Whether or not this IBM program is an LGPL violation, I can't see what the big deal is. Oh, no! Someone statically linked a Linux executable against the Linux C library. Horrors! Do you really think that they're hoarding some secret modification they've made to libc? Even if they did provide libc sources, would anybody have bothered to diff them to see if they've done something magical to printf()?
I think that nit-picking any trivial (mis)use of free software encourages the fear and uncertainty that some PHB's still have toward using anything they don't pay $$$$ for. It would be better to save the legal analyses for cases which are truly meaningful.
I think one reason the Perl Artistic license comes up for discussion less often is that the license itself isn't the important part of Perl -- the language is. Many GPL programs have been written specifically because of the license, i.e. to provide a "free" alternative to something that already exists (particularly all of the usual UNIX commands: ls, tar, etc.). In Perl's case, Perl is Perl, it's not a free version of something that was developed commercially.
So "Let's work on a GPL Napster/javac/VMWare/X/..." comes up a lot, but "let's make a GPL Perl" hardly makes any sense.
Actually, Purify was developed by Pure Software, which was bought out by Atria, which made ClearCase, becoming PureAtria, which was, in turn, bought by Rational, which makes Rose. So, if we know anything about Rational, it's that they've been in acquisition mode for a while.
I'll swear by Purify, and although I gripe about ClearCase it'd be foolish to attempt source control on a large, geographically distributed project without it (and MultiSite). Rose I'm less sold on, but it's better than nothing.
When I see the Easter Eggs that those fun-loving wacky light-hearted minions of the Dark Side over in Redmond toss into their products, I immediately wonder "how many of the fscking bugs in your fscking products could you have fixed while you were programming that fscking pinball game!?"
If Microsoft, Apple, etc. had an official place for credits, then there'd be a lot less incentive for programmers to spend lots of time creating fancy easter eggs. Besides, I'll bet that practically all Easter Egg programming is already done off the clock anyway.
In the SJMercury article, the point is made that individual credits make the technical contributors more visible and thus more likely to be recruited away. It seems only fair -- the managers are already visible because they're the ones who get quoted in company press releases. "FooBar 3.1 is the greatest advance in technology since sliced bread," said P. H. Boss, Manager, Advanced Products. You know that Mr(s). Boss had nothing to do with the actual technology.
One of the supposed reasons for dropping individual credits is that the lists have become too large, but that doesn't seem like a valid excuse. I work on software used by the motion picture industry, and while I don't get a credit in the software, the people who use the software usually get a credit in the movie, even if it's buried among thousands of credits and falls somewhere between the credit for the caterer and the obligatory 'no animals were harmed' statement.
Now that it's uncommon for a computer not to have an internet connection, and with full-time connections becoming more common, this is only the tip of the iceberg.
What won't stop invasion of privacy is so-called disclosure in license agreements and readme files. First, nobody reads those, and second, they're too vague. I think that the info that ID gathered was perfectly acceptable, while what RealJukebox did was definitely not, and yet one generic disclosure statement would cover both.
I think that what we need is something similar to anti-virus software that sits between applications and the TCP/IP stack, and limits what different applications can do, putting up warnings and confirmation dialogs as necessary. I expect that my web browser will connect to internet sites. I don't expect that of most other software, and I want to be warned whenever that happens.
This should be similar in concept to some virus protection software. I expect FORMAT.EXE to format disks. I don't expect any other program to do so, and if anything else calls the INT13h or whatever it is (apologies for the DOS-isms), I want to know about it.
Of course, clever programmers could code around anything, just as virus writers avoid detection, but if any company employed such tricks, they'd really have a lot of explaining to do.
I had one of the original Amigas in 1985, and I've never heard of Adam Chowaniec, the guy profiled in this article. I'm sure that he was yet another manager in Commodore's byzantine corporate structure who, if he had anything at all to do with the Amiga, probably hurt it more than helped it.
Just to double-check, and because it was lying around, I just checked the August, 1985 issue of Byte, which was the first in-depth look at the Amiga. Let's see who they mention: the late Jay Miner. (It's a shame they don't mention the guys behind the software, too. Although they describe the software components, no one else is mentioned by name. It's been so long now that the only names I can think of off-hand are Carl Sassenrath and RJ Mical.)
The real definition of an Amiga "pioneer" would be someone with his signature on the inside top cover of an Amiga 1000. Is there a list of these somewhere?
