Was the vote sufficiently close that the margin of error in the vote tabulation system might have been enough to swing the vote?
If you can change the vote to whatever you want, this won't be a problem to the tamperer. Also, most states must be very close before a recount is required.
Were the results of the vote significantly different from exit polls or opinion polls?
Opinion / exit polls can have a fairly large margin of error. The vote result would have to be quite a bit different before anyone is likely to be suspicious. In other words, if Wyoming went to Nader, people would wonder what was up, but if you can swing the vote by a few percent in close states, you might be able to influence the election.
Was there evidence of fraud?
Electronic voting fraud, if successful, leaves no evidence.
Redundancy is necessary to prove that the votes were not tampered with. Let's say we had a voting proceedure something like this:
Each voter fills out a ballot and places it in a box. At the end of the day, an election official takes the box into a locked room, counts all the votes, then puts all the ballots in a shredder, then burns the results. That person comes out of the room and reports the results, which everyone is expected to accept as perfectly trustworthy.
This is what electronic voting is like, if there isn't a paper trail.
If everyone insists on having a manual recount afterwards...
And why shouldn't we do a count of the paper ballots in all cases? It's not that much work, given that someone already has to manually verify the identity of the voters and look them up on a list as they come in.
...given that the existence of the paper trail confirms that the electronic vote cannot be inherently relied upon in all reasonable circumstances...
The untrustworthiness of the electronic vote should be apparent with or without a paper trail. The paper trail lends credibility to the electronic vote (or at least it does if they match).
...why bother with the electronic count to begin with? What purpose is the electronic count serving?
Electronic interfaces can be friendlier and mor e accomodating to people with disabilities, and they allow a rapid (and accurate to the extent that it hasn't been tampered with) count.
The reason it is called teleportation is that quantum effects are used to make a particle disappear from point A and reappear at point B (a suitable time later) without crossing the intervening space.
Hmm, that sounds like a nice way of solving that pesky last mile problem without having to deal with the local ILEC, cable right-of-way, radio interference, or FCC licensing. From the wikipedia article, it doesn't sound like anyone's tried this over large distances though. Are there any issues which prevent quantum teleportation from being practical as a communications channel (besides not having having been festooned with acronyms by the IEEE)?
The main problem I have isn't with the method per se, but with its supporters.
The problem is that it isn't a single method. There are many variations of the condorcet method depending on what ambiguity resolution method you use, each with their own benefits and problems. Those who advocate Condorcet tend to do so without specifying which version they're talking about. I can't compare my favorite election method with Condorcet if I don't know what I'm comparing against.
the generalized concensus was that this guy was dead
Do we know how long ago it was recorded? If he mentions Kerry, I suppose it's probably not that old. Are there any other clues, such as mention of recent events?
Does anyone have a complete transcript? I'm kind of curious how skewed his version of reality is.
Wikis are good for permanent content (documentation, notes, etc..). News / message board sites like slashdot are good for time-dependant information.
Slashdot is fun to read because it changes every day. Wikis are useful because what I write will probably still be easy to find after months or years, if it's sufficiently interesting. Hardly anyone ever reads slashdot posts more than a week old. That's mostly because slashdot's content is sorted by time. There's no practical way to search for the most interesting posts ever made about a particular topic.
If you could create a system indexed by both time and content, and could provide separate forums for fact and opinion, you'd have quite a useful CMS. Imagine mediawiki with a slashdot-style threaded comment system instead of a simple wiki-style "discussion" page. Imagine if it could restrict users from editing the most controversial articles to those with excellent karma. Imagine threaded discussions that go on for years instead of days. (And how would you manage a discussion with tens or hundreds of thousantds of comments? What sort of content filtering interface would you need?)
The disadvantage is that another rewrite might leave you stranded with a difficult upgrade path.
I find it somewhat reasurring that MediaWiki is used to run Wikipedia. Since they already have a huge amount of preexisting content, it's in their best interests to make migration from one version to the next as easy as possible.
Of the wikis I've used, I like mediawiki the best in terms of simple interface (most CMSs have a cluttered interface that bombards the user with way too many buttons). Setting it up can be painful, though (it's easy when everything is working like it should, but I've tried a few versions that just refused to run, and I have no idea why).
