Most of the examples you gave seem more like "Less restrictive copyright helps" than no copyright. Give your songs for free! The money is on the tour... Of course, as long as another band doesn't do -exactly- the same songs with a bigger marketing budget (and if everyone does it, ONE of the bands who copy you will most likely be bteter). Also, thats as long as the video of your show isn't in hi definition 7.1 surround blu ray the day after it for free (or even worse, SOLD by someone else). With absolutely -zero- copyright, its a lot less powerful as a promotion tool. (Now it works because you're only letting indiviuals step in... once corporations can rape your copyright too, things get a little grim). Oh, and without IP laws, people can rip off your name, your logo, everything, and not only sell it as free promotion to you... but make it -theirs- and use it for -themselves-.
Absence of copyright doesn't necessarily mean freedom to take credit for what someone else creates -- a right of attribution could be separated from a right to restrict copying.
For the photographer... yes, giving their pictures away is great if they get their money from hired work (all photographers I know do this). But lets remove all IP laws for a sec there... The photographer's customers aren't going to be too too happy once their FACE (your pictures that you, not only gave away, but couldn't restrict in any ways, shape and form) is on a box of cereals or used on TV to advertise condoms.
This is neither here nor there, for a couple of reasons. It isn't copyright that prevents that from happening; it's more of a privacy right. People who want to use images of recognizable people commercially need model releases in at least certain circumstances; the model release is essentially saying "you have a right to use my likeness for this purpose". But this isn't a general purpose copyright, and people don't have a copyright on how they look -- it's perfectly legal to imitate someone's appearance (at least as long as it's not for a fradulent purpose), and it's likewise legal to photograph someone.
The other reason this is irrelevant is that usually the photographer, not the client, holds the copyright on the images. That's why if you want copies of the photographs from your own wedding you usually have to go back to the original photographer and pay outrageous amounts of money for prints -- if the photographer has kept the negatives or files (usually they don't hold on to them for very long). There are photographers who will either work for hire (as opposed to being contractors), in which case the person hiring the photographer gets the copyright, but those are the exceptions (I think -- hopefully I'm wrong about that, but I don't think so). So absent a contractual agreement (which trumps copyright law), and aside from the model release issue, the photographer, holding the copyright, could still sell your wedding images in this way, but you yourself couldn't!
As Richard Stallman likes to say, we can't talk about "intellectual property" as one thing; there are copyrights, patents, trademarks, trade secrets, all of which have very different characteristics. I suppose we could add privacy rights and right of attribution as other forms of "intellectual property". Personally, I would like to eliminate copyright and patent outright (find other solutions for things like medicines -- outright subsidies come to mind as one possibility, which I actually think is more market-based than patents). Trade secrets on things that really are secret are more reasonable, as long as we don't get into the emperor has no clothes situation where companies claim that what they sell you has trade secrets, even though they're right out in the open for you to figure out. Trademarks are reasonable as non-functional means of identification, but again it shouldn't be possible to leverage them into a functional role (a game console refuses to run games that don't have the trademark embedded, and they hold onto that monopoly by enforcing the trademark).
You're basically advocating the service as opposed to selling position that a lot of people on slashdot have been advocating in the software sector for years. While it may work for you I don't think it can work on a universal level. Why would someone hire you to ghost write a book if they can't get any profit from the actual sale of that book?
Same reason that most people work for specified wages or salary -- they're paid money for it, upfront (maybe in arrears for a short time, but they do get paid).
If someone busts his ass producing something, he has the right to determine what to do with it. If he wants to give it away, fine. If he wants to restrict it a million different ways, fine. It's his work, so it's his choice.
Let's be a bit more precise about what we're talking about. The issue isn't what the hypothetical band decides what to do with it; I don't believe there's much controversy about that. I don't think that anyone is arguing that the band must make copies for everyone who wants one. What the issue is is whether, and to what extent, the band may restrict other people from doing what they want with it.
This error is very important; when someone says "a person should have a right to do whatever they want with what they create!" it sounds very obvious. When it's instead stated as "a person should have a right to restrict what others may do with their own copy of something because he created it" it doesn't sound so obvious. If you buy a computer or a car outright, does/should the seller have any post-sale claim on the item? Why should it be different in the case of an intangible item?
Now, if you want to argue that allowing unlimited downstream copying impairs the market, or de facto makes it impossible to sell something for the desired price, that may be true -- but it isn't the same thing as not having a right to ask whatever price you want. A legal right to offer something for sale doesn't mean that someone else has to pay that price, if they can get it elsewhere for less.
I'm not pleased to see a decent and honorable person like Gore lampooned for something he never even said. He received a Webby in 2005 for lifetime achievement, and Vint Cerf has also defended him.
Not to mention deliberately leaving his wireless network open for "convenience" (although securing a wireless network isn't all that hard in Linux) and so that he can share connections with neighbors. Deliberately running an open endpoint should mean having to take at least some responsibility for what flows through it, unless you're a common carrier and really are in the business of supplying pure bandwidth to your customers.
