Can you tell me a track or two where I might hear the difference?
I don't remember what my friend and I were using (actually, *he* was running the test for himself -- it was his CD, and I just happened to drop in and try it as well). It was something with plenty of percussion -- drums and cymbals. IIRC, and this was a while ago, the most noticeable MP3 artifact happened in cymbals (you can also hear this in white-noise type stuff). You hear...darn, I can't think of anything like it, but you hear "swishing" sounds in the cymbals. With Vorbis, the drums sounded a tiny bit flatter (the flatter drums didn't bother me, but the cymbals were annoying).
I wonder if it was my speakers that made the difference.
I've heard that it's much easier to hear minor things like that in headphones than in speakers -- I wouldn't know, since I don't own any nice speakers -- headphones are much cheaper.:-)
And Radio Shack headphones? What kind of audio snob are you?
A bad one, I reckon.:-)
Both of us were surprised that the artifacts could be detected at that bitrate, FWIW. My friend couldn't pick up on the Vorbis artifacts, but could manage the MP3 artifacts.
I'm a little suspicious that it isn't anything about the "quality of the ears", but just a matter of training yourself. I played the MP3, the Vorbis and the.wav back and forth (not blind) many times to try to learn the differences first. While I was able to teach myself to recognize the compressed audio, in an environment where I hadn't just done this, I doubt I'd be able to say "Aha! That's Vorbis-compressed audio!" if I just heard a random clip compressed with something.
A double-blind test, yes (A friend ran a program that randomly played a clip, then I called out what I thought, and then he told me what it was). I had decent headphones, but nothing amazing from an audiophile standpoint (about $70, Radio Shack, closed circumaural), and I'm certainly no musician. The test clip I was using had drums and cymbals. I could tell the difference between.wav and.mp3 consistently at 256 KBps CBR (encoded with LAME). Cymbals started to sound "swishy" -- they start to pick up unpleasant patterns.
With Vorbis, I could tell the difference all the way up to the maximum quality setting (to my surprise). With Vorbis, cymbals didn't sound different (to my ear, at least), but percussion sounds slightly different -- a little bit flatter, maybe.
I still preferred Vorbis, because the MP3 artifacts sounded *unpleasant* -- swishy cymbals sound bad -- but Vorbis's artifacts don't.
Note that I was using CBR with LAME (because at the time I was interested in CBR results), though that isn't what I'd use in real life.
When listening on my little computer multimedia speakers (about $25) I've found that I can't distinguish between.wav and MP3 at a lower bitrate (but I don't remember the precise results, and in any case, that wasn't a double-blind test). Note that this was two years or so ago, and both encoders may have improved since then.
When e-mail clients can match that functionality, allowing me to follow a thread of e-mail messages as easily as following a thread on Usenet, I'll find a reason to switch. When I can manually reorganize message threads (so that messages with different subjects are included), I'll be even happier. You guys at Google listening?
Lots of mail clients thread (if the option to do so is enabled -- most can also not thread). Mutt does. Even Outlook does.
What I can't hear is the difference between a properly done 192kbps/44khz mp3 rip and one made with FLAC.
Try something with cymbals.
Also, while FLAC may not be useful for a portable music player (especially in a typically noisy environment, with those little earbud things), my desktop computer has ungodly amounts of space available -- FLAC makes a lot of sense for that.
JPEG artifacts may not be visible in normal images under typical viewing conditions, and JPEG may be the most suitable format for photographic images in web pages, but nobody is pushing to replace TIFF with JPEG.:-)
I *would* be interested in knowing why there's so little support for Vorbis -- I can understand on, say, the little value-oriented devices that use ASIC MP3 decoders, but the iPod has twin ARMs, and should be able to handle Vorbis.
I just crossed from office-geek into "don't stare and walk away slowly" dude.
I don't think that it's really that weird, unless you think of "anyone trying something new" as weird. Automobiles and electric lighting were all new at one point...
I suspect that the enjoyment of a media is greater when you have more related stimuli firing that your brain can tie together.
People like bass. I doubt that this is because low notes are just really neat, but because they can *feel* the bass -- vibration and air pressure differences on their skin. A chair that has an embedded subwoofer could produce some of this effect.
Video games that have audio triggered by various in-game events, producing a synchronized audio-video simuli, have stomped silent video games.
There have been some attempts at games that produce even more synchronization between audio and video. A number of games now have not merely sound effects triggered by game events, but adaptive music. The game state affects the type of background music playing -- Lumines, Total Annihilation, all kinds of newer games. Your brain gets to tie together finger actions with music.
There have even been some video games which alter game state based on audio -- for example, Vib-Ribbon and Rez have game events triggered in synchronization with the audio as *well* as audio that adapts itself to game events (and, for that matter, outside of the US, Rez had a vibrator that one could hook up to the game via USB to provide tactile feedback). Now, Rez didn't sell all that well, but I strongly suspect that this was because it was too unusual for people to try it out -- the people I know that *do* play Rez are generally pretty rabid about it. The concept of associating more stimuli to produce a more enjoyable experience generally seems to work pretty well. The same approach seemed to work well for Dance Dance Revolution.
I've seen a lot of game designers take the idea of tying more forms of stimuli together, but they often just seem to think something along the lines of, "I should make a game based on music", and then the game doesn't really rise above anything other than a game oriented around music. There's no real reason why you couldn't take *any* of the types of games that currently have background music (i.e. almost all of them) and simply synchronize various game events to the music.
I wonder how well a full-body suit would sell, something that provides tactile stimuli that could be synchronized with audio. I mean, yes, it would be expensive, but I also see vendors for audiophiles selling multi-thousand-dollar cables, amplifiers, and headphones to try to reproduce that sensory experience crafted by the musician just a *little* more accurately -- surely, just adding more stimuli to the mix would be more worthwhile.
I wonder when the RIAA (which dearly loves moving to new formats and forcing everyone to buy their audio over again) will try shipping music in a new format, where it can contain video tracks or tactile tracks. If they want to be less ambitious, at least lighting tracks -- coming up with a standard where you have, say, eight lights each containing R, G, B elements, and allowing them to fade or flash to various colors in time to the music would seem to be a pretty cheap and compelling way to get people to purchase new music all over again.
WinAMP and various computer-based audio players try to automate the above via beat detection and use extracted data from the audio stream as input to a video image. However, as people who have tried this out know, not only is automated realtime beat and rhythm detected far from perfect, but elements of the music that are quite pe
My question is how does the money raised by these levies find its way to the copyright holders? (Artists, publishers and so on.)
Basically, all the people not signed by an RIAA label get a net 4% of the goods taken by blank media taxes (they don't have RIAA lobby dollars working for them). RIAA-signed artists get a total of 38.4% of the take, and the RIAA member publishers get 57.6% of the take. Note that this is a description from the RIAA, so that 38.4% may potentially be siphoned off into that 57.6% via fees orwhat-have-you.