It would be nice if it had web-based configuration, but I can live without it. I do wish it had a web interface to read/write permissions for anonymous users/logged in users/admin users for each page. I believe it does have some sort of security/permissions system, but I haven't figured out how to use it.
I also wish the "discussion" section for each article was more constrained - like a slashdot-style threaded comment system, preferably with a reputation system, but I don't expect them to implement that overnight.
...10 gigabit per second per user... Guaranteed switched bandwidth? But you have to hit some kind of limit at some point, right? I need more data...
By "user" they mean "institution connected to our network" not "individual person". As the previous reply said, they're limited to 40 wdm channels, one per user. To put this in perspective (from wikipedia):
The first WDM systems combined two signals and appeared around 1985. Modern systems can handle up to 160 signals and can expand a basic 10 Gbit/s fibre system to a theoretical total capacity of over 1.6 Tbit/s over a single fiber pair.
Anyone know what the shannon limit for single mode fiber is?
A 2mbit link could saturate a T1 without breaking a sweat. It would not take many rooted boxes to take out even an oc3
But, it will take 2.5 times more boxes to take out an oc3 if they have a 2 mbps cap instead of 5.
Um 2 mbits is plenty of bandwidth for video conferenceng. What are you dplaning on doing sending HDTV over it?
Sorry for the confusion, I was speaking generically about caps in general: a 768k/128k connection like I used to have wouldn't be that great for videoconferencing, but 2mbps would be more than reasonable. But, now that you mention it, hdtv (or at least near-dvd-quality) videoconferencing doesn't sound too bad (most DVDs have a bitrate of about 5-8 mbps, and that's with mediocre (compared to what a typical monitor can display) resolution and 24 fps, and non-realtime compression).
thought the whole idea of bit-torrent was to spread the load. Seems like 2mbit would be plenty for that use.
Most people pull stuff down more than they push stuff up, so download speed affects user's perception of the quality of the service more than upload.
Limiting upload puts a cap on the amount of traffic a single rooted box can generate, when participating in DOS and DDOS attacks.
Most people only spend a small fraction of the time downloading stuff, so the connection stays idle 99% of the time or so. Those who run file/web/p2p servers, though, can utilize their link more fully because their computers are "used" by a potentially large number of users on the rest of the Internet. If, for example, I hosted fedora ISOs from a web server on my home cable modem account, my upload bandwidth could easily dwarf what I could possibly download by surfing the web 24/7, even with my upload capped at about 10% of download.
The upload cap provides a disincentive from running potentially bandwidth hungry applications like videoconferencing, which require high throughput in both directions.
The upload cap provides a disincentive for people to try to use multicast trees and bit-torrent-like applications, by which a user can generate a disproportionally large quantity of traffic from a single connection, by utilizing other user's idle connections.
I tried fc2 on an old 64 meg PII awhile ago. I had to do a few tricks just to get it to install. Logging into X with gnome and nautilus running was about at the limits of its useable capabilities. I could run mozilla, but it was very, very slow.
This was a machine that I used to use every day just a couple of years ago, and it was adequate (but a little slow) then.
Who knows what the price point would be if they had have used Embedded Linux, firefox and OO instead.
On my fedora core 2 box, mozilla (with many windows open) is using over 250 megs (if I include non-resident pages its more like 600), and firing up oowriter with a blank document consumes almost 50 megs. I don't know what the memory requirements of firefox are, but is 128 megs a realistic ammount of ram, and can firefox and OO be stipped down to run on a low end computer?
I suspect this may be a case of not knowing your mrkets. In less industrialized markets, copper phone lines are rare.
I suppose that depends on where you are. I stayed in Bandung, Indonesia once for a few weeks. Electrity was unreliable, and the city water system was a maze of pvc pipe and garden hoses. I didn't see many computers, but a lot of people had TVs. There were phone lines there, and dialup internet access (I wish I'd saved the output from a traceroute I did back to the States). A moderately expensive meal was about $1. A hotel room was about $5-$10 a night. This is probably the sort of market AMD is targeting: places where basic infrastructure is present (if not reliable), but a midrange computer is still astronomically expensive for the average person.
