Slashdot posts about mathematics are usually so far wrong that I don't even try to respond to them. It is really distressing to me (having a Ph.D in mathematics) to see how shallow the general level of mathematics understanding here is.
However in this case your comment is only slightly wrong and therefore I have some hope that my reply might be a useful contribution.
You are correct that mathematical proofs are based on axioms. However there is still a crucial difference between a mathematical proof and a scientific theory. A mathematical proof is an absolute certainty. Note that I am not claiming that the underlying axioms are certain. I am only claiming that the proof itself is certain.
To put it another way, mathematicians are never certain about their underlying axioms but they are absolutely certain that if those axioms hold then the result stated in the proof also holds. It's kind of like a building with indestructible walls but no foundation.
Scientific theory is a whole different kettle of fish. You cannot prove a scientific theory with absolute certainty. In fact it is not even clear to me how one can define certainty within the framework of the scientific method. You never have any guarantee in science that future observations will be consistent with past observations.
In science you can prove a theory in the sense of preponderance of the evidence. You can even sometimes prove a theory beyond all reasonable doubt. But there is no way to eliminate the unreasonable doubts. Any endeavour based on empirical observation suffers from the fundamental limitation that you can never be sure of the next observation.
Finally, regarding 1+1=2, the foundational proof of this fact using the standard propositional axioms of mathematics really does require 362 pages. You can see the 362nd page on the bottom half of this Russell's paradox site.
I get around 200 spams a day, after filtering. How many do I need to get, how long do I have to waste each day, before you accept that it's a problem? Geez, I swear people like you are part of the problem, not the solution. Grrrr!
I have no relationship to anyone else here other than impartial bystander, but I would suggest that you not attack the people who are trying to build better filters. Even if you think filters are in general futile, you must surely admit that it is worth a shot.
I'm seriously interested in how much spam you get every day that 200 messages would slip through the filters. I get "only" about 200 spams a day before filtering, and spamassassin (using only a simple bayesian filter that you so deride) catches 99.9% of spams with less than.1% false positives for a grand total of about one spam per five days after filtering, on average.
Is your spam so different from mine that your filter's accuracy suffers tremendously? Or do you really get 200000 spams a day of which 99.9% is filtered? I'd be interested in some samples of spam that made it through your filters, to see how they would stack up against my filters.
I think the Apache SpeedLimit perl module accomplishes the effect that you want, but not in the same way as you asked for.
SpeedLimit works by limiting the request rate of each IP address. If your web site consists of many small files (which sounds like the case), then curbing the request rate is enough to cure the most abusive bots. A determined adversary can still circumvent request rate limits with wget --random-wait, but it will be more frustrating for them, and a large percentage of clients can be expected to give up altogether.
If your web site has lots of large files, then (tooting my own horn here) your best choice is to use my mod_limitipconn module together with mod_bandwidth. The mod_bandwidth sets a total limit on traffic and the mod_limitipconn ensures that any single IP address gets only its fair share of that total traffic. I would also advise you in this case to use mod_bandwidth's built in ability to exempt small files from the bandwidth limits.
If your situation is such that the vast majority of visitors intend to download your whole web site, then your best option is to seed a bittorrent tarball of your whole web site.
That should just about cover all the bases. No single one of these proposals is ideal, but you have to realize that overcapacity has no elegant solution. The goal is to manage the situation as best you can using the available tools.
Q45: What other major spam advisory lists and blocking systems are there? What are your opinions of them?
MAPS RSS: Nice when it worked. ORBS-type lists are better, you should not have to actually get proof of spam through a relay before blocking it. Due to abuse by spammers, open email relays no longer have any place on the Internet. Some may want to debate this, we won't.
John Gilmore (founder of the EFF) has for a long time been running an open relay which is not abuseable by spammers. It works by rate limiting each user of the relay.
I am very sympathetic to the complaints of harming innocent third parties, and indeed I used to be very supportive of anti-spam efforts. But these days I find that the anti-spammers are doing just as much harm to innocent parties as the spammers themselves. Real time blacklists are some of the worst offenders, since many of them (e.g. SPEWS) actively promote collateral damage as a mechanism for encouraging change.
I don't see how open relay blacklists like orbs or SPEWS can say with a straight face that they care about innocent third party damage from open relays. I consider the damage inflicted by one lost legitimate mail to be far worse than the damage inflicted by one unwanted spam mail.