I wonder if this person would appreciate being spammed through someone else's open relay.
So Ron Paul, who is (or at least claims to be) a libertarian, attracts "traditional social conservatives", who are primarily interested in having the government enforce their own beliefs on others about what consenting adults should or should not be allowed to do privately? Anybody else see anything wrong with this picture?
But that's what this whole thing is about -- AutoDesk has been sending DMCA (i. e. copyright) takedown notices to eBay over Mr. Vernor's attempts to sell copies of AutoDesk whose license he never agreed to in the first place.
The First Sale Doctrine *is* also law, and contracts and agreements do have to follow the law.
Mr. Vernor never installed (those copies, at any rate) of AutoCAD, and so never agreed to any license. If the company that sold those copies to him violated its license, that's between AutoDesk and that company, but not Mr. Vernor.
And license violation doesn't automatically mean copyright infringement, either.
On an 8080? 8086 or 8088, sure, but I don't think the 8080 was really compatible with the x86 instruction set. Similar, sure, but not compatible in either direction.
It really is scary just how powerful computers are today. I recently built a new computer, using a Xeon E3110 (everyone was out of Core 2 Duo E8400's recently, and the Xeon was only about $10 more, and I didn't feel like waiting around). I used to work at Thinking Machines, and a group of us were planning a reunion later that week, and it occurred to me that in just about every measure -- floating point, memory capacity, disk bandwidth, and even memory bandwidth -- my new machine was at least equal to a full CM-2. In some ways -- storage capacity and total I/O bandwidth -- it blows the CM-2 out of the water.
When the CM-2 was introduced in 1987, it was way faster than anything else out there -- if you could figure out how to actually program it. These days, even a distinctly midrange home system (we're not talking an "Extreme" here) gives it at least an honest run for the money. There aren't any CM-2's still running that I know of, and apparently the last running CM-5 was shut down a few years ago, although none of us who ever worked on these remarkable machines would be thrilled to be proven wrong on those two statements.
As others have said, using these things with streaming I/O doesn't make much sense.
I recently built myself a new system. The new processor (Xeon E3110, aka Core 2 Duo E8400) certainly did make boot time somewhat faster, but not dramatically so. Likewise for initial login -- the KDE desktop came up somewhat faster, but it wasn't overwhelming.
Then it occurred to me to move my root and home directory partitions from an older 250 GB 1.5 Gb/sec SATA drive to my newer 500 GB 3.0 Gb/sec compatible drive. There are more differences than just the interface speed; the most notable one is probably Native Command Queuing. This is similar to tagged command queuing on SCSI; it allows the host to queue multiple commands to the disk, which replies to 'em as ready.
*That* made a difference.
fsck insists on checking the partitions one at a time (it sequences the partitions on a particular disk even if you tell it otherwise in/etc/fstab), and it's single threaded and does no async I/O. Reiserfs check does a quick scan of the filesystem tree, which seems to take time proportional to the size of the partition, so that didn't really change very much. After that, it's a completely different story. The rest of the boot sequence now completes in maybe 10 seconds, tops (it previously took something like a minute), and login to my KDE desktop (even including Firefox) really is fast -- only a bit slower than a warm login, when everything's in memory. The issue's not lack of memory -- I had 2 GB on my old system and 4 GB now, and anyway, the big change was on my new system only when I switched drives around.
Moral of the story: if you're suffering from slow boot, make sure that your motherboard supports SATA 3.0 Gb/sec, use AHCI and native IDE (not legacy IDE), and make sure that your drive supports SATA 3.0 Gb/sec and that it isn't jumpered for 1.5 Gb/sec.
You can buy a boxed OpenSUSE system. That's selling software. In any event, what difference does it make whether you're selling "software" or "service"? Money's fungible, and what difference does it make what you're technically selling?
As far as I'm concerned, if I buy something, it should be mine, pure and simple. Yes, it is an issue of rights if I'm not allowed to do whatever I damn well please with something I buy -- a straight purchase transaction.
As far as something for nothing: I'm the project lead for Gutenprint, so I think I'm giving a little something back to the community, thank you very much. Most of the really serious free software proponents (such as Richard Stallman) have in fact contributed a tremendous amount back to the community; I'm chump change compared to them.
It has nothing to do with paying for it or not. It's *precisely* about being able to use it -- or the fruits of it. And I think you're well aware of that, given the way you're oh so carefully trying to link the "free" of freedom with the "free" of zero price. What it's more like is Staples forbidding you from trying to build your own office equipment or from using what you buy from them in any way that they don't approve of. If I buy a pencil from Staples, they don't forbid me from extracting the core and using it as an electrode for an arc lamp or some such. If I buy proprietary software, I usually *am* forbidden from repurposing it.
Companies can and do charge for open source software. Red Hat, Novell, etc. seem to be able to make a business of it. Equating commercial with proprietary is a red herring, and again I think you actually do know that.