If I was buying it, I'd rather have a nic, but the indended market is less industrialized countries, where broadband probably isn't widely available. In that context, a modem makes sense. Hopefully, it'll have a free pci slot or something.
What we need to do, IMHO, is to stop doing backoff after a connection has been established for a few thousand packets.
I don't think not backing off at all is a good idea. Your connection could change routes at any time, and you can't assume you'll be notified. On the other hand, it may be reasonable to only back off a little (say 10%) for each lost packet, instead of half. Or maybe subtract a large constant number (like 50 or 100 packets) from the window for each dropped packet, if the window size is huge.
Another option would be to drastically reduce the window only when the ration of dropped vs. successful packets exceeds some threshold (like 1 or 2%) - though this wouldn't play nice with routers that implement random early drop (RED).
I agree with you that TCP doesn't work great in all situations, and it is sometimes overly sensitive to packet drops, especially on connections with enormous bandwidth-delay products. It's also important, though, to realize the tradeoffs that TCP makes - over-agressive congestion control could potentially cause a lot more problems than overly-conservative bandwidth utilization.
-jim
(I'm sure there's hundreds of research papers out there on this topic if I cared to look)
TCP's bandwidth usage is dependent on the latency of the link. This is due to the fact that sliding windows (the number of packets that are allowed to be out on the network) have a limited size.
TCP's window size is not a fixed size. There is a maximum, but most "normal" connections are well below it. (Links with very high bandwidth and high delay may reach this limit, which can be increased somewhat with the right TCP extensions.) TCP's window size, in fact, regulates its bandwidth consumption.
Another problem with TCP is slow start and rapid back-off. IIRC, a TCP connection linearly increases its bandwidth, but will halve it immediately when a packet is lost.
That's not a bug, it's a feature! AIMD (additive increase, multiplicative decrease) is used because it's been found to work for most people. The multiplicative decrease part may seem drastic (cutting the window in half whenever a packet is lost), but it does do a good job of preventing severe packet loss due to congestion.
When multiple connections are sharing a link, TCP ends up favoring the lower latency connection. This happens because when a packet gets lost (often this is because a link is not fast enough to handle the data being sent over it), the corresponding sender fails to receive an acknowledgement, and reduces it's window size in half. Window size is incremented by some small value each time a full window of data is acknowledged. The window size of low latency connections grows much more quickly than high latency connections, so the low latency connection will have a larger window most of the time.
This is a known limitation of TCP, but I'm naturally suspicious of anyone who claims to have "fixed it" without offering specific details. Squeezing a little more bandwidth out of TCP is far less important than preventing the Internet from becoming unusable whenever a link becomes congested.
What would be wrong with that? Seriously? If the uneducated part of the general public left, this would only be beneficial to the rest of us.
Why do you think that's the demographic that would leave? Besides, it's not the general public that's the problem, it's a relatively small number of spammers and crackers. Most of the general public are just consumers. They neither contribute to nor detract from most internet services they use (such as the web), other than consuming bandwidth. The average "general public" internet user is a nice guy who doesn't write spam or create worms. They may unintentionally propogate worms through bad security, but we'll always have those kinds of people on the Internet, and we need to figure out how to deal with it instead of hoping someday they'll all go away.
no one cares/knows what Ogg is except the nerd elite
Exactly the group of people who are likely to go out and buy an ipod in the first place...
I know I'm in the minority, but most of my music is in ogg format these days. Most people couldn't care less about why I use ogg*, but I'm sure there's a sizeable minority that agrees with me.
There's really not much reason to not support ogg unless
the hardware isn't fast enough to decode it in real time,
the manufacturers are trying to push their own formats to gain leverage in a standards war, or
the manufacturers are concerned that someone will produce a "submarine patent" covering some aspect of ogg vorbis and demand royalties.
-jim
*I don't like the idea of patent-weilding middle men dictating what I can and can't do with my own bits**. Mp3 licensing, as far as I understand, is farily inexpensive, but it's still a big hurdle for open source developers. A standard Fedora install comes with ogg support, but no mp3 support.
