Not quite. It was mentioned in the series (sorry, not enough of a fan to quote the episode title) that the silver droids with the sweeping red LED eye were built by the Cylons, a reptilian race.
The Cylons were fighting some war, so they built themselves a nice little robot army and gave them a simple command: "Go blow stuff up and kill things." So the robots go out and do it and wipe out the Cylons' enemies. Then they start looking for new targets, and, well... the Cylons happen to be the closest civilization, so the bots wipe them out. And then they start wandering the universe looking for stuff to kill.
It's basically the same premise as the Berserker novels by Saberhagen.
Moral of the story: Automating killing machines with just enough intelligence to distinguish between a lifeform and a rock and giving them power supplies that last for hundreds of thousands of years is a really, really, monumentally stupid idea.
And, just for the record, if a huge armada of homicidal robots showed up and started razing my world, I really wouldn't give a crap where they came from until *AFTER* I'd managed to stop 'em.
Competitive Gaming
on
NYT on Game Mods
·
· Score: 4, Informative
I've recently been sucked into the competitive gameplay world (where teams organize into divisions, leagues, etc, tournaments are held periodically for cash and prizes, and all that good stuff). As much as I used to chuckle at the thought of "pro gamers", it turns out that there can be just as much nuance to strategy and execution to appreciate in watching a multiplayer video game as there is in watching say a football game. At least to my mind.
One cool thing about mods is that they can be used to improve games to a point where they're suitable for competition. The ETpro mod by bani for the game Wolfenstein: Enemy Territory alters some aspects of gameplay to make it more suitable stopwatch competitions.
The other thing mods can do, and this is kinda neat, is actually add in features to accomodate game spectators. Again, using ETpro as an example, bani included some small changes to help shoutcasters quickly identify players and get stats during the match. A multiview feature was also added so that a spectator could watch the game from several different points of view with a Picture-in-Picture style setup.
In the future, I see mods stepping up to fill in the roles that the original game developers either couldn't think of or didn't want to address because the competition world wasn't their target audience. I can see a mod coming out that can not only handle broadcasting video of the match, but offers optional commentary via an mp3/ogg stream from a caster and presents information kind of in the same way FOX does for football games (current scores, tickers for other matches, league stats for players, etc).
Yeah. Mods are crucial if you want to let your users take your software places you'd never even thought of before.
Section 2870 does *NOT* protect you if you're a software developer. Check out sub section 1. It includes an exemption for the company if your invention relates at the time of conception.... to the employer's business.
The company's case against you is pretty straighforward: "We're in the business of making and selling software. Your invention is software. Hand it over."
And they don't even have to pat you on the head for writing it.
I hope homeboy has more luck trying to exert his rights under the CLC than other people who've fallen for it have had.:(
The thing is (and I'm going to try avoid spoiling the plotline for those who have never read the books) that the inevitable showdown with Sauron and his minions isn't the end of the story. Tolkien went all out in the crafting of the story, and that means he covered not only Big Battles, but also what happens when it's all over and our heroes are trying to get back to their old, normal lives.
And Sauruman being defeated but not eliminated is influential (but by no means necessary) to that part of the story.
To my mind, Jackson can be forgiven for skipping over some stuff from the book to accomodate the movie. Except for leaving out Tom Bombadil. I'LL NEVER FORGIVE THAT! TO ARMS! TO ARMS! REMEMBER THE BOMBADIL!!
Nope, I haven't seen yellow dog (I'm a tightwad and since Macs are waaaaaay more expensive than PCs... just kidding:). What I've seen of other arch support has been the code included in RH's anaconda (and I admit I haven't been spelunking in anaconda for about 6 months or so).
But it's good to know that I've got the option to change arch's in midstream if I wanna.
Thanks also to madmonkey for the info about Alpha support... my XL300 is still running RH 5.2, so I've only ever seen anaconda in action on x86 archs.
I don't know if you've ever actually worked with anaconda, but (like other open source software) it's possible to hammer it do whatever you need. There is support in anaconda for non i386 archs (s390, sparc, and IIRC, vestigial traces of the alpha installer). Yes, it's going to be a pain to implement code to handle new archs (like the PPC), but there are enough examples of how to do it that it should be possible.
The one thing that makes me downright ecstatic in all this is the prospect of being able to use the "kickstart" feature of anaconda for Debian. RH's kickstart is pretty damn flexible (as opposed to FAI, FreeBSD's unattended install mode, Solaris's jumpstart, and even the Winders solutions that are available). With the kickstart, it's possible to build and install a customized system from modular parts (instead of having to rely on image based installs)... and that makes it easy to slide in updates or quickly implement new install types.
Hardware autodetection is abstracted out via kudzu (yes, it's a pain after the OS is installed, but at install time it's a godsend and makes probing hardware programmatically much easier).
On top of that, you can hack up anaconda to do some other "interesting kickstartish type stuff" (in the words of Matt Wilson).