It's irritating knowing that a percentage chunk of sales of many computer storage devices goes to lobbyists working to punch holes in free use, and even more irritating knowing that despite the fact that I have to pay for potential infringement, I *still* don't get to infringe legally.
You know...every time the RIAA complains that they need to be around to fund artists, and it's important that they exist to do so, and if they didn't get federal protection in the form of funds...I wonder what would happen if *100%* of that tax went to artists -- publishers not included. That *would* fund artists, presumably cutting out the middleman. Never going to happen, but an fun thought.
Another interesting idea -- an artist can choose to be supported by blank media sales *but* need to place all their work into the public domain *or* be supported by regular retail sales but not get any blank media sale funding.
And if your friend had his maps stolen by an African tribesman, could he successfully navigate by the stars?
It's all relative.
That being said, the reason so damn much high tech crap goes to the military is because "national security" is a great word to get ahold of federal money to siphon off to companies (defense contractors, construction types, etc) in your state.
Consider the F-22 Raptor. The F-22 has a unit cost of somewhere above $150 million. Now, I recognize that technology is great and all, but you have to wonder whether the F-22 could really maintain the 10:1 kill ratio against an F-16 or something that costs in the $15 million range. Yes, it's got doohickies and all that, but who are you planning to fight with this thing? Who do you need to out-technology? Tribesmen in Africa? Rebels in Iraq? Hell, at some point your simple crashes-due-to-accident becomes significant.
I think that every time I read about recent US military operations (Iraq, Somalia), there are problems with the sheer number of incompatible communication networks (everyone wants to develop their own network -- lots of $$$).
Identifying the scope of the commercial versus the open source problem is no easier than discerning real data on p2p usage.
Ah, yes. Those problematic open source P2P authors.
Mr. Garfield, I like hacking on P2P software. You can sic attorneys on every visible open source P2P author, and all that will happen is you will drive people underground -- and you don't need much of an underground to write all the software that anyone could ever use. You aren't going to manage to stop the production of open source P2P software.
Perhaps you'd like to look at Microsoft, Mr. Garfield. Microsoft has greater annual revenues than *all your member studios put together*. Microsoft has *clout*.
Microsoft wasn't able to quash open source development, despite spending an awful lot of money and effort on it, Mr. Garfield. I'm going to give you *very* slim chances of succeeding where they failed.
What are you going to try? PR? Microsoft did that. They called Linux a virus. They said it exposed users to liability. They said that it was insecure, and that it was *communist*, Mr. Garfield. It didn't work.
How about legislation? Maybe, if you're *really* lucky, you can manage to pay enough legislators to vote in laws criminalizing the production of software that is used to cause greater than some degree of purported damages. I don't think that you can manage that -- you'd face opposition from a lot of tech types, and a number of legislators have noticed that people don't *like* stories in the newspaper about nine-year-old girls being sued for thousands of dollars. But let's say that, despite all that, you manage it. There are a *lot* of open source programmers overseas, Mr. Garfield, and software does not understand national boundaries. The US government made export of encryption code illegal due to national security concerns for a long time. What happened? Encryption development and distribution continued, from overseas. It didn't do any good. You can't quash software development.
You going to try to track down all the people copying software and music and movies down? Mr. Garfield, one of the primary functions of a computer is to reproduce and distribute data quickly and accurately. There is *huge* demand for this, demand which far exceeds and outweighs any demand for entertainment. They have a device which does *exactly* what you don't want. There are *too many people* that want to be able to copy around movies for this to work.
How about a technical solution, Mr. Garfield? You spent plenty of effort trying to lock up DVDs -- that didn't work (you excluded Linux from your supported platforms, which was pretty stupid and put a lot of very smart Linux-using techies and crypto types to work on the problem, but even if you hadn't, it wouldn't have lasted long). You want to try again? Well, there are a lot of security types who would love to take your money and can guarantee you the moon, but it isn't going to happen.
You want to try keeping digital data from becoming analog? Good luck.
You want to try keeping analog data from becoming digital? This is a new, interesting one. You're now trying to plug a hole that requires *one* person with *one* analog-to-digital encoding device somewhere in the *world* per movie. It makes no more sense than trying to use CSS to keep people from getting at DVD content. It's just not a feasible approach.
I know that this is a really appalling concept, and one that you probably don't want to entertain. But it is possible -- just possible -- that your only solution is to reduce costs to where the convenience and guaranteed quality of buying your product from you outweighs the inconvenience of pirating. That means that you have to trim all your excess fat. That means that maybe you can't spend hundreds of millions of dollars producing and marketing a movie. Maybe you can't *have* actors that get tens of millions of dollars for every work. Maybe you need to use CG, and can't afford to recompen
You also clearly do lose if they blow up your home city.
Uh, huh.
Or if a meteorite hits me.
Thus far, the rate of meteorites hitting humans (one on record) has exceeded that of terrorists blowing up cities (none on record), and while you can paint a terribly heartstring-tugging picture using either device, I'm not worried about either.
And don't think the terrorists want to co-exist with us if we just let them rule the Middle East.
"Terrorists" rule the Middle East?
You mean Arabs?
Or perhaps Muslims?
The people running around blowing things up are largely people pissed off because they *don't* have political power other than by blowing things up. Are Palestinian militants "ruling" Israel?
They tolerate no one except themselves...
Ah. You seem to know a hell of a lot about these terrorists. You spend time with them? Have you been active in Middle Eastern politics? Or are you just talking out your ass because Fox News keeps telling you absurd scary things about them because terrorism is currently politically useful?
To be sure, as I have said above, piracy is not the whole problem - industry practices are part of the problem as well, but it is part of the problem. So what should we do about it?
I am inclined to oppose legislation unless we are absolutely certain that there is no technical solution that can come to the fore, or the market can simply be reshaped (for example, musicians getting tips or movies being sold in *extremely* large files -- so large that transfer of them is prohibitive). Or maybe competition finally driving ISPs to not allow unlimited data transfer any more, making P2P subsidized by lighter users less feasible.
The only reason to introduce legislation is to preserve the content distribution industry. While there is some cost to turmoil in the content distribution industry, I would point out that generally, if technology obsoletes an industry (such as the horse and carriage being obsoleted by the car), it's just not very feasible to try to extend the lifetime of the industry through legislation.
Traditionally, the expense of music production came in the distribution phase. The moneymaking step thus was attached to distribution, which led to music being produced based on funds provided by a music distributor (who thus assumed some financial risk). Getting the word out about good music was very hard, and so huge advertising budgets were worthwhile.
