years. Stand behind your ideals.
Duverger's Law (two-party system) is caused by plurality voting. We could end the Duopoly if we just instituted Condorcet voting. Politics is not one-dimensional, yet so many people think of everything being neatly divided into two camps. I just don't understand that. We need a voting system that lets us honestly choose who we really want, rather than voting strategically based on "winnability" considerations. Condorcet voting would force politicians to have a clear, distinct message, and then stick to that platform.
Privatized sidewalks means that you wouldn't have the sidewalks if you hadn't built them yourself, because you wouldn't have been taxed for them. And then since they are yours, and on your property, you could probably charge people for passage.
Plurality voting leads to Duverger's Law - a two-party system. You're right, having a Duopoly in power means not really having any choices at all. Both drift to the middle to gain voters.
What we need is voting reform (not campaign finance reform). We need Condorcet voting. With the electorate freed up to make choices based on platforms and ideas rather than "winnability" we'll finally see honest competition and debate of issues. We need more freedom (voting reform) not less freedom (campaign finance reform).
Since the Duopoly is not going to change the system that benefits them, it's important to start voting for third parties now. Having even a few in office will allow the issue to be raised and put in the public eye. The Libertarian and Green parties endorse IRV, which is a big mistake as it's about the only well-known voting system that is actually worse than we currently have. But at least they address the issue of voting reform.
Most people don't think of the mundane issues like the process of voting, but it's very important. It has shaped the political landscape in the USA for over a hundred years.
The problem isn't the money. I should be free to spend my money to support whatever cause I want - free expression.
The problem is the voting system. Plurality voting causes two-party systems, and a choice between two parties is not much of a choice at all.
Two parties means both drift to the center and have no meaningful differences. It's a Duopoly rule. Politics is not a one-dimensional, left/right spectrum, so why do we think two parties can adequately express all points of view? Be an advocate for Condorcet Voting.
I believe that with a voting system in place that rewards sincerity rather than encouraging "strategic thinking", politicians would be obliged to have a clear, distinct message and then stick to it.
Think about it - you make all candidates free to run a political race, rather than crippling those that aren't "major parties". You make all people free to donate whatever they want to getting their message out, which they have the right to do, and since there are more candidates to spread the money amongst you still don't have the huge warchests and media blitzes like you do now. The more freedom you have in the system, the better. Clamping down and making more regulations doesn't help.
Two parties means both drift to the center and have no meaningful differences. It's a Duopoly rule. Politics is not a one-dimensional, left/right spectrum, so why do we think two parties can adequately express all points of view? Be an advocate for Condorcet Voting.
Nations with a multi-party system often do not have a very good voting method behind it to make their selections with. For example, it's often been demonstrated that Instant Runoff Voting, though giving the appearance that third parties are given a fair shot, still leads to a two-party system. I find it extremely hard to believe that, with the multitude of political issues, most people can still fall into one of two camps. This leads me to conclude that IRV is flawed - and the math backs me up. Until a third party garners enough support to actually rival a major party, it will remain on the fringe. When it has that support, it suddenly unseats one of the major parties. It has to do with the whole concept of "elimination rounds" in any runoff system - until you have enough "primary" support to get to the final round, you don't really matter. Condorcet methods neatly solve this by realizing that your secondary-vs-tertiary preference is just as important as your primary-vs-secondary preference, and your primary-vs-tertiary preference, so it takes them all into account at the same simultaneously instead of sequentially.
I believe that with a voting system in place that rewards sincerity, politicians would be obliged to have a clear message and then stick to it.
Two parties means both drift to the center and have no meaningful differences. It's a Duopoly rule. Politics is not a one-dimensional, left/right spectrum, so why do we think two parties can adequately express all points of view? Be an advocate for Condorcet Voting.
I, too, would like to have all aspects of the story influence homepage customization. I'd turn off Anime, too.
You know that many (if not all) sections have been slashboxes for a long, long time, right? That's how I used to keep up on BSD, Science, and YRO stuff, anyway. Just activate those slashboxes and there you go.
I'd like to see all aspects of a story unified in the approach to homepage customization, too.
First off, let me say I welcome the increased customization of the home page. But I think it could be done in a different way that would be easier to extend in the future, including doing things like you suggest.
Currently there are two separate systems that determine whether or not you see something. There's the new full/abbreviated/none system that uses the "home page" flag set by the editors in conjunction with the user's sectional preferences (which used to be a simpler yes/no by section), and there's yes/no by editor.