Kudos for the Progeny boys for making this available.:) It's going to enable some cool stuff to be done with Debian.
The whole point of this is to make it so that the RPM database can be aware of what software is installed on your system... not to simplify installation.
If RPM was smart enough to actually go look in/usr/lib/ and see what's there instead of just checking it's own DB and just blankly assuming that if something wasn't installed via RPM it must not exist, then you'd be absolutely right and this would be a solution without a problem.
And let me know if/when you figure that problem out. You could do some cool stuff and make Linux (heck, and *nix-ish OS) more admin friendly almost overnight.
Fair enough... but tell me exactly how RedHat is forcing you to upgrade?
The thing that makes RedHat a million times better than Microsoft is the simple fact that I've got the source code to every product from RedHat that I or the company I work has ever purchased. That means that both using RHN and purchasing a copy of each distro is entirely optional to me. I only have to do that if you want the service, not the software.
To the original poster: Someone else mentioned checking out Debian... this is a valid answer to the problem and one a few folks I know have chosen to go with. Another solution your might want to consider is what the costs of hiring some developers to maintain RH 7.[2|3] beyond it's EOL date would be, or possibly to fork maintain a distro that's based on RHEL (either of those should end up costing less than $350K).
You misunderstand the algorithm. Unlike traditonal lie detector tests where stress indicates an attempt at deciet, in this case I'd say it's almost a requirement for validating the story.
I would expect this to be a stop-gap solution until the technology to tell whether or not the caller has dropped a load in his/her shorts is in place (the smellaphone, maybe?).
Looks like a flub in the wording... the definition is for "Installed System" in section 5:
5. Support Service Conditions. Red Hat may, at its discretion, decline to provide Support Services for Software that has been modified or changed by Customer in any way, except as directed by Red Hat. Red Hat will provide Support Services for Supported Hardware and Platforms only. Red Hat will only provide Support Services for those Installed Systems for which Customer has subscribed under this Agreement.
So s/Server/System/g and it should make more sense.
IANAL, so I don't know if that invalidates the contract or what... but it seems like a fair deal. They just want to get paid for each server that is using their premium service. Just wish whoever wrote up the contracts had been a little more on the ball in making sure the semantics were straight.
On the upside, the 5 year life cycle is a good thing from an administration point of view because you don't have to worry about the product changing its feature-set on you. RH has done a good job of trying to backport security fixes to whatever version they released with the distro (not always possible, though). This is useful because most open source projects reccomend upgrading to the latest version of their software whenever a bug is found and fixed, and this can cause problems. One example was bind 8.2.2... there was a remote exploit in there that was fixed in 8.3.X, but the 8.3.X changed the requirements for how the named.conf was supposed to be configured (I think you had to add in a default TTL or something). With 2000 servers online, rolling our new 8.3.X self-made RPMs actually caused problems since our support staff had to go in and manually update all the customers' named.conf to get it up to snuff. That's what the longer lifecycle is protecting you from.
OTOH, your hands are kind of tied. There is some cool stuff coming down the pipe, and RH will be absolutely stubborn about putting it into their older distros. The line will be "that feature is going into our next version, please upgrade when it's available".
So it all gets back to what you need the distro to do. If you'd rather just have admins on the payroll to make sure the software on the server is up and running instead of keeping a few developers on the payroll and maintaining your own in-house distro (and there are a number of companies that have chosen to take the latter route), then knuckle down, sign up for RHN, and start shelling out for entitlements. Or you can find another distro to use... Debian's stable usually has a long lifetime and they do an excellent job of keeping it patched up.
Just remember that whichever option you choose, there will be much pain and expense, and before AS is EOL'd you will probably be replaced by some young gimp who calls you a "crack head" for making whichever choice it is that you will make.
Tell you what, folks. If no one comes down here to Crazy Darl's Unix Emporium and buys a license in the next hour, I'm gonna club this baby seal. That's right, I'd club a baby seal to make a better deal. And I'll do it, too... cuz I'm Craaaaazy Darl.
1. SCO's stock is currently trading at $3.55 / share 2. SCO's Annual Stockholder Meeting is Friday, May 16 at 9am at their headquarters in Lindon, UT.:-)
I just keep getting this really amusing image of a horde of Linux users showing up for the Suits' Fandango and asking annoying questions like:
"I appreciate your concern for protecting my investment by defending the company's IP and all, but I have to ask.... before dumping all this money into lawyers and court costs, did you even bother to send out an email (which would have been free) to the people who are allegedly infringing on our IP and simply asking them to stop?"
"Does that not hurt our chances of winning any legal decisions since we failed to excercise even a minimal amount of prevention?"
"Does releasing our own version of Linux using the same infringing IP after filing the lawsuit against IBM not hurt our chances of winning?"
"Can you think of one good reason why I, as a shareholder, shouldn't demand Darl's head on a silver platter right now for gross incompetence in how he has handled this situation? Can I call for a vote on firing him, or do I need to sue the Board or something?"