Changing a object's orbit, within the small altitude changes we are talking about, involves essentially the same velocity change whether or not you are bringing it one kilometer higher or one kilometer lower.
There is one exception: changing the eccentricity of an orbit is easier than shifting to a different altitude at the same eccentricity.
Crashing an orbiting object into earth does not actually require changing the apogee of the orbit. All you have to do is change the perigee i.e. the orbital eccentricity. Once the object hits the drag of the atmosphere, you don't need to spend any more energy on slowing it down.
if you'd bother to follow the links i make at the top of this thread, you'd see precisely how close HST is to the ISS.
Just because the orbits are close doesn't mean the objects are close. The links you gave do not show the heights of the objects. You realize that space is three dimensional right? As a matter of fact the ISS flies considerably higher than the HST. It takes MUCH more fuel to boost an object's height than to send it crashing into earth.
In fact, even if they were in the same orbit, at the same heights, they could still be 180 degrees out of phase with each other within that orbit. Bringing the two craft together in that scenario would cost more than trashing the hubble and relaunching it.
You do realize that you can't just stop in orbit and wait for the other vessel to catch up. The whole point of an orbit is that you have to keep moving to stay in the orbit.
I want you to consider that the shuttles have quite easilly made it to both HST and ISS on several occasions
This is absolutely false. I challenge you to present even one space shuttle mission that visited both the HST and the ISS. It's never happened before.
Can I get my Ogg files to one of these players in Linux, or do I have to use Windows? Can anyone explain how that works before I buy one of these? I don't do Windows anymore, I have my Ogg files on my harddrive - in Linux, how do I move them to these players? Thanks.
The only way to transfer songs is Linux at the moment is through the ethernet interface, because the USB interface does not implement the mass storage standard.
You have to connect the Karma player to your LAN (using either dhcp or keying in a static IP into the device). Then you point your web browser to the embedded http server on the player and download a java program that you can use to transfer songs. The program is not fancy but it has all the essential features it needs for daily use.
It must be mentioned that you cannot upgrade the firmware from Linux yet.
On the whole, I would say that the process of transferring songs in Linux is harder than with the Korean made players, but still easier than trying to get the iPod to work in Linux since at least Linux is supposed to work out of the box.
Why did they choose to work at a factory than work where they did before?
Prisoner's dilemma, caused by
Relativity of status, and exacerbated by
Incomplete information.
Let me try to explain in more detail. I assume you are familiar with the
prisoner's dilemma. It is the classic game theory example of locally optimal play that is globally non-optimal. In this case, what happens is that each individual worker's optimal strategy is to work in a factory, but the global optimal strategy is for no one to work in a factory at all.
The reasons for this dilemma lie in the second and third reasons. Most farm workers are enticed to work in a factory by higher wages. By earning higher wages than their peers, the factory workers gain purchasing power and are better off. The problem is that purchasing power is relative, not absolute. Once everybody is a factory worker, no one has any extra purchasing power anymore, and everyone's economic status is the same as before except that they now have factory jobs instead of farm jobs.
Incomplete information is a separate but contributing factor that makes the problem worse: most workers are unaware of the true health costs of factory work, and most people (workers or not) tend to discount future costs too heavily anyway as part of human nature, so many people who should never be working in a factory in a perfect market end up working in factories anyway.
Copyright law does not give a 100% ban on any redistribution. This is what fair use is all about. Allowing people to distribute small portions for personal use.
Even ignoring the fact that KISS is not a person, the fair use law says nothing about personal use. In fact, the law says:
... the fair use of a copyrighted work... for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright.
I really don't think that what KISS is doing qualifies under any of the categories above. Admittedly, the phrase "such as" is not exclusive, but if the purported fair use does not even remotely resemble any of the listed examples, then it is very hard to justify.
Copying an mp3 is NOT fair use. Firstly because of the nature of the work - i.e. it is made with the intention of making a profit for the distributors
You seem to think that the intention of the author matters for fair use. In fact, the intention of the author is not mentioned at all in the text of the law. What is relevant is "the purpose and character of the use", i.e. whether or not the use is profit oriented. Copying mp3s for personal use in your car is not profit oriented, but what KISS is doing is very much profit oriented. So KISS actually has much less of a leg to stand on than a home user copying mp3s for personal use.