My idea uses a point-based scheme to decide whether or not a story appears on the home page. It incorporates all aspects of the story into one system. In the user prefs, you promote/demote stories by one point based on section, topic, editor posting it, and whatever other characteristics of the story exposed for this. NumPosts over X where you decide what X is, or Y% of comments at +4 or better where you decide what Y is, or submitter is a friend ... these are all possibilities. Then the editor posting it gives it a +1 if he marks it "home page" news. Tally up the points, and if it's positive it shows up full-text. If it's zero, it shows up minimized. If negative, it won't show up on your home page (but will in its section of course).
For example, a JonKatz story might have your personal -1, and the +1 for "front page" yields a 0 - you see only the minimized version. But say you really like stories in the BSD section (+1), and Jon happens to be writing about BSD. Then you would see the full story. Assuming no user customization, the default works just like it does today: stories at +1 (home page stories) show up full-text, all others are at 0 and show up abbreviated.
I've mailed this to Taco and posted it as a feature request for slashcode several years ago.
I'd also like to see support for regionalization added to slash. I'd bet there are lots of slash sites that would like to post news that's specific to a certain geographic locale, but don't want to make it front page news for everyone. With regionalization and a customization system like this, Slashdot could mention local conferences (which would be a useful service to those in that area) without getting flamed by the 99% that don't live there.
I really like that fact that I can eliminate slashboxes from the right side of my page now. My slashbox column was taller than the story column. Now that it is easy to follow all stories I'm interested in at once, I get a leaner page.
Section-appropriate colors on the index page? Egads...it would be the Blinding Nightmarish Rainbow From Hell. Please, not that.
The sectional favicons is a nice idea though.
Preference cycles simply reflect reality. I might prefer B to A because of his stance on education, C to B because of her opinions on foreign trade, and A to C because of his military background. It's possible, though extremely unlikely IMO, that a similar cycle will manifest itself in a collective body. The fact that the voting system reflects reality isn't a weakness; it simply goes to show that finding a concensus in a large group of people is a very difficult thing to do!
The basic definition of a Condorcet method (the winner is the guy who can beat anybody and everybody else head-to-head) is really pretty intuitive. (You don't need to start using words like "pairwise" and "matrix" to explain this!) So are the rules for disambiguating cycles. There are more than one, sure, so just pick one and go with it. I don't see how the public would be any more confused than they are with the Electoral College.
IRV a huge improvement over plurality? Surely you jest! IRV fails even more voting criteria than plurality does! See the chart?
IRV doesn't even meet monotonicity for crying out loud...you can cause a candidate to fall back in the race by ranking him higher than his competitors! What good is a system like that?
With IRV, you can't sub-aggregate the votes by districts either...once you "throw away" the "runoff rounds" there's no way to sum all the districts together. (The summability criterion. Not on the chart, but another failure for IRV.) Condorcet, OTOH, gives you a very neat summary (A vs B, A vs C, A vs D, B vs C, etc) that can be tallied up at any district level. IRV would have you counting (potentially) 280 million votes at the US level (for example), Condorcet would have you combining 50 state totals.
IRV gives a nice illusion that your full preference is being recorded and makes a difference, but that's all it is — an illusion.
Say this with me: Instant Runoff is flawed, use Condorcet instead; Instant Runoff is flawed, use Condorcet instead; Instant Runoff is flawed, use Condorcet instead. IRV has serious problems which are easily found with a simple Google search. Condorcet is no harder to use than IRV, doesn't yield bizarre unintuitive results like IRV, and removes incentives to vote dishonestly.
With Opera already supported on three Unices (Linux, FreeBSD, Solaris) why isn't it available for more? Surely there couldn't be much more work involved than recompiling if things have been engineered correctly. Those of us using NetBSD, Irix, OpenBSD, and others would really like to use Opera too!
Grinding your NIC down, that's great.
My "low end" box is a Mac SE/30 running NetBSD (of course!). It's about as low-end as you can get in the Mac world and still run Unix. (Yeah, some other models are also 16 MHz but a bit slower because of other design issues, and there is one 68020 machine.) It's the most expandable of the bottom-end Macs, which is why I chose it - mine has 80MB and a 9GB drive, and I've heard of larger.
Why do this? Because I can. It's a rock solid little box, and performs its tasks very well.