I'll stick to computer-related hacks, otherwise the list is too broad. (How could a trojan horse program compete with The Trojan Horse?) Bearing that in mind, here are a few hacks that may be relatively minor but impressed me nonetheless:
The program that played music (usually Daisy) via RFI picked up by a nearby AM radio. I first encountered an 8080 version of it, but it may go back further than that.
The ZIL (or whatever it was called) engine that ran Zork and all the other Infocom games on every platform known to man in the early-mid 80's was a nice hack. Plus, it inspired some minor hacks in the form of some track loaders we used so that we could buy the game in one format (usually something oddball like Tandy 2000) and transfer the game data to another format.
Emulators are interesting in that it's impressive that they work at all, and amazing when they work well. I'd give the most credit to Magic Sac, which was, I think, the first "hostile port" of the MacOS to another platform (Atari ST); to UAE for doing the "impossible" by emulating the Amiga; and to MAME for the sheer scope of it.
PARNET was a "network" for Amigas that ran over the parallel port and actually worked well enough to be useful.
The Amiga hardware included a number of clever hacks and inspired still more: Hold-And-Modify mode graphics; copper-list-dependent graphics modes (SHAM etc.); overscanned desktops; parallel floppy duplicators (that actually "broadcast" the data to more than one drive at once); scan doublers/flicker fixers; the A2024 monitor; lack of cut-and-paste worked around by OCR'ing the frame buffer...
One thread that runs through most if not all of these hacks is that they make a computer work in some way that was never intended by the original designers. That, to me, is a key ingredient that distinguishes a hack from a non-hack.
According to this article, the complete system runs about $2500
Actually, $2500 buys a bare-bones system with only the special enclosure w/ cooling system, motherboard, and CPU. Add your own memory, controllers, drives, peripherals, monitor(s), etc.
I am very impressed with the quality of the answers (as well as the questions). I'll bet that we see the concepts here rehashed in articles on ZDNet and the like, which is the way it should be, rather than Slashdot rehashing pre-dumbed-down ZDNet/Wired/c|net articles.
To whomever picks interviewees: it's stating the obvious, but the quality of this one was on a different planet from the one with that self-professed "security expert" kid.
Favorite quote:
[Jerry Pournelle's] descriptions of working with DOS and later Windows hardware were in part responsible for my buying Macintoshes again...
There are several intertwined issues here, so it's not easy to boil everything down to a simple "it should[n't] be Open Source" answer:
People are conditioned to want their computers to run faster. The amount of time and effort some people spend to overclock and benchmark their computers is often far out of proportion to the actual benefit they get from their computer's speed. It's not surprising that people treat their SETI@home processing speed as a benchmark.
The fact that SETI@home puts up statistics that have turned this experiment into a competition to complete the most work units reinforces that behavior.
At least one company (perhaps SGI, but I can't remember for sure) has mentioned their SETI@home crunching speed in some marketing literature, again emphasizing speed over quality of results.
As several in this discussion have pointed out, making the clients faster won't help the project because the bottleneck is that SETI@home can't prepare the units fast enough. However...
If the client software were improved, clients could potentially do more sophisticated processing in roughly the same time, improving the science. However...
This could make the clients seem even slower than they already are, which wouldn't sit well with the kiddies who are more interested in their rank or how fast it makes their box seem than the science involved.
So what lessons could be learned if this or a similar experiment were to be done again?
Deemphasize the ranking of work units completed. Perhaps if the concept of a fixed work unit could be dropped altogether (i.e. make the "size" of a work unit something arbitrary so that they couldn't be compared). This would possibly prevent the client from being used more as a benchmark than for its true purpose.
Plan for hacked clients and spoofed results by sending out enough test work units and by cross-checking results with multiple clients enough to have confidence in the results backed up by statistics.
With enough cross-checking, you might as well Open Source the client.
I would be interested to hear if there is a (theoretically) foolproof way to use distributed clients to produce results with confidence if you accept that some clients will be spoofed.
I have a coupon for a discount on an SGI 1400L server that I got by attending a "Linux University" put on by SGI the other morning, and it also says Linux is a registered trademark of Linux Torvalds in very small type along the edge. In fact, that's the only reason I haven't thrown out the coupon yet.
Last I checked, SGI actually did exist; I haven't seen any evidence of existence from this company, though.