Let's be very clear what the issues with proprietary software are:
1) You can't (and you're usually legally forbidden from even trying to) figure out what's going wrong with it, if it isn't working the way you need. If there's a security hole, for example, you're completely at the mercy of the vendor. If the vendor isn't interested, or wants to force you to pay for an upgrade, or goes out of business, you're stuck.
2) You typically don't have any real control over your data, assuming that it's stored in a proprietary file format. If the vendor goes south, and the version of software I have won't run on another machine when my current one breaks, I'm stuck. If the vendor remotely disables it, I'm also stuck.
I'm willing to trust someone, as long as what I'm being asked to trust is transparent and trusts me. Proprietary software vendors obviously don't trust me, or they wouldn't build DRM into everything, or have all this activation nonsense, or forbid me from disassembling it, or whatnot. Why should I trust them?
The problem is that bandwidth (and for that matter latency, only more so) hasn't kept pace with capacity. So yes, we have a lot more storage capacity, but getting the data onto and off the disk hasn't improved by nearly as much.
It's relatively not *too* bad if you're working with large files that can stream. A system I bought in 1994 had a 420 MB disk, which was state of the art at the time. Its bandwidth was on the order of 1 MB/sec. In contrast, the 500 MB disks I'm using now get about 60 MB/sec (internal SATA, at any rate -- USB disks are still limited to 20 MB/sec). That's about 1200x the storage with 60x the transfer rate, so the relative transfer performance (a word I just made up) is about 5% of what it was then.
Latency's another matter altogether. Both seek time and rotational latency are about half what they were then (rotational latency based on 7200 RPM today vs. 3600 RPM in the mid 1990's). So if you're latency-bound, you're really in tough shape. If you're streaming ogg files or what have you, you don't have this problem, but if you're paging to disk (or use applications that create a lot of small files, or scan directories containing lots o'files) you're really in a world of hurt.
Enterprise SAS disks tend to be a lot lower in capacity (74 and 150 GB are common sizes), but rotate at 15000 RPM. So you're spreading out your data over a lot more disks, improving your net throughput, and you're suffering much less from latency. If your application's multi-threaded, with plenty of threads performing queued I/O and plenty of workers, you can make progress even while you're waiting for other I/O ops to complete.
Most of Indiana, as I recall, is at the extreme western edge of the eastern time zone, so the astronomical day is already very late relative to the clock day (on standard time, astronomical noon is close to 12:30, so on DST it's close to 1:30). So that's an extreme example. One could have picked the other extreme (say, Maine, or Illinois in the central time zone), and the results might have been different.
why sites that solicit controversial posts just plain flat out don't keep log files, even momentarily (surely it's possible to configure any server to not keep any records *at all*). Granted, they could try go after the ISP, but unless the ISP is logging all connection attempts from anywhere, it's going to be a lot harder to get that information.
1. No more lifetime-plus copyrights. You get a couple decades and that's it. If it doesn't become popular until 40 years after the fact, well that's just rough luck.
That might work, it might not. I agree that in most countries, durations are now excessively long, but if they became too short there is a genuine risk of stifling innovation. If all copyright worldwide was 40 years from publication, the market would now be saturated with free public domain material. You think TVs full of repeats now? It would be ten times worse. Broadcast the free or the expensive? Star Trek: TOS or Enterprise? It's a no-brainer.
We're on the cusp of a creative crunch: copyright protection in the 20th century has created unprecedented growth in creative output, but as the fruits of that growth fall into the public domain, economics will cease to favour such innovation. The industry can't even prepare for this: the cost of star actors and directors is just getting more and more stupid, and further and further from competitive.
What's the problem with reruns, anyway? In any case, if TV stations broadcast unpopular material, they'll feel it in their ratings. Copyright's supposed to provide an incentive to create (whether it does is another matter); it's not supposed to permanently lock out competition. As for the cost of star actors and directors: either they take a pay cut, or people find ways around it (use lesser known actors who are just as good).
4. No right to control distribution. You can choose the cost per copy but after than anyone has a right to make copies so long as they pay you for them.
Plus I still don't want the KKK to be able to use my works in their propaganda, paid for or not.
They can make certain uses of it anyway, under fair use.
5. No exclusive right to make derivative works. If I want to write a story in your world setting, I can do it and sell it as much as I want, as long as it accompanies a legally acquired copy of your work as well.
Third-party derivative works risk my ability to profit from sequels. It allows another party to change the public perception of my world, diluting my vision and damaging my work. Imagine you're writing a series of books, and in between the publication of parts 2 and 3 an incredibly popular fan-fiction scene starts up and kills off my characters. Three books are released by different authors furthering the world in an opposite direction from which I intended -- including the two aforementioned characters saving the world in episode 9. A stupendously large chunk of my readership are now disengaged from my reality and are no longer interested. I'm left selling less books, and most of the books I am selling are just more reprints of the first book that'll never be read. Probably on a CD licensed under your proposal 4.