There is something else that you are leaving out as well: the Audio Home Recording Act of 1992 permits consumers to make limited noncommercial copies of recorded audio for personal use. This permission is above and beyond the normal permissions granted under fair use. There is absolutely no analogue of the AHRA that applies to software. Even if the personal copying of mp3 were somehow found to be outside of fair use, the AHRA would still apply in the mp3 situation. However in the KISS situation there is no other law that would permit what KISS is doing.
I admit my numbers are partly made up. They are based on my experiences distributing DVD fansubs to people. Do you regularly distribute recordable DVD discs to large numbers of random people in the general public? If so, I would love to hear your own compatibility experiences. If not, then I stand by my claims on the grounds that I have more actual experience than you.
Even if my numbers are made up, it's not clear that hard statistics for a useless figure are better than soft statistics for a useful figure.
We seem to be viewing compatibility from two different and incompatible perspectives.
You're asking the question "what player should I get to maintain compatibility with recordable DVD?" While this is a wonderful question to ask and answer, it's not the same question as the question I am faced with when choosing media.
When choosing media, the question I am faced with is "what media should I use to maintain compatibility with the DVD players that other people (not me) have already purchased?" Your advice about what player models to purchase is solid advice, but completely irrelevant for me in my position. As a video distributor, I'm not in a position to dictate what players other people purchase. I have to choose a disc that plays in their players. I don't get to choose what players they buy, since in most cases the people I send my discs to have already purchased their DVD players.
I own a compaq presario 2100 as well (proof), and while I am happy with the battery, the hard drive is not user replaceable as far as I can tell.
I've opened all the side panels and bottom panels and none of them leads to a hard drive. As best I can tell, the hard drive is near the front of the laptop under the touchpad, but I haven't been able to access it yet, even after unscrewing all the screws on the case and opening the laptop as far as I dared.
If you know how to replace the hard drive on a compaq presario 2100us, I'd love to hear from you.
As for the question of whether to leave the battery in, I usually take it out for long term storage because the buchmann battery faq says li-ion batteries store best at 40% charge level and cold temperatures. While I don't go so far as to refrigerate my batteries, I can't help but think that storing them at 40% charge, 25 degC is better than storing them in my laptop where they would be at 100% charge, 40 degC. So far my oldest battery (1 year old) is at about 88% of original charge, which is in line with the figures given on the web page considering that it has seen more usage than just sitting in storage.
According to DVDrHelp here, -Rs have about 92% compatability, while +Rs have about 86% compatability. Hardly a huge difference.
It actually is a huge difference, but most people don't realize why.
As I already explained previously in this post, the "percentage of DVD models" statistic is almost useless for answering questions of compatibility. If I am concerned about playing one disc on many different players, then I don't care about how many player models are compatible; instead I care about how many player units are compatible.
The distinction is subtle but important. Not all player models have the same market share. In fact you will find that the DVD+R-incompatible models tend to have a far larger market share than the DVD-R incompatible models. If you count player models, then your 92/86 figure is accurate. But if you count the number of actual DVD players sold, taking into account the nonuniform market share distribution of DVD player models, you'll find that the actual compatibility difference is something like 95/70, and that really matters to me.
I have a Hauppauge WinTV PVR USB capture card for use with my laptop under linux. Note that the PVR part is important -- there is another Hauppauge card called WinTV USB which in my opinion is far inferior. The difference is that the PVR card has a hardware mpeg encoder and it encodes the video into mpeg before sending it down the USB wire, which gives you a lot more video quality for the limited bandwidth available under USB.
I wrote a linux-oriented review of this card on the dvdrhelp web site (search for the May 10, 2003 review) which contains everything that I have to say about using this card in linux.
I tend to play when the jackpot is 3N where the chance of winning is 1/N, since I like poker and this situation only comes up every few years, but to take everything into account, you should wait until the jackpot is about 6N
Some states let you pick your own numbers on your lottery tickets. You can substantially increase your chances of not sharing the prize by picking relatively high numbers, which are less likely to be picked by others since most other people's "lucky" numbers tend to be low numbers.
Picking the high numbers doesn't affect your own chance of winning, but it has a very definite effect on your chance of having to share the prize if you win. Unfortunately I don't have specific figures to quantify the size of this effect.
I have an iRiver 180T and I love it. The battery life is exceptional and the earphones are the best in the market.
I have no experience with the iRiver 180T, but I have a very hard time believing the earphones are the best in the market.