Today, we have very good systems for finding new music or movies that we like (and those systems get better by the day). IMDB, mp3.com, MovieLens, and so forth. This means that the marketing service that is tied into distribution is rapidly losing value. What about the distribution services? Well, it turns out that not only is distribution getting a lot easier and cheaper (via electronic distribution), but there is no initial cost to press a huge run and ship them to retail stores. That kind of eliminates much of the economic role of the music distributor as a necessary element in music production. Naturally, these industries are going to do what they can to protect themselves -- this is hardly cold and unfeeling, as Sally Smith, the record marketing directory for Acme Audio, wants to keep being able to buy Christmas toys for her kids. Ultimately, however, I'm not convinced that traditional distributors are still necessary for music production and consumption to occur.
I don't like the approach that this Conyers gentleman is aiming for. He's saying "well, if you don't like it, give me a better solution, and in the meantime, we're going to create this crummy law."
Frankly, the local live jazz groups don't need the RIAA to function. They're doing just fine. I'm not sure why I need to subsidize the RIAA's continued existence with my tax dollars and in increased prices for various electronic goods and in reduced functionality.
What I want to know is if they are planning on [...] adding features like sound support, Sound support is handled by a sound server, which fortunately runs independently of X.
While you can certainly set up network-remote sound, few people do. It just turns out that there are incredibly few applications where sound is actually used -- most of 'em are things like media players and games. With games, you frequently want to run the thing locally for performance reasons, and with media player, it's just more efficient to access the (generally compressed) media files over the network than it is to decompress the audio and stream it decompressed over the network.
Resolution changes don't require a reboot, just a restart of X. I think that he was talking about changing the pixel dimensions of the desktop. You can cycle through all the auto-detected or specified resolutions in modern X (control-alt-keypad+ or -keypad-). If you don't want to have the desktop scroll with your mouse, try xrandr -s 640x480 or similar.
Except for full-screen applications using DGA, changing the color depth is out (but, to be honest, palettes were a horrible hack, and I'm very glad to have seen the last of them -- I can't think of any modern apps that require, say, 8-bit color).
I actually hope this was helpful, but if I was just the unwitting victim of flamebait, I can roll with it.
Actually, it was a troll, but even trolls can spawn useful information.
This strikes me as a silly idea and a move in the wrong direction.
Why does there have to be a wrong direction?
It's trying something new. Either it will work out or it won't (and if it does work out, there will probably have to be revisions to the idea).
There is an *incredible* number of incredibly useful information systems that do not exist that have the potential to exist, now that we have the Internet widely available. They could be the next most important way to exchange information -- someone just has to come up with the system and nurse it. We haven't yet scratched the surface -- we don't have any idea what can be done.
In the past few years, I've seen the rise of:
* MMORPGs -- "virtual reality" with huge numbers of people actually existing in real life, playing, exploring and talking together, without regard for physical location. I have a number of friends that have fanned out across the United States, but can still spend more time together than people they live next door to, just because they have forums to do so now.
* Instant Messaging systems -- A system that grants the ability to contact most people with almost zero delay time, collaborate (pasting text and links), carry on masses of real time conversations at once, etc.
* blogs -- A way to rapidly publish, identify, and propagate new memes, with a reputation system built in (if someone has written good articles before, perhaps they will continue to do so). CNN isn't my sole (or primary) source of interesting information any more, which means that control of information channels is *much* weaker than it was even recently.
* reddit -- collaboratively rated "blog". A truly adaptive "content of interest" stream. IMHO, the next generation beyond just reading RSS feeds of blogs.
* del.icio.us -- collaboratively rated bookmarking, useful for researching a topic quickly.
* Wikipedia -- whether you call it an "encyclopedia" or not, there's no denying that this store of overview-level knowledge on many, many topics is incredibly valuable.
* Freenet -- we have (abeit still not in a particularly Joe-Sixpack-usable package) truly anonymous interaction offered us.
That's just off the top of my head. There are new ideas just bubbling up all over. What's the cost of trying something wrong? Maybe someone insults your idea and you pay some server fees. The Internet is a *long*, *long* way from being a mature environment -- there are new, completely untapped things coming into being every day.
I don't think anyone thinks that Digital Universe is going to be unilaterally better than Wikipedia, but who knows? Maybe it will work, and maybe it will be better in some ways than WP. In any event, is has the ability to feed off Wikipedia, and provides a mechanism to access copyrighted content (whereas WP is limited to public-domain and free-use content).
Seagate drives have always sucked. I've hurled more dead seagate drives down the concrete path than any other brand of hard drive.
I've had both Seagate and WD drives fail.
To be fair, Seagate drives tend to also be more expensive and somewhat smaller than their equivalent competitors -- it may be less that Seagate has some wonderful technology and more that they don't push as hard into the value market. Pay a little bit more for a slightly more conservatively built drive, get more reliability. SCSI-class drives are still too small expensive for me to use as general-purpose drives, but I've very happy inching away from Maxtor and WD's value IDE drives.
There was a point in time when I bought the cheapest, largest hard drives I could. The times when I did that are over. There's a large amount of time involved in reconfiguring the software on a computer, even if you don't lose an iota of irreplacable personal data (unlikely).
Today, I keep a spare hard drive in my system, and run a cron job nightly that mounts it, rsyncs all the important data on the other hard drives to the spare hard drive, and unmounts it. Of course, this means more wear and tear on the drives -- however, when one fails, it's not a huge deal, since I've lost at most a day's work, plus the time to reinstall the software (minimal in human time, on a Linux box, where the whole thing can be automated). In addition, if you're running with the -t flag, the churning that the hard drives undergo is not really worse than their nightly run from updatedb -- only metadata gets examined on old files, and new data gets copied over once. The spare hard drive cost me maybe $120 -- the knowledge that no one hard drive failure can blow away my data makes me sleep *much* better at night.
(I know that RAID has been getting popular in the desktop market -- I've heard of at least as many problems caused by people setting up home RAID setups as problems solved. I like the nightly backup approach.)
Also, this approach lets me roll back to last night's copy of anything, if I accidentally blow a file away.
I wholeheartedly endorse the use of one extra hard drive.
One other nice thing about this approach is that you can stick an extra PATA drive in on the same bus as another drive without worrying about the performance impact (since this thing should only be running when you aren't using the computer).
I always wondered why, when the hard drive is the single most difficult-to-recover-from-in-the-event-of-a-failur e component, that people spend less on hard drives than almost anything else in their computer. Most of the time, a modern CPU is starved for data, but people still keep trying to get the fastest CPU they can afford, and just don't give a damn about data integrity.
I don't see why editing your bio, espcially to correct errors, would be such a terrible crime worthy of news.
Because Wikipedia is cool and works well; it's hard to bash it.
It's much easier for WP's detractors to use ad hominem attacks against people involved with WP. Dig up a point where Wales violated WP TOS? Post it all over!
...while the fundementalist churches that have created much of the terror over the past several years are free to break the laws and violate the tax codes.