**Not to be confused with the copyright-weilding middle men at the RIAA, who would like to control access to intellectual property through a combination of legal means and proprietary formats. Beware of middle men who don't add value to a transaction!
Re:Too rainy for Linus' convertible in Portland?
on
Linus Interviewed
·
· Score: 1
My question now is for you guys living in or knowing about Portland, will it too rainy for Linus' convertible in Portland?
It's been unusually dry so far this fall, but I don't expect that'll last too much longer. He'll probably have to leave the top up all winter and most of the spring and what's left of the fall. In the summer, it's actually pretty dry here.
My personal definition of libertarianism excludes actions that harm others, except by mutual consent.
I believe that is Ayn Rand's philosophy as well. Someone else summed it up as "Your right to swing your fist ends at my nose." I think people object to Ayn Rand for other reasons - such as her "altruism is the root of all evil" notions. Many of her arguments for refuting communism were quite good (and she was an excellent writer), but I'm not sure I'd want to live in her ideal society.
I'd be interested to know specifically how some libertarians distinguish their ideology from objectivism.
Q. Why did you choose to live in Portland, and what's your impression so far of the Northwest? One person told me you moved there because it looks like Finland. Is that true?
A. Well, the Northwest is certainly more like Finland in the sense that California is not like Finland.
But, no, I don't think that was the reason. Although part of it was definitely that we thought that Portland was more "livable," being smaller and less busy than Silicon Valley. Whether that is because I grew up in Finland, I don't know.
And being from Finland, the horror tales of constant rain didn't scare me as much as they do the native Californians.
I (as a Portlander) for one welcome our new Finish overlord.
Markets, as described by Akerlof, are neither complex nor competative.
Unless competitive means something different to you, Akerlof's used-car market example is definitely competitive. He shows how competing used car salesmen, acting in their own self interest, can destroy the market for used cars, in the sense that it becomes unprofitable to sell high quality used cars. His example is simple, but you'd have trouble proving that similar effects can avoided in more complex markets merely by virtue of their complexity.
Maybe I should describe his scenario so there's no confusion. I found some good explanations on the web once, but don't remember where. Alas, I haven't been able to find the original paper on the web. This is the scenario (numbers are probably different in the actual paper):
Used car salesmen sell two kinds of cars. Lemons have a value of $200 to the seller (that's what it cost the buyer to acquire it in the first place) and $300 to the buyer, while cherries have a value of $2000 to the seller and $3000 to the buyer. Initially, fifty percent of the used cars on the market are cherries, the other half are lemons.
If all parties could classify used cars as either cherry or lemmon, all transactions would be satisfactory (since the cars are worth more to the buyer than the seller).
Unfortunately, customers cannot determine by simple inspection whether a car is a cherry or a lemon. The seller, however, knows which it is. (Hence is the asymmetry.)
When a customer buys a car, given a fifty percent chance of buying a lemon with value $300 and a cherry with value $3000, the customer will not pay more than $1650 (the customer is assumed to have no aversion to risk).
Consequently, the seller loses money ($350) on every cherry he sells, and makes money on every lemon. In such a market, all sellers acting in their own self-interest will stop selling cherries and only sell lemons.
Customers like Kroger or Walmart. Customers that, if ingredients become a concern, can and will be able to test your product ingredients.
First, that would require a lot more work than just reading a label. Second, Kroger and Walmart aren't going to go to that kind of effort to verify the quality of food from every small supplier. Consequently, they will only carry foods from the biggest suppliers (maybe they do anyways, but that's a separate topic), reducing product choice. Third, they won't be motivated to go to that kind of effort unless the consumer (or some consumer group) is likely to do the same.
Regulated markets only benefit regulators.
That's not always true. Markets only regulate themselves in an efficient manner when competition exists and buyer's demand is not based on false (or non-existent) information. Government is usually less accountable than a well functioning market, but a democratically elected govenment is usually more accountable than a poorly functioning market.
Slashdot Mirror
If you can change the vote to whatever you want, this won't be a problem to the tamperer. Also, most states must be very close before a recount is required.