Again, copyright is supposed to provide an incentive to create to benefit society, not to achieve maximum revenue potential for the creator. Fan fiction also expands the market for the entire setting.
Let's just say that I don't agree with you on this. To me, this looks like an attempt to cast unwarranted aspersions on non-restrictive licensing in general, at least as applied to photography.
Creative Commons is just another copyright license (more precisely, set of licenses, but I don't believe that to be terribly relevant). Everything you say applies equally to any other kind of copyright license. Someone could misappropriate a work under a different license, or neglect to get a model release, and to the customer it's totally irrelevant whether the license used is CC or anything else.
It appears to me that what you're trying to say is that any form of distribution other than a signed contract in which money changes hands in return for strictly defined usage rights, and acquiring photographs from any but the best known (with the implication that those are the most trustworthy) sources is dangerous. If that's what you believe, that's fine, but if such a trustworthy source were to use CC, why would there be any greater risk than with any other license? Or is your real point that no such "serious" photographer would use a CC license? Or that the only way to ensure "safety" is for the photographer to restrict distribution?
I could envision a business model in which a photographer makes large numbers of stock photographs (say, that photographer's entire body of work other than specific hired jobs) available under a CC license (requiring attribution), using this library in part as a form of advertisement for custom photography services. For these hired jobs, the copyright of the work product would be given to the client as a work-for-hire, freeing the photographer from having to keep track of the later use of this work. The photographer doesn't mind anyone using the stock photos free of charge for any purpose whatsoever, but wants maximum distribution to maximize the free advertising. In other words, flood the market with free work, and make money on the specific paid jobs. This, by the way, is a business model I'd like to try some day. Ignoring issues such as whether this is an effective way to make money or not (maybe you believe that it isn't and I believe that it is, but if I do this, it's my livelihood, not yours, that I'm putting on the line), what is the fundamental flaw of this as a practice?
We've already seen that there are some people -- a tiny minority, to be sure -- who are willing to sacrifice the lives of everyone on board. Removing this kind of weakness, which would surely be inexpensive to engineer, would seem to be one of those things with a relatively good cost/benefit ratio.
Security is only as good as its weakest point. There's no reason that passengers need any access -- even theoretical access -- to the avionics and maintenance systems of the aircraft. Being able to see the flight data is certainly nice, but it can be done safely, as someone else described (via a hardware-enforced write-only mechanism such as a one way optical transistor or whatever that kind of thing is called) and if for some reason it couldn't be done safely, it would be a minor loss. Commodity networking technology is very nice, but there's a time and place for physical partition also.
There *is* a separate wire going to each seat back, if you think about it. As long as the lights are centrally powered (as opposed to a battery in each seat back with wireless control -- and I won't go into that), the wire is already there, whether there's some kind of central bus, or there's a computer in each seat back, or whatever.
The only question is how general-purpose that connection is, or whether it's a dedicated wire for that light.
"Not completely connected" is a very strange phrase. I could say that my laptop is "not completely connected" to the internet because there's a router between them. But either there's a connection between the two networks or there isn't. I don't know what it means to be connected at some points and not at others.
The pilots certainly do need access to some of the cabin systems, for the seatbelt sign, for example. They may also need to be able to turn the cabin network off altogether. But those switches should have no signal connection of any kind to the maintenance and monitoring/control systems. The two networks should be physically partititioned.
The way I read the article, there really are some connections between the networks (my guess is that it was simply cheaper or more convenient to link them), and the FAA's not happy with that state of affairs. I can't say I blame them.
Somehow I have a suspicion that someone will crack this sooner or later, and the TSA will react by banning use of laptops or something equally foolish, rather than addressing the more basic fact that the plane's systems have not been hardened appropriately (in this case, by being physically partitioned).
Whatever rules have been made, there's a fundamental difference: when you steal something from someone, that person no longer has possession of it. Copyright infringement may reduce the value of something, but it doesn't actually prevent the copyright holder from using it.
For the N+1'th time: *copyright infringement is not stealing*. It's copyright infringement. It's making a copy that the law doesn't authorize. That's not the same thing as actually taking something away from someone. Not to mention that DRM doesn't allow the consumer the freedom to do with the content what they want to.
People have to devise business models that don't rely on being able to sell copies of things that inherently have near zero unit production cost (the incremental cost of producing another copy of a recorded song is near zero -- note that I am specifically ignoring the development cost here). It's a fairly basic principle that in a competitive market the price of a good eventually falls to its incremental production cost. Trying to impose an artificially high price in this situation leads to a black market. Either everyone is going to be a criminal, or this artificial scarcity has to be broken and a solution friendlier to the market has to be found.
There are other business models, though:
* Use recorded music to promote live concerts. Under this model, a musician uses recordings as essentially advertising for live performance. That's a good that cannot be copied -- Neil Young is the only Neil Young, and for many people there is no possible substitute, for example.