I wonder what the world would be like without said churches. I mean, if I had a button that would eliminate Christianity from the world, I'd probably just push it, but we don't really have hard evidence that the world would be better off in such a case.
It could be that religion does important things that are non-obvious that outweigh the obvious bad things.
If we don't spy on everyone, have the terrorists won?
Frankly, I don't really give a damn whether or not Al Qaeda wins or not (i.e. establishes Islamic goverments throughout Middle East -- Bush gave him a real head start by removing the secular Hussein -- and kicks US bases and military force out of the area).
What I care about is whether or not *I lose*.
I clearly *do* lose if I lose Constitutional liberties.
Also, if you read the article, the problem is about building a better model to catch people who want to do harm.
Hitler also reduced civil liberties in the name of protecting people from terrorists.
The question is whether the benefits of your more effective model are worth the costs.
I'd also like to point out that if, indeed, Clinton, Carter, or Reagan did something unconstitutional, that certainly does not mean that Bush can also do unconstitutional things.
The reason slavery is unacceptable today is because the vast majority of the world believes that it is wrong.
The reason slavery is unacceptable today is because of geeks producing advances in technology.
Slavery started up when we stopped hunting and gathering and started staying in one place and needed masses of manpower to do agricultural work.
Then the techies of the past, the Eli Whitneys and the Cyrus McCormicks, improved agricultural efficiency (in the US, 84% of the population in 1810 was involved in agricultural work. Today only 1.9% is. The Industrial Revolution made manufacturing a lot more relatively valuable than agriculture. Combine those two factors. Now you have a big drop in demand for unskilled muscle power, and keeping a bunch of slaves around to do manual labor just doesn't seem to be all that good of an idea.
The morality of societies adapt wonderfully to the times. For example, polygamy makes a *lot* of sense in an environment where you have lots of hand-to-hand wars in which lots of men are killed. You don't need lots of men to retain your population's reproduction rate (which is a big chunk of why men traditionally composed armies) -- the number of women available is the bottleneck. Now you have a bunch of war widows. If you marry then off to some of the surviving men, you don't lose your next generation of children (and the society that chooses to do that is asking for trouble in one generation, when the guys in the next city-state are more numerous and looking for land).
The real knights in shining armour, the ones going out and making everyone wealthier and happier, are the ones pushing technology ahead. It's easy to forget that. Your graph theory paper fights slavery!:-)
the real answer would be to fully ignore this bullshit.
It's not even just this furor -- this is just the present set of claims about why WP doesn't work.
I use Wikipedia many times a day. I consider it as important as Google. I see tons of posts on Slashdot from people bitterly criticizing Wikipedia. All I can say is, it works. Surprisingly so, to me, but it does really work. Maybe at some point in the future it will stop, but right now, it's great.
I remember a period of time when people like kelkoo were managing to spam the bajeezus out of Google. There were many people on Slashdot saying that Google had lost its value, how everyone should switch to an alternate search engine, etc. Uh, huh. If that's the case, people will figure it out themselves -- you don't need to keep hollering at them.
I'm sure that Wikipedia will evolve over time, and maybe someone will fork it with some different design ideas, and that fork will win out. But the people claiming that WP is not useful are just *wrong*. You can always find some article on WP that is incorrect, but you'd have to ignore the vast quantities of useful, well-written information. I've read more history in the past year on WP than I ever thought I'd read in a lifetime -- unlike most of the history classes I'd taken in the past, WP is facinating and allows one to easily dig for more information.
I personally think that it's because so much computer security theory is based around the idea of preventing any exploits or attacks at all, instead of around survivability, and that really bugs people who normally work on computer security. It drove me nuts -- I've spent time doing P2P design, and at first all I could think about was what appeared to me to be gaping holes in Wikipedia's functioning. Anyone can vandalize almost *anything* on Wikipedia! There are so many subtle ways to attack it! There's so much of the fallible human element involved! And yet...Wikipedia works. Clearly, my model of the way such a system needed to work in order to be useful was wrong -- Wikipedia wasn't what was wrong. I had undervalued survivability, because in the past, systems that I'd looked at that had allowed attacks had simply *failed*. Wikipedia doesn't.
The environment is always changing, and I'm sure that Wikipedia will evolve with it, and forks of Wikipedia will probably explore different ideas. Wikipedia is a potential source of more social and informational research than I can even begin to imagine. The point is, though, Wikipedia simply is not the dead-end road that it seemed to be when I first glanced at it -- and I think that many other people are making the same error that I was upon first seeing it.
My argument here isn't going to help or hurt WP. If something is genuinely useful, people will flock to it in the long term, and if it becomes not useful, people will leave. However, I think that the reasons that people criticize WP so heavily are interesting and worthy of discussion.
Wikipedia has collected more facts over time, but it reads worse.
I disagree.
The *average quality of new content* may be lower.
However, the quality of any given *existing* content rarely drops (other than through vandalism). The quality of existing content nearly monotonically rises.
All you're saying is that content that previously *wasn't covered at all* is not yet up to the level that mature articles have (and, possibly, articles once started with with the original group of Wikipedians).
So if your test for the quality of Wikipedia is to choose a random article and examine how well the article reads -- yes, you may measure a drop in quality over time. However, the alternative would be for those masses of added articles to not exist at *all*. If you look at just existing articles and examine whether they are getting better or worse, I would say that they are definitely improving.
I'd like to thank you for the time and talent you've devoted to Wikipedia. It it is a wonderful resource, and competes with Google as one of the most helpful systems on the Internet for obtaining useful information quickly.
At some point, Microsoft needs to stop being so ridiculously anti-open source if they are to survive. They aren't going to win "the battle against the GPL", no matter how big they are. Not only is it not winnable, it's not their fight. Every other major tech company has long since been planning and implementing strategy to coexist with open source, and Microsoft is *still* trying to figure out how to subvert or attack open source. It's just silly.
It's nice to see a few Microsoft people collaborating with open source folks in a (no matter how minor) way to help everyone come out ahead. For once, I can see working with Microsoft without worrying about a knife in the back or a poison pill. The problem is that every time in the past that Microsoft claimed to be working on something "open", they were just trying to figure out some way to screw people over -- with their patent-encumbered SPF alternative, say. It makes people *very* nervous about giving Microsoft an inch.
It's also really nice, no matter how minor to see someone saying something polite and honest in a blog about what is ultimately a competitor. One of the nicest things about open source is that you have a big group of people being open about what they're doing, and often critiquing their project -- everyone in the world can read lkml, and there's often harsh criticism of components of Linux on it. If folks at Microsoft can be honest and say "Hmm, this person was doing something well -- we are going to learn from them", I'm much more inclined to think that they're doing a good job.
Slashdot Mirror
Can you tell me a track or two where I might hear the difference?