Hmm, let's see. To have a web site, I think you need a computer. You should get a computer, or find somebody to loan you theirs.
Seriously, the only clue you've given us is that your web site "has great potential" (what doesn't?) and that it might have some use for banners and a database. That's a lot like saying "I have an idea for a computer program that has great potential; what language and OS should I use?"
Despite what the headline implies, I doubt that MS deliberately broke Notes. They probably just didn't catch it during testing. My guess is that their test coverage is pretty good for other Microsoft products, but not great for non-MS stuff. A certain amount of that is to be expected. They can't test tens of thousands of programs for compatibility with every bug fix they release.
However, my suspicion is that MS hardly cares any more whether its OS works well with anything other than MS products. Now that they have the dominant office suite, the dominant web browser, and are pushing MS alternatives to practically every other mass-market software there is, why should they care whether anybody other than MS can compete and develop stable programs for Windows?
They don't adequately test 3rd-party software compatibility, and the problem is that they can get away with it.
The ridiculously high number of API calls for Windows (and the fact that they're constantly increasing) only makes sense if they don't care about 3rd-party developers being able to keep up.
Think about it: if Windows didn't have the monopoly on desktop OS that it enjoys now, would anybody in their right mind choose to develop software for it? Would they really want to learn the 2500-or-so API calls, only to have an unknown number of them be obsolete when DirectWhatever 9.0 comes out in another month (timed to coincide with the splashy release of MSWhatever 1.0)?
While I'm disappointed that TRUSTe has defined their scope to be very narrowly focused on web pages, that's all they've ever claimed to be. What we really need isn't validation of web sites, it's validation of any software application, especially ones that don't obviously have anything to do with the web.
We're now at the point where internet connections are common enough and alternative payment methods like banner ads in non-browser software are starting to catch on, and that means that every software application really needs to come with a full disclosure of what conditions, if any, will cause it to establish a network connection, what data will be sent, and what will be done with the data after it is uploaded. As much as I normally loathe regulation, this is one that should be enforced by law. For 95% of the software out there, the disclosure would be pretty simple: none. For most of the remainder, the answer is still pretty simple, e.g. "FooMail sends and receives email messages at periodic intervals under control of the user". The remainder is what we need to watch out for, e.g. "InnocentJukeBox scans your hard drive for new, non-copy-protected MP3 files and sends a list of new additions to www.riaa.com every time an internet connection is established."
The disclosure could even be vague, like "Word2002 periodically uploads personal data and usage patterns to microsoft.com" but who in their right mind would consent to use an application like that?
Unfortunately, I doubt that enough people give a damn. Real Networks stock seems to be up following a new web site launch, now that the privacy issue has exceeded the attention span of the media.
IMHO, you should give a lot more of a rat's ass about digital television, even if you don't care about the current crap being broadcast. If things get done right (i.e. the gov't and entrenched broadcasters don't screw things up), this will be a convergence of television and computing. The result? Imagine pipes coming into your house with the bandwidth to handle hundreds of channels of hi-res video and commodity displays capable of 1920x1080.
Wakko Warner asks: What I want to know is, what happens in 6 years when stations quit broadcasting in their current format. Will my non-cable-box-connected, normal, cable-ready televisions still be fine? Or will I have to toss them and buy new ones
AFAIK, while there is a timetable for ending NTSC transmission completely someday, which would require you to toss your equipment or get some kind of converter box, that transmission cutoff won't actually kick in until a high percentage of sets in use are compatible with the new digital standard. Practically speaking, that may not happen for several years beyond the nominal cutoff date.
tgd claims: The 35mm crap people see in movie theaters can't hold anything on HD.
I agree that HD is dramatically better than current video standards (NTSC, PAL), but disagree that it's better than film. Film still has a much higher contrast range and resolution than any HDTV standard, and most of the HD material you'll see for years to come is going to continue to originate on 35mm film.
At the moment, the limited HD material that's out there may look quite good compared to a worn film print in a typical movie theater, but if you could see the results side by side, with equal care going into each, I think your conclusion would be different.
Disclaimer: I work for a company that builds systems for digitizing film for, among other applications, HDTV mastering.
Existing UIs will need to be tweaked (a little, but perhaps not too much) when big displays like this become more common. I've spent some time using a high-end SGI with side-by-side 24" 16x9 widescreen monitors, and it was very easy to lose track of the mouse pointer when it was out of the center of my field of view.