The etymotic earphones that I use, which as far as I know are the best sounding earbud headphones in the world, cost (for the headphones alone) almost three times as much as the iRiver mp3 player. If iRiver can sell better headphones than the etymotics for 1/3 the price, and throw in an mp3 player along for free, then I'd be very interested in buying one.
Um, I'm pretty sure the PTO gets paid whether your patent is granted or not. So there's no profit motive for them in granting patents.
What you say is literally true, but it is way too shallow an analysis. Yes, it is true that on any given application, the application fee has already been paid. But the only way to encourage lots of applications is to approve lots of applications.
If the patent office denies a large fraction of applications, then they will discourage applications in the first place, and their revenue from application fees will go down. It is very easy to see why the patent office has a financial incentive to approve as many patent applications as possible.
I have no relationship with Linux Gazette other than once having contributed an article to LG (issue #35). My reading of the situation is that the trademark belongs to the originators of the magazine (i.e. the people who have recently left SSC), rather than the company that got involved almost a year later (i.e. SSC).
I would be shocked and surprised if a hosting company could acquire trademark rights to a web site merely by hosting the web site. If this were so then Rackspace.com would find itself right now in possession of a very large number of trademark rights. Now I agree that SSC provided free hosting instead of paid hosting, but I fail to see how the fact that SSC provided its hosting for free changes anything.
Likewise, although I realize SSC has contributed much effort to LG since SSC got involved with LG, I do not see these contributions establishing any trademark rights either. After all, *I* have contributed to LG before as well, and you don't see me going around asserting that I have trademark rights to LG.
SSC should do the right thing and admit that it has no trademark rights to the name LG, relinquish the linuxgazette.com domain to the founders of LG, and publish their CMS under some other name. It is clear to me that LG is not a CMS, never has been a CMS, and that SSC is going to have a very difficult time arguing that the CMS is truer to the LG name than the rival publication.
I should also point out that even if SSC somehow manages to win a legal case and keep the LG name, it will be blackballed by a sizable fraction (possibly even the majority) of the linux community, who, like me, view the founding volunteers of LG as the true keepers of the LG torch.
It's totally permissible to run their Enterprise products on as many machines as you like without any RHN subscription or support contract. Just buy one copy, and install it everywhere.
I hate to be negative, but I do not actually think that this is legal.
Look carefully at the RHEL EULA. Here are some quotes, emphasis added:
The term "Installed Systems" means the number of Systems on which Customer installs the Software. The term "System" means any hardware on which the Software is installed, which may be, without limitation, a server, a work station, a virtual machine, a blade, a partition or an engine, as applicable. The initial number of Installed Systems is the number of copies of the Software that Customer purchases.
If Customer wishes to increase the number of Installed System, then Customer
will purchase from Red Hat additional Services for each additional Installed System. During the term of this Agreement and for one (1) year thereafter, Customer expressly grants to Red Hat the right to audit Customer's facilities and records from time to time in order to verify Customer's compliance...
Of course Redhat is not allowed to impose this EULA on third party GPL software, but the problem is that not all of RHEL is third party GPL software. A lot of it is third party free software under licenses other than the GPL. Moreover a substantial fraction of the GPL software in RHEL is actually owned by Redhat themselves.
So, ironically, the only way to install RHEL on multiple machines without support is not to buy RHEL and compile your own copy.
On the best of gear, professionals can't tell which track is original, between high bitrate mp3 and uncompressed.
This may be true for most tracks, but it is definitely not true for all tracks. For example I have never heard any mp3 copy of the notorious fatboy slim track, at any bitrate, that sounds like the original, and I am nowhere near a professional. I think the vast majority of people with normal hearing would be able to identify the lossy copy of the fatboy slim track at any bitrate nearly 100% of the time.
BTW, your ER-4P is not that accurate, it actually enhances bass response for portable listening.. it's ety's jump into the portable market.
I actually like the extra bass, so call it euphonic distortion if you will... but on the other hand I also have the 4P to 4S converter dongle which still sounds to me better than speakers.
Many of the tracks people listen to were made to listen to on speakers, not headphones. They are not binaural recordings, desgined for 100% stereo isolation you get with headphones (Unless you use amps with crossfeed, that's another story).
I do happen to use amps with crossfeed with the 4S (but not the 4P, since an amp hinders portability).