:-)
:-)
.wav back and forth (not blind) many times to try to learn the differences first. While I was able to teach myself to recognize the compressed audio, in an environment where I hadn't just done this, I doubt I'd be able to say "Aha! That's Vorbis-compressed audio!" if I just heard a random clip compressed with something.
I don't remember what my friend and I were using (actually, *he* was running the test for himself -- it was his CD, and I just happened to drop in and try it as well). It was something with plenty of percussion -- drums and cymbals. IIRC, and this was a while ago, the most noticeable MP3 artifact happened in cymbals (you can also hear this in white-noise type stuff). You hear...darn, I can't think of anything like it, but you hear "swishing" sounds in the cymbals. With Vorbis, the drums sounded a tiny bit flatter (the flatter drums didn't bother me, but the cymbals were annoying).
I wonder if it was my speakers that made the difference.
I've heard that it's much easier to hear minor things like that in headphones than in speakers -- I wouldn't know, since I don't own any nice speakers -- headphones are much cheaper.
And Radio Shack headphones? What kind of audio snob are you?
A bad one, I reckon.
Both of us were surprised that the artifacts could be detected at that bitrate, FWIW. My friend couldn't pick up on the Vorbis artifacts, but could manage the MP3 artifacts.
I'm a little suspicious that it isn't anything about the "quality of the ears", but just a matter of training yourself. I played the MP3, the Vorbis and the
Did you do the test?
.wav and .mp3 consistently at 256 KBps CBR (encoded with LAME). Cymbals started to sound "swishy" -- they start to pick up unpleasant patterns.
.wav and MP3 at a lower bitrate (but I don't remember the precise results, and in any case, that wasn't a double-blind test).
A double-blind test, yes (A friend ran a program that randomly played a clip, then I called out what I thought, and then he told me what it was). I had decent headphones, but nothing amazing from an audiophile standpoint (about $70, Radio Shack, closed circumaural), and I'm certainly no musician. The test clip I was using had drums and cymbals. I could tell the difference between
With Vorbis, I could tell the difference all the way up to the maximum quality setting (to my surprise). With Vorbis, cymbals didn't sound different (to my ear, at least), but percussion sounds slightly different -- a little bit flatter, maybe.
I still preferred Vorbis, because the MP3 artifacts sounded *unpleasant* -- swishy cymbals sound bad -- but Vorbis's artifacts don't.
Note that I was using CBR with LAME (because at the time I was interested in CBR results), though that isn't what I'd use in real life.
When listening on my little computer multimedia speakers (about $25) I've found that I can't distinguish between
Note that this was two years or so ago, and both encoders may have improved since then.
MagnaTune: returning guilt to life.
How about 7-Zip? Open Source, Free, supports more formats than WinZip, nice shell extension menu interface...
When e-mail clients can match that functionality, allowing me to follow a thread of e-mail messages as easily as following a thread on Usenet, I'll find a reason to switch. When I can manually reorganize message threads (so that messages with different subjects are included), I'll be even happier. You guys at Google listening?
Lots of mail clients thread (if the option to do so is enabled -- most can also not thread). Mutt does. Even Outlook does.
What I can't hear is the difference between a properly done 192kbps/44khz mp3 rip and one made with FLAC.
:-)
Try something with cymbals.
Also, while FLAC may not be useful for a portable music player (especially in a typically noisy environment, with those little earbud things), my desktop computer has ungodly amounts of space available -- FLAC makes a lot of sense for that.
JPEG artifacts may not be visible in normal images under typical viewing conditions, and JPEG may be the most suitable format for photographic images in web pages, but nobody is pushing to replace TIFF with JPEG.
I *would* be interested in knowing why there's so little support for Vorbis -- I can understand on, say, the little value-oriented devices that use ASIC MP3 decoders, but the iPod has twin ARMs, and should be able to handle Vorbis.
I just crossed from office-geek into "don't stare and walk away slowly" dude.
I don't think that it's really that weird, unless you think of "anyone trying something new" as weird. Automobiles and electric lighting were all new at one point...
I suspect that the enjoyment of a media is greater when you have more related stimuli firing that your brain can tie together.
People like bass. I doubt that this is because low notes are just really neat, but because they can *feel* the bass -- vibration and air pressure differences on their skin. A chair that has an embedded subwoofer could produce some of this effect.
Video games that have audio triggered by various in-game events, producing a synchronized audio-video simuli, have stomped silent video games.
There have been some attempts at games that produce even more synchronization between audio and video. A number of games now have not merely sound effects triggered by game events, but adaptive music. The game state affects the type of background music playing -- Lumines, Total Annihilation, all kinds of newer games. Your brain gets to tie together finger actions with music.
There have even been some video games which alter game state based on audio -- for example, Vib-Ribbon and Rez have game events triggered in synchronization with the audio as *well* as audio that adapts itself to game events (and, for that matter, outside of the US, Rez had a vibrator that one could hook up to the game via USB to provide tactile feedback). Now, Rez didn't sell all that well, but I strongly suspect that this was because it was too unusual for people to try it out -- the people I know that *do* play Rez are generally pretty rabid about it. The concept of associating more stimuli to produce a more enjoyable experience generally seems to work pretty well. The same approach seemed to work well for Dance Dance Revolution.
I've seen a lot of game designers take the idea of tying more forms of stimuli together, but they often just seem to think something along the lines of, "I should make a game based on music", and then the game doesn't really rise above anything other than a game oriented around music. There's no real reason why you couldn't take *any* of the types of games that currently have background music (i.e. almost all of them) and simply synchronize various game events to the music.
I wonder how well a full-body suit would sell, something that provides tactile stimuli that could be synchronized with audio. I mean, yes, it would be expensive, but I also see vendors for audiophiles selling multi-thousand-dollar cables, amplifiers, and headphones to try to reproduce that sensory experience crafted by the musician just a *little* more accurately -- surely, just adding more stimuli to the mix would be more worthwhile.
I wonder when the RIAA (which dearly loves moving to new formats and forcing everyone to buy their audio over again) will try shipping music in a new format, where it can contain video tracks or tactile tracks. If they want to be less ambitious, at least lighting tracks -- coming up with a standard where you have, say, eight lights each containing R, G, B elements, and allowing them to fade or flash to various colors in time to the music would seem to be a pretty cheap and compelling way to get people to purchase new music all over again.
WinAMP and various computer-based audio players try to automate the above via beat detection and use extracted data from the audio stream as input to a video image. However, as people who have tried this out know, not only is automated realtime beat and rhythm detected far from perfect, but elements of the music that are quite pe
My question is how does the money raised by these levies find its way to the copyright holders? (Artists, publishers and so on.)
Basically, all the people not signed by an RIAA label get a net 4% of the goods taken by blank media taxes (they don't have RIAA lobby dollars working for them). RIAA-signed artists get a total of 38.4% of the take, and the RIAA member publishers get 57.6% of the take. Note that this is a description from the RIAA, so that 38.4% may potentially be siphoned off into that 57.6% via fees orwhat-have-you.