How is it that a site can get its MP3 files cataloged by a search engine but restricts downloads with password-only access? It seems to me that any legitimate search engine would catch on and refuse to catalog such a site. Surely the search engine could spot check and attempt to download a file to verify that a site is open. That could still be spoofed by leaving the site open long enough to be cataloged, but it would go a long way toward taking care of this issue.
(While I think the BPA is on drugs for attempting to link MP3's to organized crime, I don't like the porn banner ad scam either.)
I wonder of Coca-cola's rebuttal is truly because of erroneous news reports or if it's corporate backpedaling.
With a bit of marketing savvy, they could have spun this to make it sound like the machines are lowering the price when the temperature goes down. Then I'll bet the tone of discussion here would have been quite different. There's already well-known model for changing drink prices over time to match demand. It's called happy hour, and because it's advertised like a discount (instead of "during Happy Hour, we don't gouge you quite as bad as we normally do") nobody complains.
Oh, please, not yet another complaint about the size of "hello, world" compiled. Let me guess: you've never written anything larger than the Fahrenheit-to-Celsius converter in K&R. This has absolutely nothing to do with any real world problem. If you want a program to say hello, here you go:
#!/bin/sh
echo "hello, world"
Use the right tool for the job.
Judging C/C++ compilers by comparing "hello, world" sizes is like buying a car based on the horsepower of its starter motor.
Real Genius had a couple of examples of hacking detail that weren't too bad: a screen shot showed a brute force password guesser being used to break into a system, and at one point, they burned an EPROM and plugged it into a PC board on a B-1, and yes, it really looked like an EPROM chip and I think they even plugged it in the right way.
On the other hand, I'm sure that movie would have made anyone familiar with lasers cringe.
Someone at Pez, Inc. finally got word of how eBay grew out of Pierre Omidyar helping his girlfriend sell her Pez collection on the web. Since then, in the late twentieth century's version of the tulip craze, eBay has a market cap larger than $20 billion (as large as all of Hershey, Nabisco, Keebler, and Wrigley combined), with Omidyar's personal holdings amounting to something like $5 billion. So, obviously, the next time somebody tries to launch a multi-billion dollar empire based on a Pez dispenser, their lawyers will be ready to take their cut.
Do a search on eBay for auctions by idsi, and you'll find several auctions like this one, which uses a picture of the Official Red Hat box, and mentions only in small type at the bottom of a multi-page ad that they're just selling a bare CD.
In fact, they're even deep-linking the image of the box from Red Hat's site.
Slashdot Mirror
I think that nit-picking any trivial (mis)use of free software encourages the fear and uncertainty that some PHB's still have toward using anything they don't pay $$$$ for. It would be better to save the legal analyses for cases which are truly meaningful.
So "Let's work on a GPL Napster/javac/VMWare/X/..." comes up a lot, but "let's make a GPL Perl" hardly makes any sense.
I'll swear by Purify, and although I gripe about ClearCase it'd be foolish to attempt source control on a large, geographically distributed project without it (and MultiSite). Rose I'm less sold on, but it's better than nothing.
If Microsoft, Apple, etc. had an official place for credits, then there'd be a lot less incentive for programmers to spend lots of time creating fancy easter eggs. Besides, I'll bet that practically all Easter Egg programming is already done off the clock anyway.
One of the supposed reasons for dropping individual credits is that the lists have become too large, but that doesn't seem like a valid excuse. I work on software used by the motion picture industry, and while I don't get a credit in the software, the people who use the software usually get a credit in the movie, even if it's buried among thousands of credits and falls somewhere between the credit for the caterer and the obligatory 'no animals were harmed' statement.
What won't stop invasion of privacy is so-called disclosure in license agreements and readme files. First, nobody reads those, and second, they're too vague. I think that the info that ID gathered was perfectly acceptable, while what RealJukebox did was definitely not, and yet one generic disclosure statement would cover both.
I think that what we need is something similar to anti-virus software that sits between applications and the TCP/IP stack, and limits what different applications can do, putting up warnings and confirmation dialogs as necessary. I expect that my web browser will connect to internet sites. I don't expect that of most other software, and I want to be warned whenever that happens.
This should be similar in concept to some virus protection software. I expect FORMAT.EXE to format disks. I don't expect any other program to do so, and if anything else calls the INT13h or whatever it is (apologies for the DOS-isms), I want to know about it.