* The "street performer protocol" -- provide the next song, or chapter of a book, or movie, only when people have paid enough for the previous one. That means you have to give away enough freebies to get recognized, then "I'll release the next song when people have paid $10,000 for this one". It certainly at least reduces the potential unlimited upside (although music that's being released directly by the artist might find people more willing to pay even after release than when it's being released by a large recording company). But most of us don't exactly have an unlimited upside from our work; why should creative artists be any different?
What this does is effectively charge a price for the scarce good -- the work that hasn't been released yet.
Stephen King tried a variation on this that didn't work maybe 5 years ago, and then tried to use it as a proof that SPP won't work. The variation was to only release the next book if 75% of the people who downloaded the previous one paid for it. Can you see why that's not even remotely the same thing? He wasn't putting a price on the next book, without caring who paid for it; he was still trying to impose a royalty model with perhaps a 25% discount. For the SPP to work correctly, you have to specify *only* the total price for producing the next work. If you want more money, raise the price for the work after that.
If this spells the end for large entertainment conglomerates, it's a small loss (at most) to everyone else. Like anyone else in a declining sector of the economy (buggy whips, anyone?), they can find something else to do. They're essentially only acting as middlemen in a market that doesn't need that kind of thing.
bbaston on Groklaw received his XO laptop (see this story) seems to think that it will work really well for that purpose. I'm going to find out in a few weeks. Give one, get one!
And they're all going out of their way to point out that it's running Linux, not Windows, that you shouldn't expect your Windows programs to run on it, and that you should understand what it is and what it isn't. Is this really an astroturf campaign to get people to think for themselves?
Slashdot Mirror
Absence of copyright doesn't necessarily mean freedom to take credit for what someone else creates -- a right of attribution could be separated from a right to restrict copying.
This is neither here nor there, for a couple of reasons. It isn't copyright that prevents that from happening; it's more of a privacy right. People who want to use images of recognizable people commercially need model releases in at least certain circumstances; the model release is essentially saying "you have a right to use my likeness for this purpose". But this isn't a general purpose copyright, and people don't have a copyright on how they look -- it's perfectly legal to imitate someone's appearance (at least as long as it's not for a fradulent purpose), and it's likewise legal to photograph someone.
The other reason this is irrelevant is that usually the photographer, not the client, holds the copyright on the images. That's why if you want copies of the photographs from your own wedding you usually have to go back to the original photographer and pay outrageous amounts of money for prints -- if the photographer has kept the negatives or files (usually they don't hold on to them for very long). There are photographers who will either work for hire (as opposed to being contractors), in which case the person hiring the photographer gets the copyright, but those are the exceptions (I think -- hopefully I'm wrong about that, but I don't think so). So absent a contractual agreement (which trumps copyright law), and aside from the model release issue, the photographer, holding the copyright, could still sell your wedding images in this way, but you yourself couldn't!
As Richard Stallman likes to say, we can't talk about "intellectual property" as one thing; there are copyrights, patents, trademarks, trade secrets, all of which have very different characteristics. I suppose we could add privacy rights and right of attribution as other forms of "intellectual property". Personally, I would like to eliminate copyright and patent outright (find other solutions for things like medicines -- outright subsidies come to mind as one possibility, which I actually think is more market-based than patents). Trade secrets on things that really are secret are more reasonable, as long as we don't get into the emperor has no clothes situation where companies claim that what they sell you has trade secrets, even though they're right out in the open for you to figure out. Trademarks are reasonable as non-functional means of identification, but again it shouldn't be possible to leverage them into a functional role (a game console refuses to run games that don't have the trademark embedded, and they hold onto that monopoly by enforcing the trademark).
Same reason that most people work for specified wages or salary -- they're paid money for it, upfront (maybe in arrears for a short time, but they do get paid).
Let's be a bit more precise about what we're talking about. The issue isn't what the hypothetical band decides what to do with it; I don't believe there's much controversy about that. I don't think that anyone is arguing that the band must make copies for everyone who wants one. What the issue is is whether, and to what extent, the band may restrict other people from doing what they want with it.
This error is very important; when someone says "a person should have a right to do whatever they want with what they create!" it sounds very obvious. When it's instead stated as "a person should have a right to restrict what others may do with their own copy of something because he created it" it doesn't sound so obvious. If you buy a computer or a car outright, does/should the seller have any post-sale claim on the item? Why should it be different in the case of an intangible item?
Now, if you want to argue that allowing unlimited downstream copying impairs the market, or de facto makes it impossible to sell something for the desired price, that may be true -- but it isn't the same thing as not having a right to ask whatever price you want. A legal right to offer something for sale doesn't mean that someone else has to pay that price, if they can get it elsewhere for less.
This has been quite thoroughly debunked; see http://www.snopes.com/quotes/internet.asp
I'm not pleased to see a decent and honorable person like Gore lampooned for something he never even said. He received a Webby in 2005 for lifetime achievement, and Vint Cerf has also defended him.