It's irritating knowing that a percentage chunk of sales of many computer storage devices goes to lobbyists working to punch holes in free use, and even more irritating knowing that despite the fact that I have to pay for potential infringement, I *still* don't get to infringe legally.
You know...every time the RIAA complains that they need to be around to fund artists, and it's important that they exist to do so, and if they didn't get federal protection in the form of funds...I wonder what would happen if *100%* of that tax went to artists -- publishers not included. That *would* fund artists, presumably cutting out the middleman. Never going to happen, but an fun thought.
Another interesting idea -- an artist can choose to be supported by blank media sales *but* need to place all their work into the public domain *or* be supported by regular retail sales but not get any blank media sale funding.
And if your friend had his maps stolen by an African tribesman, could he successfully navigate by the stars?
It's all relative.
That being said, the reason so damn much high tech crap goes to the military is because "national security" is a great word to get ahold of federal money to siphon off to companies (defense contractors, construction types, etc) in your state.
Consider the F-22 Raptor. The F-22 has a unit cost of somewhere above $150 million. Now, I recognize that technology is great and all, but you have to wonder whether the F-22 could really maintain the 10:1 kill ratio against an F-16 or something that costs in the $15 million range. Yes, it's got doohickies and all that, but who are you planning to fight with this thing? Who do you need to out-technology? Tribesmen in Africa? Rebels in Iraq? Hell, at some point your simple crashes-due-to-accident becomes significant.
I think that every time I read about recent US military operations (Iraq, Somalia), there are problems with the sheer number of incompatible communication networks (everyone wants to develop their own network -- lots of $$$).
Identifying the scope of the commercial versus the open source problem is no easier than discerning real data on p2p usage.
Ah, yes. Those problematic open source P2P authors.
Mr. Garfield, I like hacking on P2P software. You can sic attorneys on every visible open source P2P author, and all that will happen is you will drive people underground -- and you don't need much of an underground to write all the software that anyone could ever use. You aren't going to manage to stop the production of open source P2P software.
Perhaps you'd like to look at Microsoft, Mr. Garfield. Microsoft has greater annual revenues than *all your member studios put together*. Microsoft has *clout*.
Microsoft wasn't able to quash open source development, despite spending an awful lot of money and effort on it, Mr. Garfield. I'm going to give you *very* slim chances of succeeding where they failed.
What are you going to try? PR? Microsoft did that. They called Linux a virus. They said it exposed users to liability. They said that it was insecure, and that it was *communist*, Mr. Garfield. It didn't work.
How about legislation? Maybe, if you're *really* lucky, you can manage to pay enough legislators to vote in laws criminalizing the production of software that is used to cause greater than some degree of purported damages. I don't think that you can manage that -- you'd face opposition from a lot of tech types, and a number of legislators have noticed that people don't *like* stories in the newspaper about nine-year-old girls being sued for thousands of dollars. But let's say that, despite all that, you manage it. There are a *lot* of open source programmers overseas, Mr. Garfield, and software does not understand national boundaries. The US government made export of encryption code illegal due to national security concerns for a long time. What happened? Encryption development and distribution continued, from overseas. It didn't do any good. You can't quash software development.
You going to try to track down all the people copying software and music and movies down? Mr. Garfield, one of the primary functions of a computer is to reproduce and distribute data quickly and accurately. There is *huge* demand for this, demand which far exceeds and outweighs any demand for entertainment. They have a device which does *exactly* what you don't want. There are *too many people* that want to be able to copy around movies for this to work.
How about a technical solution, Mr. Garfield? You spent plenty of effort trying to lock up DVDs -- that didn't work (you excluded Linux from your supported platforms, which was pretty stupid and put a lot of very smart Linux-using techies and crypto types to work on the problem, but even if you hadn't, it wouldn't have lasted long). You want to try again? Well, there are a lot of security types who would love to take your money and can guarantee you the moon, but it isn't going to happen.
You want to try keeping digital data from becoming analog? Good luck.
You want to try keeping analog data from becoming digital? This is a new, interesting one. You're now trying to plug a hole that requires *one* person with *one* analog-to-digital encoding device somewhere in the *world* per movie. It makes no more sense than trying to use CSS to keep people from getting at DVD content. It's just not a feasible approach.
I know that this is a really appalling concept, and one that you probably don't want to entertain. But it is possible -- just possible -- that your only solution is to reduce costs to where the convenience and guaranteed quality of buying your product from you outweighs the inconvenience of pirating. That means that you have to trim all your excess fat. That means that maybe you can't spend hundreds of millions of dollars producing and marketing a movie. Maybe you can't *have* actors that get tens of millions of dollars for every work. Maybe you need to use CG, and can't afford to recompen
You also clearly do lose if they blow up your home city.
Uh, huh.
Or if a meteorite hits me.
Thus far, the rate of meteorites hitting humans (one on record) has exceeded that of terrorists blowing up cities (none on record), and while you can paint a terribly heartstring-tugging picture using either device, I'm not worried about either.
And don't think the terrorists want to co-exist with us if we just let them rule the Middle East.
"Terrorists" rule the Middle East?
You mean Arabs?
Or perhaps Muslims?
The people running around blowing things up are largely people pissed off because they *don't* have political power other than by blowing things up. Are Palestinian militants "ruling" Israel?
They tolerate no one except themselves...
Ah. You seem to know a hell of a lot about these terrorists. You spend time with them? Have you been active in Middle Eastern politics? Or are you just talking out your ass because Fox News keeps telling you absurd scary things about them because terrorism is currently politically useful?
To be sure, as I have said above, piracy is not the whole problem - industry practices are part of the problem as well, but it is part of the problem. So what should we do about it?
I am inclined to oppose legislation unless we are absolutely certain that there is no technical solution that can come to the fore, or the market can simply be reshaped (for example, musicians getting tips or movies being sold in *extremely* large files -- so large that transfer of them is prohibitive). Or maybe competition finally driving ISPs to not allow unlimited data transfer any more, making P2P subsidized by lighter users less feasible.
The only reason to introduce legislation is to preserve the content distribution industry. While there is some cost to turmoil in the content distribution industry, I would point out that generally, if technology obsoletes an industry (such as the horse and carriage being obsoleted by the car), it's just not very feasible to try to extend the lifetime of the industry through legislation.
Traditionally, the expense of music production came in the distribution phase. The moneymaking step thus was attached to distribution, which led to music being produced based on funds provided by a music distributor (who thus assumed some financial risk). Getting the word out about good music was very hard, and so huge advertising budgets were worthwhile.