Also, I admit I am guilty of confusing the issue somewhat, as the original issue was about lossless vs. lossy instead of headphone quality. However, given that a portable is always going to be listened with headphones, the problem of stereo isolation is not relevant to the lossless vs. lossy debate. After all there is no way you can argue that lossy compression somehow ameliorates the stereo isolation problem.
Why support FLAC? Granted, I love a free, lossless codec. It's great for listening on your computer... but you listen to your portable MP3 player with headphones. You're probably not going to notice the difference between lossless and lossy compression.
Believe me there are headphones where you can tell the difference. The $300 Etymotic ER-4P headphones are more than portable enough for a portable player and produce better sound than all but maybe a half dozen (no exaggeration) full size headphone models. In fact for regular stereo audio (i.e. not surround sound), a good pair of headphones is almost guaranteed to sound better than the same amount of money spent on speakers, because speakers have to contend with reflection noise off your walls.
So I'd say you have it backwards -- computer listening doesn't really benefit much from lossless audio, but headphone listening sure can.
Even if you don't feel like spending $300 on headphones, there are still many lesser headphones for which FLAC is worthwhile. Don't judge headphone quality based on the cheap headphones included with the player.
Hey, I hope you don't mind me fishing for responses, but I have a few things I'd like to ask you about the Rio Karma. I have not seen these things discussed in any online reviews so far.
Does the Karma support dynamic playlist building? That is, can you program a playlist on the device while it is playing music?
Does the Karma support gapless playback? I've heard about the crossfade feature, but I'm much more interested in gapless transitions between tracks.
Does it display non-western charsets in the song titles? Chinese, Japanese, Korean, etc. Even accented ISO-8859-1 European characters would be a good start.
Finally, does it play vorbis files at all bitrates? libvorbis 1.0 can encode 48kbps up to about 320 kbps. I was quite disappointed that my neuros was unable to play anything below 64kbps. Some of Garf's demo files contain entire songs encoded in vorbis at as low as 4kbps. I have not heard of anyone who has tried to play these back on a portable.
Note that I am very close to getting a Karma anyway, even if the answers are all no... but if you could tell me that some of them are yes, then that would really seal the deal.
Slashdot Mirror
Even hard core gaming sites like Sharky Extreme are now recommending optical mice exclusively in all their hardware guides.
However in this case your comment is only slightly wrong and therefore I have some hope that my reply might be a useful contribution.
You are correct that mathematical proofs are based on axioms. However there is still a crucial difference between a mathematical proof and a scientific theory. A mathematical proof is an absolute certainty. Note that I am not claiming that the underlying axioms are certain. I am only claiming that the proof itself is certain.
To put it another way, mathematicians are never certain about their underlying axioms but they are absolutely certain that if those axioms hold then the result stated in the proof also holds. It's kind of like a building with indestructible walls but no foundation.
Scientific theory is a whole different kettle of fish. You cannot prove a scientific theory with absolute certainty. In fact it is not even clear to me how one can define certainty within the framework of the scientific method. You never have any guarantee in science that future observations will be consistent with past observations.
In science you can prove a theory in the sense of preponderance of the evidence. You can even sometimes prove a theory beyond all reasonable doubt. But there is no way to eliminate the unreasonable doubts. Any endeavour based on empirical observation suffers from the fundamental limitation that you can never be sure of the next observation.
Finally, regarding 1+1=2, the foundational proof of this fact using the standard propositional axioms of mathematics really does require 362 pages. You can see the 362nd page on the bottom half of this Russell's paradox site.
I have no relationship to anyone else here other than impartial bystander, but I would suggest that you not attack the people who are trying to build better filters. Even if you think filters are in general futile, you must surely admit that it is worth a shot.
I'm seriously interested in how much spam you get every day that 200 messages would slip through the filters. I get "only" about 200 spams a day before filtering, and spamassassin (using only a simple bayesian filter that you so deride) catches 99.9% of spams with less than .1% false positives for a grand total of about one spam per five days after filtering, on average.
Is your spam so different from mine that your filter's accuracy suffers tremendously? Or do you really get 200000 spams a day of which 99.9% is filtered? I'd be interested in some samples of spam that made it through your filters, to see how they would stack up against my filters.
SpeedLimit works by limiting the request rate of each IP address. If your web site consists of many small files (which sounds like the case), then curbing the request rate is enough to cure the most abusive bots. A determined adversary can still circumvent request rate limits with wget --random-wait, but it will be more frustrating for them, and a large percentage of clients can be expected to give up altogether.