Not to mention deliberately leaving his wireless network open for "convenience" (although securing a wireless network isn't all that hard in Linux) and so that he can share connections with neighbors. Deliberately running an open endpoint should mean having to take at least some responsibility for what flows through it, unless you're a common carrier and really are in the business of supplying pure bandwidth to your customers.
I wonder if this person would appreciate being spammed through someone else's open relay.
So Ron Paul, who is (or at least claims to be) a libertarian, attracts "traditional social conservatives", who are primarily interested in having the government enforce their own beliefs on others about what consenting adults should or should not be allowed to do privately? Anybody else see anything wrong with this picture?
It's not polite to turn on preload in Fasterfox!
But that's what this whole thing is about -- AutoDesk has been sending DMCA (i. e. copyright) takedown notices to eBay over Mr. Vernor's attempts to sell copies of AutoDesk whose license he never agreed to in the first place.
The First Sale Doctrine *is* also law, and contracts and agreements do have to follow the law.
Mr. Vernor never installed (those copies, at any rate) of AutoCAD, and so never agreed to any license. If the company that sold those copies to him violated its license, that's between AutoDesk and that company, but not Mr. Vernor.
And license violation doesn't automatically mean copyright infringement, either.
On an 8080? 8086 or 8088, sure, but I don't think the 8080 was really compatible with the x86 instruction set. Similar, sure, but not compatible in either direction.
It really is scary just how powerful computers are today. I recently built a new computer, using a Xeon E3110 (everyone was out of Core 2 Duo E8400's recently, and the Xeon was only about $10 more, and I didn't feel like waiting around). I used to work at Thinking Machines, and a group of us were planning a reunion later that week, and it occurred to me that in just about every measure -- floating point, memory capacity, disk bandwidth, and even memory bandwidth -- my new machine was at least equal to a full CM-2. In some ways -- storage capacity and total I/O bandwidth -- it blows the CM-2 out of the water.
When the CM-2 was introduced in 1987, it was way faster than anything else out there -- if you could figure out how to actually program it. These days, even a distinctly midrange home system (we're not talking an "Extreme" here) gives it at least an honest run for the money. There aren't any CM-2's still running that I know of, and apparently the last running CM-5 was shut down a few years ago, although none of us who ever worked on these remarkable machines would be thrilled to be proven wrong on those two statements.
As others have said, using these things with streaming I/O doesn't make much sense.
/etc/fstab), and it's single threaded and does no async I/O. Reiserfs check does a quick scan of the filesystem tree, which seems to take time proportional to the size of the partition, so that didn't really change very much. After that, it's a completely different story. The rest of the boot sequence now completes in maybe 10 seconds, tops (it previously took something like a minute), and login to my KDE desktop (even including Firefox) really is fast -- only a bit slower than a warm login, when everything's in memory.
I recently built myself a new system. The new processor (Xeon E3110, aka Core 2 Duo E8400) certainly did make boot time somewhat faster, but not dramatically so. Likewise for initial login -- the KDE desktop came up somewhat faster, but it wasn't overwhelming.
Then it occurred to me to move my root and home directory partitions from an older 250 GB 1.5 Gb/sec SATA drive to my newer 500 GB 3.0 Gb/sec compatible drive. There are more differences than just the interface speed; the most notable one is probably Native Command Queuing. This is similar to tagged command queuing on SCSI; it allows the host to queue multiple commands to the disk, which replies to 'em as ready.
*That* made a difference.
fsck insists on checking the partitions one at a time (it sequences the partitions on a particular disk even if you tell it otherwise in
The issue's not lack of memory -- I had 2 GB on my old system and 4 GB now, and anyway, the big change was on my new system only when I switched drives around.
Moral of the story: if you're suffering from slow boot, make sure that your motherboard supports SATA 3.0 Gb/sec, use AHCI and native IDE (not legacy IDE), and make sure that your drive supports SATA 3.0 Gb/sec and that it isn't jumpered for 1.5 Gb/sec.
You can buy a boxed OpenSUSE system. That's selling software. In any event, what difference does it make whether you're selling "software" or "service"? Money's fungible, and what difference does it make what you're technically selling?
As far as I'm concerned, if I buy something, it should be mine, pure and simple. Yes, it is an issue of rights if I'm not allowed to do whatever I damn well please with something I buy -- a straight purchase transaction.
As far as something for nothing: I'm the project lead for Gutenprint, so I think I'm giving a little something back to the community, thank you very much. Most of the really serious free software proponents (such as Richard Stallman) have in fact contributed a tremendous amount back to the community; I'm chump change compared to them.
It has nothing to do with paying for it or not. It's *precisely* about being able to use it -- or the fruits of it. And I think you're well aware of that, given the way you're oh so carefully trying to link the "free" of freedom with the "free" of zero price. What it's more like is Staples forbidding you from trying to build your own office equipment or from using what you buy from them in any way that they don't approve of. If I buy a pencil from Staples, they don't forbid me from extracting the core and using it as an electrode for an arc lamp or some such. If I buy proprietary software, I usually *am* forbidden from repurposing it.