Of course, clever programmers could code around anything, just as virus writers avoid detection, but if any company employed such tricks, they'd really have a lot of explaining to do.
Just to double-check, and because it was lying around, I just checked the August, 1985 issue of Byte, which was the first in-depth look at the Amiga. Let's see who they mention: the late Jay Miner. (It's a shame they don't mention the guys behind the software, too. Although they describe the software components, no one else is mentioned by name. It's been so long now that the only names I can think of off-hand are Carl Sassenrath and RJ Mical.)
The real definition of an Amiga "pioneer" would be someone with his signature on the inside top cover of an Amiga 1000. Is there a list of these somewhere?
- The program that played music (usually Daisy) via RFI picked up by a nearby AM radio. I first encountered an 8080 version of it, but it may go back further than that.
- The ZIL (or whatever it was called) engine that ran Zork and all the other Infocom games on every platform known to man in the early-mid 80's was a nice hack. Plus, it inspired some minor hacks in the form of some track loaders we used so that we could buy the game in one format (usually something oddball like Tandy 2000) and transfer the game data to another format.
- Emulators are interesting in that it's impressive that they work at all, and amazing when they work well. I'd give the most credit to Magic Sac, which was, I think, the first "hostile port" of the MacOS to another platform (Atari ST); to UAE for doing the "impossible" by emulating the Amiga; and to MAME for the sheer scope of it.
- PARNET was a "network" for Amigas that ran over the parallel port and actually worked well enough to be useful.
- The Amiga hardware included a number of clever hacks and inspired still more: Hold-And-Modify mode graphics; copper-list-dependent graphics modes (SHAM etc.); overscanned desktops; parallel floppy duplicators (that actually "broadcast" the data to more than one drive at once); scan doublers/flicker fixers; the A2024 monitor; lack of cut-and-paste worked around by OCR'ing the frame buffer...
One thread that runs through most if not all of these hacks is that they make a computer work in some way that was never intended by the original designers. That, to me, is a key ingredient that distinguishes a hack from a non-hack.Actually, $2500 buys a bare-bones system with only the special enclosure w/ cooling system, motherboard, and CPU. Add your own memory, controllers, drives, peripherals, monitor(s), etc.
To whomever picks interviewees: it's stating the obvious, but the quality of this one was on a different planet from the one with that self-professed "security expert" kid.
Favorite quote:
- People are conditioned to want their computers to run faster. The amount of time and effort some people spend to overclock and benchmark their computers is often far out of proportion to the actual benefit they get from their computer's speed. It's not surprising that people treat their SETI@home processing speed as a benchmark.
- The fact that SETI@home puts up statistics that have turned this experiment into a competition to complete the most work units reinforces that behavior.
- At least one company (perhaps SGI, but I can't remember for sure) has mentioned their SETI@home crunching speed in some marketing literature, again emphasizing speed over quality of results.
- As several in this discussion have pointed out, making the clients faster won't help the project because the bottleneck is that SETI@home can't prepare the units fast enough. However...
- If the client software were improved, clients could potentially do more sophisticated processing in roughly the same time, improving the science. However...
- This could make the clients seem even slower than they already are, which wouldn't sit well with the kiddies who are more interested in their rank or how fast it makes their box seem than the science involved.
So what lessons could be learned if this or a similar experiment were to be done again?- Deemphasize the ranking of work units completed. Perhaps if the concept of a fixed work unit could be dropped altogether (i.e. make the "size" of a work unit something arbitrary so that they couldn't be compared). This would possibly prevent the client from being used more as a benchmark than for its true purpose.
- Plan for hacked clients and spoofed results by sending out enough test work units and by cross-checking results with multiple clients enough to have confidence in the results backed up by statistics.
- With enough cross-checking, you might as well Open Source the client.
I would be interested to hear if there is a (theoretically) foolproof way to use distributed clients to produce results with confidence if you accept that some clients will be spoofed.Last I checked, SGI actually did exist; I haven't seen any evidence of existence from this company, though.
Seriously, the only clue you've given us is that your web site "has great potential" (what doesn't?) and that it might have some use for banners and a database. That's a lot like saying "I have an idea for a computer program that has great potential; what language and OS should I use?"
However, my suspicion is that MS hardly cares any more whether its OS works well with anything other than MS products. Now that they have the dominant office suite, the dominant web browser, and are pushing MS alternatives to practically every other mass-market software there is, why should they care whether anybody other than MS can compete and develop stable programs for Windows?