Today, we have very good systems for finding new music or movies that we like (and those systems get better by the day). IMDB, mp3.com, MovieLens, and so forth. This means that the marketing service that is tied into distribution is rapidly losing value. What about the distribution services? Well, it turns out that not only is distribution getting a lot easier and cheaper (via electronic distribution), but there is no initial cost to press a huge run and ship them to retail stores. That kind of eliminates much of the economic role of the music distributor as a necessary element in music production. Naturally, these industries are going to do what they can to protect themselves -- this is hardly cold and unfeeling, as Sally Smith, the record marketing directory for Acme Audio, wants to keep being able to buy Christmas toys for her kids. Ultimately, however, I'm not convinced that traditional distributors are still necessary for music production and consumption to occur.
I don't like the approach that this Conyers gentleman is aiming for. He's saying "well, if you don't like it, give me a better solution, and in the meantime, we're going to create this crummy law."
Frankly, the local live jazz groups don't need the RIAA to function. They're doing just fine. I'm not sure why I need to subsidize the RIAA's continued existence with my tax dollars and in increased prices for various electronic goods and in reduced functionality.
What I want to know is if they are planning on [...] adding features like sound support,
Sound support is handled by a sound server, which fortunately runs independently of X.
While you can certainly set up network-remote sound, few people do. It just turns out that there are incredibly few applications where sound is actually used -- most of 'em are things like media players and games. With games, you frequently want to run the thing locally for performance reasons, and with media player, it's just more efficient to access the (generally compressed) media files over the network than it is to decompress the audio and stream it decompressed over the network.
Resolution changes don't require a reboot, just a restart of X.
I think that he was talking about changing the pixel dimensions of the desktop. You can cycle through all the auto-detected or specified resolutions in modern X (control-alt-keypad+ or -keypad-). If you don't want to have the desktop scroll with your mouse, try xrandr -s 640x480 or similar.
Except for full-screen applications using DGA, changing the color depth is out (but, to be honest, palettes were a horrible hack, and I'm very glad to have seen the last of them -- I can't think of any modern apps that require, say, 8-bit color).
I actually hope this was helpful, but if I was just the unwitting victim of flamebait, I can roll with it.
Actually, it was a troll, but even trolls can spawn useful information.
This strikes me as a silly idea and a move in the wrong direction.
Why does there have to be a wrong direction?
It's trying something new. Either it will work out or it won't (and if it does work out, there will probably have to be revisions to the idea).
There is an *incredible* number of incredibly useful information systems that do not exist that have the potential to exist, now that we have the Internet widely available. They could be the next most important way to exchange information -- someone just has to come up with the system and nurse it. We haven't yet scratched the surface -- we don't have any idea what can be done.
In the past few years, I've seen the rise of:
* MMORPGs -- "virtual reality" with huge numbers of people actually existing in real life, playing, exploring and talking together, without regard for physical location. I have a number of friends that have fanned out across the United States, but can still spend more time together than people they live next door to, just because they have forums to do so now.
* Instant Messaging systems -- A system that grants the ability to contact most people with almost zero delay time, collaborate (pasting text and links), carry on masses of real time conversations at once, etc.
* blogs -- A way to rapidly publish, identify, and propagate new memes, with a reputation system built in (if someone has written good articles before, perhaps they will continue to do so). CNN isn't my sole (or primary) source of interesting information any more, which means that control of information channels is *much* weaker than it was even recently.
* reddit -- collaboratively rated "blog". A truly adaptive "content of interest" stream. IMHO, the next generation beyond just reading RSS feeds of blogs.
* del.icio.us -- collaboratively rated bookmarking, useful for researching a topic quickly.
* Wikipedia -- whether you call it an "encyclopedia" or not, there's no denying that this store of overview-level knowledge on many, many topics is incredibly valuable.
* Freenet -- we have (abeit still not in a particularly Joe-Sixpack-usable package) truly anonymous interaction offered us.
That's just off the top of my head. There are new ideas just bubbling up all over. What's the cost of trying something wrong? Maybe someone insults your idea and you pay some server fees. The Internet is a *long*, *long* way from being a mature environment -- there are new, completely untapped things coming into being every day.
I don't think anyone thinks that Digital Universe is going to be unilaterally better than Wikipedia, but who knows? Maybe it will work, and maybe it will be better in some ways than WP. In any event, is has the ability to feed off Wikipedia, and provides a mechanism to access copyrighted content (whereas WP is limited to public-domain and free-use content).
Seagate drives have always sucked. I've hurled more dead seagate drives down the concrete path than any other brand of hard drive.
r e component, that people spend less on hard drives than almost anything else in their computer. Most of the time, a modern CPU is starved for data, but people still keep trying to get the fastest CPU they can afford, and just don't give a damn about data integrity.
I've had both Seagate and WD drives fail.
To be fair, Seagate drives tend to also be more expensive and somewhat smaller than their equivalent competitors -- it may be less that Seagate has some wonderful technology and more that they don't push as hard into the value market. Pay a little bit more for a slightly more conservatively built drive, get more reliability. SCSI-class drives are still too small expensive for me to use as general-purpose drives, but I've very happy inching away from Maxtor and WD's value IDE drives.
There was a point in time when I bought the cheapest, largest hard drives I could. The times when I did that are over. There's a large amount of time involved in reconfiguring the software on a computer, even if you don't lose an iota of irreplacable personal data (unlikely).
Today, I keep a spare hard drive in my system, and run a cron job nightly that mounts it, rsyncs all the important data on the other hard drives to the spare hard drive, and unmounts it. Of course, this means more wear and tear on the drives -- however, when one fails, it's not a huge deal, since I've lost at most a day's work, plus the time to reinstall the software (minimal in human time, on a Linux box, where the whole thing can be automated). In addition, if you're running with the -t flag, the churning that the hard drives undergo is not really worse than their nightly run from updatedb -- only metadata gets examined on old files, and new data gets copied over once. The spare hard drive cost me maybe $120 -- the knowledge that no one hard drive failure can blow away my data makes me sleep *much* better at night.
(I know that RAID has been getting popular in the desktop market -- I've heard of at least as many problems caused by people setting up home RAID setups as problems solved. I like the nightly backup approach.)
Also, this approach lets me roll back to last night's copy of anything, if I accidentally blow a file away.
I wholeheartedly endorse the use of one extra hard drive.
One other nice thing about this approach is that you can stick an extra PATA drive in on the same bus as another drive without worrying about the performance impact (since this thing should only be running when you aren't using the computer).
I always wondered why, when the hard drive is the single most difficult-to-recover-from-in-the-event-of-a-failu
I don't see why editing your bio, espcially to correct errors, would be such a terrible crime worthy of news.
Because Wikipedia is cool and works well; it's hard to bash it.
It's much easier for WP's detractors to use ad hominem attacks against people involved with WP. Dig up a point where Wales violated WP TOS? Post it all over!
...while the fundementalist churches that have created much of the terror over the past several years are free to break the laws and violate the tax codes.