Companies can and do charge for open source software. Red Hat, Novell, etc. seem to be able to make a business of it. Equating commercial with proprietary is a red herring, and again I think you actually do know that.
Let's be very clear what the issues with proprietary software are:
1) You can't (and you're usually legally forbidden from even trying to) figure out what's going wrong with it, if it isn't working the way you need. If there's a security hole, for example, you're completely at the mercy of the vendor. If the vendor isn't interested, or wants to force you to pay for an upgrade, or goes out of business, you're stuck.
2) You typically don't have any real control over your data, assuming that it's stored in a proprietary file format. If the vendor goes south, and the version of software I have won't run on another machine when my current one breaks, I'm stuck. If the vendor remotely disables it, I'm also stuck.
I'm willing to trust someone, as long as what I'm being asked to trust is transparent and trusts me. Proprietary software vendors obviously don't trust me, or they wouldn't build DRM into everything, or have all this activation nonsense, or forbid me from disassembling it, or whatnot. Why should I trust them?
The problem is that bandwidth (and for that matter latency, only more so) hasn't kept pace with capacity. So yes, we have a lot more storage capacity, but getting the data onto and off the disk hasn't improved by nearly as much.
It's relatively not *too* bad if you're working with large files that can stream. A system I bought in 1994 had a 420 MB disk, which was state of the art at the time. Its bandwidth was on the order of 1 MB/sec. In contrast, the 500 MB disks I'm using now get about 60 MB/sec (internal SATA, at any rate -- USB disks are still limited to 20 MB/sec). That's about 1200x the storage with 60x the transfer rate, so the relative transfer performance (a word I just made up) is about 5% of what it was then.
Latency's another matter altogether. Both seek time and rotational latency are about half what they were then (rotational latency based on 7200 RPM today vs. 3600 RPM in the mid 1990's). So if you're latency-bound, you're really in tough shape. If you're streaming ogg files or what have you, you don't have this problem, but if you're paging to disk (or use applications that create a lot of small files, or scan directories containing lots o'files) you're really in a world of hurt.
Enterprise SAS disks tend to be a lot lower in capacity (74 and 150 GB are common sizes), but rotate at 15000 RPM. So you're spreading out your data over a lot more disks, improving your net throughput, and you're suffering much less from latency. If your application's multi-threaded, with plenty of threads performing queued I/O and plenty of workers, you can make progress even while you're waiting for other I/O ops to complete.
Most of Indiana, as I recall, is at the extreme western edge of the eastern time zone, so the astronomical day is already very late relative to the clock day (on standard time, astronomical noon is close to 12:30, so on DST it's close to 1:30). So that's an extreme example. One could have picked the other extreme (say, Maine, or Illinois in the central time zone), and the results might have been different.
why sites that solicit controversial posts just plain flat out don't keep log files, even momentarily (surely it's possible to configure any server to not keep any records *at all*). Granted, they could try go after the ISP, but unless the ISP is logging all connection attempts from anywhere, it's going to be a lot harder to get that information.
What's the problem with reruns, anyway? In any case, if TV stations broadcast unpopular material, they'll feel it in their ratings. Copyright's supposed to provide an incentive to create (whether it does is another matter); it's not supposed to permanently lock out competition. As for the cost of star actors and directors: either they take a pay cut, or people find ways around it (use lesser known actors who are just as good).
They can make certain uses of it anyway, under fair use.
Again, copyright is supposed to provide an incentive to create to benefit society, not to achieve maximum revenue potential for the creator. Fan fiction also expands the market for the entire setting.
See Against Intellectual Monopoly for another take on the whole issue.
I sent the following email to Dan Heller.
Let's just say that I don't agree with you on this. To me, this looks like an attempt to cast unwarranted aspersions on non-restrictive licensing in general, at least as applied to photography.
Creative Commons is just another copyright license (more precisely, set of licenses, but I don't believe that to be terribly relevant). Everything you say applies equally to any other kind of copyright license. Someone could misappropriate a work under a different license, or neglect to get a model release, and to the customer it's totally irrelevant whether the license used is CC or anything else.
It appears to me that what you're trying to say is that any form of distribution other than a signed contract in which money changes hands in return for strictly defined usage rights, and acquiring photographs from any but the best known (with the implication that those are the most trustworthy) sources is dangerous. If that's what you believe, that's fine, but if such a trustworthy source were to use CC, why would there be any greater risk than with any other license? Or is your real point that no such "serious" photographer would use a CC license? Or that the only way to ensure "safety" is for the photographer to restrict distribution?
I could envision a business model in which a photographer makes large numbers of stock photographs (say, that photographer's entire body of work other than specific hired jobs) available under a CC license (requiring attribution), using this library in part as a form of advertisement for custom photography services. For these hired jobs, the copyright of the work product would be given to the client as a work-for-hire, freeing the photographer from having to keep track of the later use of this work. The photographer doesn't mind anyone using the stock photos free of charge for any purpose whatsoever, but wants maximum distribution to maximize the free advertising. In other words, flood the market with free work, and make money on the specific paid jobs. This, by the way, is a business model I'd like to try some day. Ignoring issues such as whether this is an effective way to make money or not (maybe you believe that it isn't and I believe that it is, but if I do this, it's my livelihood, not yours, that I'm putting on the line), what is the fundamental flaw of this as a practice?