If your web site has lots of large files, then (tooting my own horn here) your best choice is to use my mod_limitipconn module together with mod_bandwidth. The mod_bandwidth sets a total limit on traffic and the mod_limitipconn ensures that any single IP address gets only its fair share of that total traffic. I would also advise you in this case to use mod_bandwidth's built in ability to exempt small files from the bandwidth limits.
If your situation is such that the vast majority of visitors intend to download your whole web site, then your best option is to seed a bittorrent tarball of your whole web site.
That should just about cover all the bases. No single one of these proposals is ideal, but you have to realize that overcapacity has no elegant solution. The goal is to manage the situation as best you can using the available tools.
From the SPEWS FAQ:
I am very sympathetic to the complaints of harming innocent third parties, and indeed I used to be very supportive of anti-spam efforts. But these days I find that the anti-spammers are doing just as much harm to innocent parties as the spammers themselves. Real time blacklists are some of the worst offenders, since many of them (e.g. SPEWS) actively promote collateral damage as a mechanism for encouraging change.
I don't see how open relay blacklists like orbs or SPEWS can say with a straight face that they care about innocent third party damage from open relays. I consider the damage inflicted by one lost legitimate mail to be far worse than the damage inflicted by one unwanted spam mail.
There is one exception: changing the eccentricity of an orbit is easier than shifting to a different altitude at the same eccentricity.
Crashing an orbiting object into earth does not actually require changing the apogee of the orbit. All you have to do is change the perigee i.e. the orbital eccentricity. Once the object hits the drag of the atmosphere, you don't need to spend any more energy on slowing it down.
Just because the orbits are close doesn't mean the objects are close. The links you gave do not show the heights of the objects. You realize that space is three dimensional right? As a matter of fact the ISS flies considerably higher than the HST. It takes MUCH more fuel to boost an object's height than to send it crashing into earth.
In fact, even if they were in the same orbit, at the same heights, they could still be 180 degrees out of phase with each other within that orbit. Bringing the two craft together in that scenario would cost more than trashing the hubble and relaunching it.
You do realize that you can't just stop in orbit and wait for the other vessel to catch up. The whole point of an orbit is that you have to keep moving to stay in the orbit.
I want you to consider that the shuttles have quite easilly made it to both HST and ISS on several occasions
This is absolutely false. I challenge you to present even one space shuttle mission that visited both the HST and the ISS. It's never happened before.
The only way to transfer songs is Linux at the moment is through the ethernet interface, because the USB interface does not implement the mass storage standard.
You have to connect the Karma player to your LAN (using either dhcp or keying in a static IP into the device). Then you point your web browser to the embedded http server on the player and download a java program that you can use to transfer songs. The program is not fancy but it has all the essential features it needs for daily use.
It must be mentioned that you cannot upgrade the firmware from Linux yet.
On the whole, I would say that the process of transferring songs in Linux is harder than with the Korean made players, but still easier than trying to get the iPod to work in Linux since at least Linux is supposed to work out of the box.
- Prisoner's dilemma, caused by
- Relativity of status, and exacerbated by
- Incomplete information.
Let me try to explain in more detail. I assume you are familiar with the prisoner's dilemma. It is the classic game theory example of locally optimal play that is globally non-optimal. In this case, what happens is that each individual worker's optimal strategy is to work in a factory, but the global optimal strategy is for no one to work in a factory at all.The reasons for this dilemma lie in the second and third reasons. Most farm workers are enticed to work in a factory by higher wages. By earning higher wages than their peers, the factory workers gain purchasing power and are better off. The problem is that purchasing power is relative, not absolute. Once everybody is a factory worker, no one has any extra purchasing power anymore, and everyone's economic status is the same as before except that they now have factory jobs instead of farm jobs.
Incomplete information is a separate but contributing factor that makes the problem worse: most workers are unaware of the true health costs of factory work, and most people (workers or not) tend to discount future costs too heavily anyway as part of human nature, so many people who should never be working in a factory in a perfect market end up working in factories anyway.