Opinion / exit polls can have a fairly large margin of error. The vote result would have to be quite a bit different before anyone is likely to be suspicious. In other words, if Wyoming went to Nader, people would wonder what was up, but if you can swing the vote by a few percent in close states, you might be able to influence the election.
Electronic voting fraud, if successful, leaves no evidence.
-jim
Redundancy is necessary to prove that the votes were not tampered with. Let's say we had a voting proceedure something like this:
Each voter fills out a ballot and places it in a box. At the end of the day, an election official takes the box into a locked room, counts all the votes, then puts all the ballots in a shredder, then burns the results. That person comes out of the room and reports the results, which everyone is expected to accept as perfectly trustworthy.
This is what electronic voting is like, if there isn't a paper trail.
And why shouldn't we do a count of the paper ballots in all cases? It's not that much work, given that someone already has to manually verify the identity of the voters and look them up on a list as they come in.
The untrustworthiness of the electronic vote should be apparent with or without a paper trail. The paper trail lends credibility to the electronic vote (or at least it does if they match).
Electronic interfaces can be friendlier and mor e accomodating to people with disabilities, and they allow a rapid (and accurate to the extent that it hasn't been tampered with) count.
-jim
Hmm, that sounds like a nice way of solving that pesky last mile problem without having to deal with the local ILEC, cable right-of-way, radio interference, or FCC licensing. From the wikipedia article, it doesn't sound like anyone's tried this over large distances though. Are there any issues which prevent quantum teleportation from being practical as a communications channel (besides not having having been festooned with acronyms by the IEEE)?
-jim
The main problem I have isn't with the method per se, but with its supporters.
The problem is that it isn't a single method. There are many variations of the condorcet method depending on what ambiguity resolution method you use, each with their own benefits and problems. Those who advocate Condorcet tend to do so without specifying which version they're talking about. I can't compare my favorite election method with Condorcet if I don't know what I'm comparing against.
-jim
It's news to me, but apparently Oreilly has an article about it. Alas, no pictures.
-jim
Do we know how long ago it was recorded? If he mentions Kerry, I suppose it's probably not that old. Are there any other clues, such as mention of recent events?
Does anyone have a complete transcript? I'm kind of curious how skewed his version of reality is.
-jim
Wikis are good for permanent content (documentation, notes, etc..). News / message board sites like slashdot are good for time-dependant information.
Slashdot is fun to read because it changes every day. Wikis are useful because what I write will probably still be easy to find after months or years, if it's sufficiently interesting. Hardly anyone ever reads slashdot posts more than a week old. That's mostly because slashdot's content is sorted by time. There's no practical way to search for the most interesting posts ever made about a particular topic.
If you could create a system indexed by both time and content, and could provide separate forums for fact and opinion, you'd have quite a useful CMS. Imagine mediawiki with a slashdot-style threaded comment system instead of a simple wiki-style "discussion" page. Imagine if it could restrict users from editing the most controversial articles to those with excellent karma. Imagine threaded discussions that go on for years instead of days. (And how would you manage a discussion with tens or hundreds of thousantds of comments? What sort of content filtering interface would you need?)
-jim
I find it somewhat reasurring that MediaWiki is used to run Wikipedia. Since they already have a huge amount of preexisting content, it's in their best interests to make migration from one version to the next as easy as possible.
Of the wikis I've used, I like mediawiki the best in terms of simple interface (most CMSs have a cluttered interface that bombards the user with way too many buttons). Setting it up can be painful, though (it's easy when everything is working like it should, but I've tried a few versions that just refused to run, and I have no idea why).
It would be nice if it had web-based configuration, but I can live without it. I do wish it had a web interface to read/write permissions for anonymous users/logged in users/admin users for each page. I believe it does have some sort of security/permissions system, but I haven't figured out how to use it.
I also wish the "discussion" section for each article was more constrained - like a slashdot-style threaded comment system, preferably with a reputation system, but I don't expect them to implement that overnight.
-jim
By "user" they mean "institution connected to our network" not "individual person". As the previous reply said, they're limited to 40 wdm channels, one per user. To put this in perspective (from wikipedia):
Anyone know what the shannon limit for single mode fiber is?