I wonder what the world would be like without said churches. I mean, if I had a button that would eliminate Christianity from the world, I'd probably just push it, but we don't really have hard evidence that the world would be better off in such a case.
It could be that religion does important things that are non-obvious that outweigh the obvious bad things.
If we don't spy on everyone, have the terrorists won?
Frankly, I don't really give a damn whether or not Al Qaeda wins or not (i.e. establishes Islamic goverments throughout Middle East -- Bush gave him a real head start by removing the secular Hussein -- and kicks US bases and military force out of the area).
What I care about is whether or not *I lose*.
I clearly *do* lose if I lose Constitutional liberties.
When you make an international call, it is not private and you can have no real expectation of privacy.
Fine. Why can't I legally use end-to-end encryption, then?
Also, if you read the article, the problem is about building a better model to catch people who want to do harm.
Hitler also reduced civil liberties in the name of protecting people from terrorists.
The question is whether the benefits of your more effective model are worth the costs.
I'd also like to point out that if, indeed, Clinton, Carter, or Reagan did something unconstitutional, that certainly does not mean that Bush can also do unconstitutional things.
The reason slavery is unacceptable today is because the vast majority of the world believes that it is wrong.
:-)
The reason slavery is unacceptable today is because of geeks producing advances in technology.
Slavery started up when we stopped hunting and gathering and started staying in one place and needed masses of manpower to do agricultural work.
Then the techies of the past, the Eli Whitneys and the Cyrus McCormicks, improved agricultural efficiency (in the US, 84% of the population in 1810 was involved in agricultural work. Today only 1.9% is. The Industrial Revolution made manufacturing a lot more relatively valuable than agriculture. Combine those two factors. Now you have a big drop in demand for unskilled muscle power, and keeping a bunch of slaves around to do manual labor just doesn't seem to be all that good of an idea.
The morality of societies adapt wonderfully to the times. For example, polygamy makes a *lot* of sense in an environment where you have lots of hand-to-hand wars in which lots of men are killed. You don't need lots of men to retain your population's reproduction rate (which is a big chunk of why men traditionally composed armies) -- the number of women available is the bottleneck. Now you have a bunch of war widows. If you marry then off to some of the surviving men, you don't lose your next generation of children (and the society that chooses to do that is asking for trouble in one generation, when the guys in the next city-state are more numerous and looking for land).
The real knights in shining armour, the ones going out and making everyone wealthier and happier, are the ones pushing technology ahead. It's easy to forget that. Your graph theory paper fights slavery!
the real answer would be to fully ignore this bullshit.
It's not even just this furor -- this is just the present set of claims about why WP doesn't work.
I use Wikipedia many times a day. I consider it as important as Google. I see tons of posts on Slashdot from people bitterly criticizing Wikipedia. All I can say is, it works. Surprisingly so, to me, but it does really work. Maybe at some point in the future it will stop, but right now, it's great.
I remember a period of time when people like kelkoo were managing to spam the bajeezus out of Google. There were many people on Slashdot saying that Google had lost its value, how everyone should switch to an alternate search engine, etc. Uh, huh. If that's the case, people will figure it out themselves -- you don't need to keep hollering at them.
I'm sure that Wikipedia will evolve over time, and maybe someone will fork it with some different design ideas, and that fork will win out. But the people claiming that WP is not useful are just *wrong*. You can always find some article on WP that is incorrect, but you'd have to ignore the vast quantities of useful, well-written information. I've read more history in the past year on WP than I ever thought I'd read in a lifetime -- unlike most of the history classes I'd taken in the past, WP is facinating and allows one to easily dig for more information.
I personally think that it's because so much computer security theory is based around the idea of preventing any exploits or attacks at all, instead of around survivability, and that really bugs people who normally work on computer security. It drove me nuts -- I've spent time doing P2P design, and at first all I could think about was what appeared to me to be gaping holes in Wikipedia's functioning. Anyone can vandalize almost *anything* on Wikipedia! There are so many subtle ways to attack it! There's so much of the fallible human element involved! And yet...Wikipedia works. Clearly, my model of the way such a system needed to work in order to be useful was wrong -- Wikipedia wasn't what was wrong. I had undervalued survivability, because in the past, systems that I'd looked at that had allowed attacks had simply *failed*. Wikipedia doesn't.
The environment is always changing, and I'm sure that Wikipedia will evolve with it, and forks of Wikipedia will probably explore different ideas. Wikipedia is a potential source of more social and informational research than I can even begin to imagine. The point is, though, Wikipedia simply is not the dead-end road that it seemed to be when I first glanced at it -- and I think that many other people are making the same error that I was upon first seeing it.
My argument here isn't going to help or hurt WP. If something is genuinely useful, people will flock to it in the long term, and if it becomes not useful, people will leave. However, I think that the reasons that people criticize WP so heavily are interesting and worthy of discussion.
Wikipedia has collected more facts over time, but it reads worse.
I disagree.
The *average quality of new content* may be lower.
However, the quality of any given *existing* content rarely drops (other than through vandalism). The quality of existing content nearly monotonically rises.
All you're saying is that content that previously *wasn't covered at all* is not yet up to the level that mature articles have (and, possibly, articles once started with with the original group of Wikipedians).
So if your test for the quality of Wikipedia is to choose a random article and examine how well the article reads -- yes, you may measure a drop in quality over time. However, the alternative would be for those masses of added articles to not exist at *all*. If you look at just existing articles and examine whether they are getting better or worse, I would say that they are definitely improving.
I'd like to thank you for the time and talent you've devoted to Wikipedia. It it is a wonderful resource, and competes with Google as one of the most helpful systems on the Internet for obtaining useful information quickly.
At some point, Microsoft needs to stop being so ridiculously anti-open source if they are to survive. They aren't going to win "the battle against the GPL", no matter how big they are. Not only is it not winnable, it's not their fight. Every other major tech company has long since been planning and implementing strategy to coexist with open source, and Microsoft is *still* trying to figure out how to subvert or attack open source. It's just silly.
It's nice to see a few Microsoft people collaborating with open source folks in a (no matter how minor) way to help everyone come out ahead. For once, I can see working with Microsoft without worrying about a knife in the back or a poison pill. The problem is that every time in the past that Microsoft claimed to be working on something "open", they were just trying to figure out some way to screw people over -- with their patent-encumbered SPF alternative, say. It makes people *very* nervous about giving Microsoft an inch.
It's also really nice, no matter how minor to see someone saying something polite and honest in a blog about what is ultimately a competitor. One of the nicest things about open source is that you have a big group of people being open about what they're doing, and often critiquing their project -- everyone in the world can read lkml, and there's often harsh criticism of components of Linux on it. If folks at Microsoft can be honest and say "Hmm, this person was doing something well -- we are going to learn from them", I'm much more inclined to think that they're doing a good job.