They don't adequately test 3rd-party software compatibility, and the problem is that they can get away with it.
The ridiculously high number of API calls for Windows (and the fact that they're constantly increasing) only makes sense if they don't care about 3rd-party developers being able to keep up.
Think about it: if Windows didn't have the monopoly on desktop OS that it enjoys now, would anybody in their right mind choose to develop software for it? Would they really want to learn the 2500-or-so API calls, only to have an unknown number of them be obsolete when DirectWhatever 9.0 comes out in another month (timed to coincide with the splashy release of MSWhatever 1.0)?
We're now at the point where internet connections are common enough and alternative payment methods like banner ads in non-browser software are starting to catch on, and that means that every software application really needs to come with a full disclosure of what conditions, if any, will cause it to establish a network connection, what data will be sent, and what will be done with the data after it is uploaded. As much as I normally loathe regulation, this is one that should be enforced by law. For 95% of the software out there, the disclosure would be pretty simple: none. For most of the remainder, the answer is still pretty simple, e.g. "FooMail sends and receives email messages at periodic intervals under control of the user". The remainder is what we need to watch out for, e.g. "InnocentJukeBox scans your hard drive for new, non-copy-protected MP3 files and sends a list of new additions to www.riaa.com every time an internet connection is established."
The disclosure could even be vague, like "Word2002 periodically uploads personal data and usage patterns to microsoft.com" but who in their right mind would consent to use an application like that?
Unfortunately, I doubt that enough people give a damn. Real Networks stock seems to be up following a new web site launch, now that the privacy issue has exceeded the attention span of the media.
IMHO, you should give a lot more of a rat's ass about digital television, even if you don't care about the current crap being broadcast. If things get done right (i.e. the gov't and entrenched broadcasters don't screw things up), this will be a convergence of television and computing. The result? Imagine pipes coming into your house with the bandwidth to handle hundreds of channels of hi-res video and commodity displays capable of 1920x1080.
AFAIK, while there is a timetable for ending NTSC transmission completely someday, which would require you to toss your equipment or get some kind of converter box, that transmission cutoff won't actually kick in until a high percentage of sets in use are compatible with the new digital standard. Practically speaking, that may not happen for several years beyond the nominal cutoff date.
I agree that HD is dramatically better than current video standards (NTSC, PAL), but disagree that it's better than film. Film still has a much higher contrast range and resolution than any HDTV standard, and most of the HD material you'll see for years to come is going to continue to originate on 35mm film.
At the moment, the limited HD material that's out there may look quite good compared to a worn film print in a typical movie theater, but if you could see the results side by side, with equal care going into each, I think your conclusion would be different.
Disclaimer: I work for a company that builds systems for digitizing film for, among other applications, HDTV mastering.
Existing UIs will need to be tweaked (a little, but perhaps not too much) when big displays like this become more common. I've spent some time using a high-end SGI with side-by-side 24" 16x9 widescreen monitors, and it was very easy to lose track of the mouse pointer when it was out of the center of my field of view.
(While I think the BPA is on drugs for attempting to link MP3's to organized crime, I don't like the porn banner ad scam either.)
With a bit of marketing savvy, they could have spun this to make it sound like the machines are lowering the price when the temperature goes down. Then I'll bet the tone of discussion here would have been quite different. There's already well-known model for changing drink prices over time to match demand. It's called happy hour, and because it's advertised like a discount (instead of "during Happy Hour, we don't gouge you quite as bad as we normally do") nobody complains.
Use the right tool for the job.
Judging C/C++ compilers by comparing "hello, world" sizes is like buying a car based on the horsepower of its starter motor.
On the other hand, I'm sure that movie would have made anyone familiar with lasers cringe.
Someone at Pez, Inc. finally got word of how eBay grew out of Pierre Omidyar helping his girlfriend sell her Pez collection on the web. Since then, in the late twentieth century's version of the tulip craze, eBay has a market cap larger than $20 billion (as large as all of Hershey, Nabisco, Keebler, and Wrigley combined), with Omidyar's personal holdings amounting to something like $5 billion. So, obviously, the next time somebody tries to launch a multi-billion dollar empire based on a Pez dispenser, their lawyers will be ready to take their cut.
In fact, they're even deep-linking the image of the box from Red Hat's site.