-jim
But, it will take 2.5 times more boxes to take out an oc3 if they have a 2 mbps cap instead of 5.
Sorry for the confusion, I was speaking generically about caps in general: a 768k/128k connection like I used to have wouldn't be that great for videoconferencing, but 2mbps would be more than reasonable. But, now that you mention it, hdtv (or at least near-dvd-quality) videoconferencing doesn't sound too bad (most DVDs have a bitrate of about 5-8 mbps, and that's with mediocre (compared to what a typical monitor can display) resolution and 24 fps, and non-realtime compression).
More is better.
-jim
There are quite a few reasons to limit upload:
-jim
I tried fc2 on an old 64 meg PII awhile ago. I had to do a few tricks just to get it to install. Logging into X with gnome and nautilus running was about at the limits of its useable capabilities. I could run mozilla, but it was very, very slow.
This was a machine that I used to use every day just a couple of years ago, and it was adequate (but a little slow) then.
-jim
On my fedora core 2 box, mozilla (with many windows open) is using over 250 megs (if I include non-resident pages its more like 600), and firing up oowriter with a blank document consumes almost 50 megs. I don't know what the memory requirements of firefox are, but is 128 megs a realistic ammount of ram, and can firefox and OO be stipped down to run on a low end computer?
-jim
I suppose that depends on where you are. I stayed in Bandung, Indonesia once for a few weeks. Electrity was unreliable, and the city water system was a maze of pvc pipe and garden hoses. I didn't see many computers, but a lot of people had TVs. There were phone lines there, and dialup internet access (I wish I'd saved the output from a traceroute I did back to the States). A moderately expensive meal was about $1. A hotel room was about $5-$10 a night. This is probably the sort of market AMD is targeting: places where basic infrastructure is present (if not reliable), but a midrange computer is still astronomically expensive for the average person.
-jim
If I was buying it, I'd rather have a nic, but the indended market is less industrialized countries, where broadband probably isn't widely available. In that context, a modem makes sense. Hopefully, it'll have a free pci slot or something.
-jim
I don't think not backing off at all is a good idea. Your connection could change routes at any time, and you can't assume you'll be notified. On the other hand, it may be reasonable to only back off a little (say 10%) for each lost packet, instead of half. Or maybe subtract a large constant number (like 50 or 100 packets) from the window for each dropped packet, if the window size is huge.
Another option would be to drastically reduce the window only when the ration of dropped vs. successful packets exceeds some threshold (like 1 or 2%) - though this wouldn't play nice with routers that implement random early drop (RED).
I agree with you that TCP doesn't work great in all situations, and it is sometimes overly sensitive to packet drops, especially on connections with enormous bandwidth-delay products. It's also important, though, to realize the tradeoffs that TCP makes - over-agressive congestion control could potentially cause a lot more problems than overly-conservative bandwidth utilization.
-jim
(I'm sure there's hundreds of research papers out there on this topic if I cared to look)
I summon... Rent-a-Zilla
-jim
TCP's window size is not a fixed size. There is a maximum, but most "normal" connections are well below it. (Links with very high bandwidth and high delay may reach this limit, which can be increased somewhat with the right TCP extensions.) TCP's window size, in fact, regulates its bandwidth consumption.
That's not a bug, it's a feature! AIMD (additive increase, multiplicative decrease) is used because it's been found to work for most people. The multiplicative decrease part may seem drastic (cutting the window in half whenever a packet is lost), but it does do a good job of preventing severe packet loss due to congestion.
When multiple connections are sharing a link, TCP ends up favoring the lower latency connection. This happens because when a packet gets lost (often this is because a link is not fast enough to handle the data being sent over it), the corresponding sender fails to receive an acknowledgement, and reduces it's window size in half. Window size is incremented by some small value each time a full window of data is acknowledged. The window size of low latency connections grows much more quickly than high latency connections, so the low latency connection will have a larger window most of the time.
This is a known limitation of TCP, but I'm naturally suspicious of anyone who claims to have "fixed it" without offering specific details. Squeezing a little more bandwidth out of TCP is far less important than preventing the Internet from becoming unusable whenever a link becomes congested.