We've already seen that there are some people -- a tiny minority, to be sure -- who are willing to sacrifice the lives of everyone on board. Removing this kind of weakness, which would surely be inexpensive to engineer, would seem to be one of those things with a relatively good cost/benefit ratio.
Security is only as good as its weakest point. There's no reason that passengers need any access -- even theoretical access -- to the avionics and maintenance systems of the aircraft. Being able to see the flight data is certainly nice, but it can be done safely, as someone else described (via a hardware-enforced write-only mechanism such as a one way optical transistor or whatever that kind of thing is called) and if for some reason it couldn't be done safely, it would be a minor loss. Commodity networking technology is very nice, but there's a time and place for physical partition also.
There *is* a separate wire going to each seat back, if you think about it. As long as the lights are centrally powered (as opposed to a battery in each seat back with wireless control -- and I won't go into that), the wire is already there, whether there's some kind of central bus, or there's a computer in each seat back, or whatever.
The only question is how general-purpose that connection is, or whether it's a dedicated wire for that light.
"Not completely connected" is a very strange phrase. I could say that my laptop is "not completely connected" to the internet because there's a router between them. But either there's a connection between the two networks or there isn't. I don't know what it means to be connected at some points and not at others.
The pilots certainly do need access to some of the cabin systems, for the seatbelt sign, for example. They may also need to be able to turn the cabin network off altogether. But those switches should have no signal connection of any kind to the maintenance and monitoring/control systems. The two networks should be physically partititioned.
The way I read the article, there really are some connections between the networks (my guess is that it was simply cheaper or more convenient to link them), and the FAA's not happy with that state of affairs. I can't say I blame them.
Somehow I have a suspicion that someone will crack this sooner or later, and the TSA will react by banning use of laptops or something equally foolish, rather than addressing the more basic fact that the plane's systems have not been hardened appropriately (in this case, by being physically partitioned).
Whatever rules have been made, there's a fundamental difference: when you steal something from someone, that person no longer has possession of it. Copyright infringement may reduce the value of something, but it doesn't actually prevent the copyright holder from using it.
For the N+1'th time: *copyright infringement is not stealing*. It's copyright infringement. It's making a copy that the law doesn't authorize. That's not the same thing as actually taking something away from someone. Not to mention that DRM doesn't allow the consumer the freedom to do with the content what they want to.
People have to devise business models that don't rely on being able to sell copies of things that inherently have near zero unit production cost (the incremental cost of producing another copy of a recorded song is near zero -- note that I am specifically ignoring the development cost here). It's a fairly basic principle that in a competitive market the price of a good eventually falls to its incremental production cost. Trying to impose an artificially high price in this situation leads to a black market. Either everyone is going to be a criminal, or this artificial scarcity has to be broken and a solution friendlier to the market has to be found.
There are other business models, though:
* Use recorded music to promote live concerts. Under this model, a musician uses recordings as essentially advertising for live performance. That's a good that cannot be copied -- Neil Young is the only Neil Young, and for many people there is no possible substitute, for example.
* The "street performer protocol" -- provide the next song, or chapter of a book, or movie, only when people have paid enough for the previous one. That means you have to give away enough freebies to get recognized, then "I'll release the next song when people have paid $10,000 for this one". It certainly at least reduces the potential unlimited upside (although music that's being released directly by the artist might find people more willing to pay even after release than when it's being released by a large recording company). But most of us don't exactly have an unlimited upside from our work; why should creative artists be any different?
What this does is effectively charge a price for the scarce good -- the work that hasn't been released yet.
Stephen King tried a variation on this that didn't work maybe 5 years ago, and then tried to use it as a proof that SPP won't work. The variation was to only release the next book if 75% of the people who downloaded the previous one paid for it. Can you see why that's not even remotely the same thing? He wasn't putting a price on the next book, without caring who paid for it; he was still trying to impose a royalty model with perhaps a 25% discount. For the SPP to work correctly, you have to specify *only* the total price for producing the next work. If you want more money, raise the price for the work after that.
If this spells the end for large entertainment conglomerates, it's a small loss (at most) to everyone else. Like anyone else in a declining sector of the economy (buggy whips, anyone?), they can find something else to do. They're essentially only acting as middlemen in a market that doesn't need that kind of thing.
bbaston on Groklaw received his XO laptop (see this story) seems to think that it will work really well for that purpose. I'm going to find out in a few weeks. Give one, get one!
And they're all going out of their way to point out that it's running Linux, not Windows, that you shouldn't expect your Windows programs to run on it, and that you should understand what it is and what it isn't. Is this really an astroturf campaign to get people to think for themselves?