Even ignoring the fact that KISS is not a person, the fair use law says nothing about personal use. In fact, the law says:
I really don't think that what KISS is doing qualifies under any of the categories above. Admittedly, the phrase "such as" is not exclusive, but if the purported fair use does not even remotely resemble any of the listed examples, then it is very hard to justify.Copying an mp3 is NOT fair use. Firstly because of the nature of the work - i.e. it is made with the intention of making a profit for the distributors
You seem to think that the intention of the author matters for fair use. In fact, the intention of the author is not mentioned at all in the text of the law. What is relevant is "the purpose and character of the use ", i.e. whether or not the use is profit oriented. Copying mp3s for personal use in your car is not profit oriented, but what KISS is doing is very much profit oriented. So KISS actually has much less of a leg to stand on than a home user copying mp3s for personal use.
There is something else that you are leaving out as well: the Audio Home Recording Act of 1992 permits consumers to make limited noncommercial copies of recorded audio for personal use. This permission is above and beyond the normal permissions granted under fair use. There is absolutely no analogue of the AHRA that applies to software. Even if the personal copying of mp3 were somehow found to be outside of fair use, the AHRA would still apply in the mp3 situation. However in the KISS situation there is no other law that would permit what KISS is doing.
Even if my numbers are made up, it's not clear that hard statistics for a useless figure are better than soft statistics for a useful figure.
You're asking the question "what player should I get to maintain compatibility with recordable DVD?" While this is a wonderful question to ask and answer, it's not the same question as the question I am faced with when choosing media.
When choosing media, the question I am faced with is "what media should I use to maintain compatibility with the DVD players that other people (not me) have already purchased?" Your advice about what player models to purchase is solid advice, but completely irrelevant for me in my position. As a video distributor, I'm not in a position to dictate what players other people purchase. I have to choose a disc that plays in their players. I don't get to choose what players they buy, since in most cases the people I send my discs to have already purchased their DVD players.
I've opened all the side panels and bottom panels and none of them leads to a hard drive. As best I can tell, the hard drive is near the front of the laptop under the touchpad, but I haven't been able to access it yet, even after unscrewing all the screws on the case and opening the laptop as far as I dared.
If you know how to replace the hard drive on a compaq presario 2100us, I'd love to hear from you.
As for the question of whether to leave the battery in, I usually take it out for long term storage because the buchmann battery faq says li-ion batteries store best at 40% charge level and cold temperatures. While I don't go so far as to refrigerate my batteries, I can't help but think that storing them at 40% charge, 25 degC is better than storing them in my laptop where they would be at 100% charge, 40 degC. So far my oldest battery (1 year old) is at about 88% of original charge, which is in line with the figures given on the web page considering that it has seen more usage than just sitting in storage.
It actually is a huge difference, but most people don't realize why.
As I already explained previously in this post, the "percentage of DVD models" statistic is almost useless for answering questions of compatibility. If I am concerned about playing one disc on many different players, then I don't care about how many player models are compatible; instead I care about how many player units are compatible.
The distinction is subtle but important. Not all player models have the same market share. In fact you will find that the DVD+R-incompatible models tend to have a far larger market share than the DVD-R incompatible models. If you count player models, then your 92/86 figure is accurate. But if you count the number of actual DVD players sold, taking into account the nonuniform market share distribution of DVD player models, you'll find that the actual compatibility difference is something like 95/70, and that really matters to me.
I wrote a linux-oriented review of this card on the dvdrhelp web site (search for the May 10, 2003 review) which contains everything that I have to say about using this card in linux.
Some states let you pick your own numbers on your lottery tickets. You can substantially increase your chances of not sharing the prize by picking relatively high numbers, which are less likely to be picked by others since most other people's "lucky" numbers tend to be low numbers.
Picking the high numbers doesn't affect your own chance of winning, but it has a very definite effect on your chance of having to share the prize if you win. Unfortunately I don't have specific figures to quantify the size of this effect.
I have no experience with the iRiver 180T, but I have a very hard time believing the earphones are the best in the market.
The etymotic earphones that I use, which as far as I know are the best sounding earbud headphones in the world, cost (for the headphones alone) almost three times as much as the iRiver mp3 player. If iRiver can sell better headphones than the etymotics for 1/3 the price, and throw in an mp3 player along for free, then I'd be very interested in buying one.
What you say is literally true, but it is way too shallow an analysis. Yes, it is true that on any given application, the application fee has already been paid. But the only way to encourage lots of applications is to approve lots of applications.
If the patent office denies a large fraction of applications, then they will discourage applications in the first place, and their revenue from application fees will go down. It is very easy to see why the patent office has a financial incentive to approve as many patent applications as possible.