-jim
Why do you think that's the demographic that would leave? Besides, it's not the general public that's the problem, it's a relatively small number of spammers and crackers. Most of the general public are just consumers. They neither contribute to nor detract from most internet services they use (such as the web), other than consuming bandwidth. The average "general public" internet user is a nice guy who doesn't write spam or create worms. They may unintentionally propogate worms through bad security, but we'll always have those kinds of people on the Internet, and we need to figure out how to deal with it instead of hoping someday they'll all go away.
-jim
Seattle Post-Intelligencer
The Guardian
-jim
Exactly the group of people who are likely to go out and buy an ipod in the first place...
I know I'm in the minority, but most of my music is in ogg format these days. Most people couldn't care less about why I use ogg*, but I'm sure there's a sizeable minority that agrees with me.
There's really not much reason to not support ogg unless
-jim
*I don't like the idea of patent-weilding middle men dictating what I can and can't do with my own bits**. Mp3 licensing, as far as I understand, is farily inexpensive, but it's still a big hurdle for open source developers. A standard Fedora install comes with ogg support, but no mp3 support.
**Not to be confused with the copyright-weilding middle men at the RIAA, who would like to control access to intellectual property through a combination of legal means and proprietary formats. Beware of middle men who don't add value to a transaction!
It's been unusually dry so far this fall, but I don't expect that'll last too much longer. He'll probably have to leave the top up all winter and most of the spring and what's left of the fall. In the summer, it's actually pretty dry here.
-jim
I believe that is Ayn Rand's philosophy as well. Someone else summed it up as "Your right to swing your fist ends at my nose." I think people object to Ayn Rand for other reasons - such as her "altruism is the root of all evil" notions. Many of her arguments for refuting communism were quite good (and she was an excellent writer), but I'm not sure I'd want to live in her ideal society.
I'd be interested to know specifically how some libertarians distinguish their ideology from objectivism.
-jim
I (as a Portlander) for one welcome our new Finish overlord.
-jim
Unless competitive means something different to you, Akerlof's used-car market example is definitely competitive. He shows how competing used car salesmen, acting in their own self interest, can destroy the market for used cars, in the sense that it becomes unprofitable to sell high quality used cars. His example is simple, but you'd have trouble proving that similar effects can avoided in more complex markets merely by virtue of their complexity.
Maybe I should describe his scenario so there's no confusion. I found some good explanations on the web once, but don't remember where. Alas, I haven't been able to find the original paper on the web. This is the scenario (numbers are probably different in the actual paper):
Used car salesmen sell two kinds of cars. Lemons have a value of $200 to the seller (that's what it cost the buyer to acquire it in the first place) and $300 to the buyer, while cherries have a value of $2000 to the seller and $3000 to the buyer. Initially, fifty percent of the used cars on the market are cherries, the other half are lemons.
If all parties could classify used cars as either cherry or lemmon, all transactions would be satisfactory (since the cars are worth more to the buyer than the seller).
Unfortunately, customers cannot determine by simple inspection whether a car is a cherry or a lemon. The seller, however, knows which it is. (Hence is the asymmetry.)
When a customer buys a car, given a fifty percent chance of buying a lemon with value $300 and a cherry with value $3000, the customer will not pay more than $1650 (the customer is assumed to have no aversion to risk).
Consequently, the seller loses money ($350) on every cherry he sells, and makes money on every lemon. In such a market, all sellers acting in their own self-interest will stop selling cherries and only sell lemons.
First, that would require a lot more work than just reading a label. Second, Kroger and Walmart aren't going to go to that kind of effort to verify the quality of food from every small supplier. Consequently, they will only carry foods from the biggest suppliers (maybe they do anyways, but that's a separate topic), reducing product choice. Third, they won't be motivated to go to that kind of effort unless the consumer (or some consumer group) is likely to do the same.
That's not always true. Markets only regulate themselves in an efficient manner when competition exists and buyer's demand is not based on false (or non-existent) information. Government is usually less accountable than a well functioning market, but a democratically elected govenment is usually more accountable than a poorly functioning market.
-jim