I would be shocked and surprised if a hosting company could acquire trademark rights to a web site merely by hosting the web site. If this were so then Rackspace.com would find itself right now in possession of a very large number of trademark rights. Now I agree that SSC provided free hosting instead of paid hosting, but I fail to see how the fact that SSC provided its hosting for free changes anything.
Likewise, although I realize SSC has contributed much effort to LG since SSC got involved with LG, I do not see these contributions establishing any trademark rights either. After all, *I* have contributed to LG before as well, and you don't see me going around asserting that I have trademark rights to LG.
SSC should do the right thing and admit that it has no trademark rights to the name LG, relinquish the linuxgazette.com domain to the founders of LG, and publish their CMS under some other name. It is clear to me that LG is not a CMS, never has been a CMS, and that SSC is going to have a very difficult time arguing that the CMS is truer to the LG name than the rival publication.
I should also point out that even if SSC somehow manages to win a legal case and keep the LG name, it will be blackballed by a sizable fraction (possibly even the majority) of the linux community, who, like me, view the founding volunteers of LG as the true keepers of the LG torch.
I hate to be negative, but I do not actually think that this is legal.
Look carefully at the RHEL EULA. Here are some quotes, emphasis added:
Of course Redhat is not allowed to impose this EULA on third party GPL software, but the problem is that not all of RHEL is third party GPL software. A lot of it is third party free software under licenses other than the GPL. Moreover a substantial fraction of the GPL software in RHEL is actually owned by Redhat themselves.So, ironically, the only way to install RHEL on multiple machines without support is not to buy RHEL and compile your own copy.
Look for fatboy.wav on the LAME test samples page.
This may be true for most tracks, but it is definitely not true for all tracks. For example I have never heard any mp3 copy of the notorious fatboy slim track, at any bitrate, that sounds like the original, and I am nowhere near a professional. I think the vast majority of people with normal hearing would be able to identify the lossy copy of the fatboy slim track at any bitrate nearly 100% of the time.
BTW, your ER-4P is not that accurate, it actually enhances bass response for portable listening.. it's ety's jump into the portable market.
I actually like the extra bass, so call it euphonic distortion if you will ... but on the other hand I also have the 4P to 4S converter dongle which still sounds to me better than speakers.
Many of the tracks people listen to were made to listen to on speakers, not headphones. They are not binaural recordings, desgined for 100% stereo isolation you get with headphones (Unless you use amps with crossfeed, that's another story).
I do happen to use amps with crossfeed with the 4S (but not the 4P, since an amp hinders portability).
Also, I admit I am guilty of confusing the issue somewhat, as the original issue was about lossless vs. lossy instead of headphone quality. However, given that a portable is always going to be listened with headphones, the problem of stereo isolation is not relevant to the lossless vs. lossy debate. After all there is no way you can argue that lossy compression somehow ameliorates the stereo isolation problem.
Believe me there are headphones where you can tell the difference. The $300 Etymotic ER-4P headphones are more than portable enough for a portable player and produce better sound than all but maybe a half dozen (no exaggeration) full size headphone models. In fact for regular stereo audio (i.e. not surround sound), a good pair of headphones is almost guaranteed to sound better than the same amount of money spent on speakers, because speakers have to contend with reflection noise off your walls.
So I'd say you have it backwards -- computer listening doesn't really benefit much from lossless audio, but headphone listening sure can.
Even if you don't feel like spending $300 on headphones, there are still many lesser headphones for which FLAC is worthwhile. Don't judge headphone quality based on the cheap headphones included with the player.
Does the Karma support dynamic playlist building? That is, can you program a playlist on the device while it is playing music?
Does the Karma support gapless playback? I've heard about the crossfade feature, but I'm much more interested in gapless transitions between tracks.
Does it display non-western charsets in the song titles? Chinese, Japanese, Korean, etc. Even accented ISO-8859-1 European characters would be a good start.
Finally, does it play vorbis files at all bitrates? libvorbis 1.0 can encode 48kbps up to about 320 kbps. I was quite disappointed that my neuros was unable to play anything below 64kbps. Some of Garf's demo files contain entire songs encoded in vorbis at as low as 4kbps. I have not heard of anyone who has tried to play these back on a portable.
Note that I am very close to getting a Karma anyway, even if the answers are all no... but if you could tell me that some of them are yes, then that would really seal the deal.