That's a pretty big oversimplification of the matter. Basically, David Cutler left Digital after DEC decided to ax a project he was working on to build a new chip and an OS to go along with it, Microsoft offered him a job to work on the chip and the OS, and Cutler managed to recruit most of his dev team from DEC for the new MS squad.
So saying that NT is just VMS part II isn't really accurate, but the same guilty parties are involved. If you can find it, there was a book called _ShowStopper! The Breakneck Race to Create Windows NT_ that does a pretty good job of chronicling the history of NT during its early days.
I don't think changing it is really going to help. The guy's main complaint is that students aren't aware of the realities of working in the Real World. Better PR isn't going to solve that situation, but giving students a preview of Hell^Wlife in the Real World(tm) will.
Making students work on an open source project (preferrably for longer than just 3 months) would be a great way to do this. I can't think of a better introduction to working on a team, code maintenance, dealing with users and their vague requirements, bug reports consisting of "It doesn't work," and how good of an idea code reuse can be than that. Or at the very least make internships mandatory for the degree.
The bottom line is that you can't tell people some things... you just have to drop them in the tank with the sharks and let them figure it out on their own. Having more realistic fictional characters in mainstream pop culture media isn't going to do that.
Another is the Borda count, a point system devised by the 18th-century French mathematician Jean Charles Borda, which is now used to rank college football and basketball teams.
Oh, that's just brilliant.
Because if there's one thing that everyone can agree on in the US of A, it's who is the top ranked college football team. Yep... that'd remove all ambiguity and uncertainty from the whole election process for sure and we could always be certain of a clear winner.
Outstanding.
I still say that we should get rid of elections and just have the candidates fight for their position in a no-holds-barred cage match... "Let combat be the judge" and all that. Heck, at least it would boost the CSPAN ratings.
A couple of years ago I heard a story on All Things Considered that was about a gentleman who had opened what is basically a free/open repository of genome info on families with autism.
The reason he'd done this is because his son was autistic, and he discovered that biomed research firms were fscking patenting autistic genes to ensure that if a cure was ever found they'd reap huge profits from the treatment. This landrush for intellectual property also had the nasty side effect of killing just about any collaborative research effort for a cure or treatment... no one could get samples to work with. And the companies doing this weren't even trying to protect their own research projects.... they were just sitting on the damn knowledge (or rather, rights to the knowledge).
Now we've got stuff like the CAN Foundation and AGRE, so hopefully there's a shot at developing an answer other than, "Uhhhh.... just stick Rain Man in the looney bin and get on with your life." Hopefully a cure can be found, and failing that a treatment or at least we can figure out what the fsck causes it.
But I can't help but wonder if this "epidemic" might not have been preventable (or at least mitigated) if some greedy bastards had actually used their talents to help other people instead of making the downpayment on the Lexus.
While RedHat and Debian and Caldera and Mandrake were indeed competitors, they weren't our major ones. These weren't the people we had to beat. It was Microsoft, Novell, and Sun (on the low-end). So it was worse than just taking on RedHat.
Also at that time (1999), Linux was starting to be taken seriously by the IT world and you were starting to get a lot of folks from those companies looking in the general direction of Linux. None of the commercial distros had the resources to pick up all of the experimenters and there was more than enough to go around and this is why we got the whole "Big Four" thing (RH, Caldera, TL, and SuSE).
The problem is, though, that you had to get your numbers up and keep them up, or else people like Oracle, IBM, and SGI didn't want to talk to you. Yeah, it was hard in the beginning since RH had the advantage of being a first mover and therefore a larger market, but they weren't invulnerable at that time. But the weak spot was something that could be quickly reinforced if you didn't move fast enough.
So market share, even if it isn't enough to make you number one in the market, is important. You get the revenue benefits (as pointed out in my first post) plus you get the clout with other vendors you need to have on your side.
The second thing is cluster. Even though no one else in *LINUX* space was offering a loadbalanced kit, it wasn't the only load balancer solution out there. And hardware load balancers end up doing a better job and being more reliable than the software based ones... that's just a fact. And if you've got enough money and the need to make sure that your website is up 24/7, you aren't going to cheap out and spend $2K on a software solution that doesn't meet all your needs when for $5K you can get a switch that does.
The one smart thing TL did to salvage the "cluster" situation was to actually produce a real clustering (beowulf, not load balancing) solution with EnFuzion later on. I still don't think it made a lot of money, but I would guess that it did better than TLCluster did.
IMO, the "opportunity" in the load balancing world was a mirage, and I don't really know of any other way to make a buck in any business other than competing head to head with a market leader (save being the first mover in that market).... if you've got some ideas though, I'm all ears.:-)
This sad event neatly illustrates one of the problems with OSS--when a company gets in trouble there's very little incentive for someone like IBM to ride in on their white horse and rescue the company.
Eh? So?
The company was not genetically viable. If it was, it wouldn't have its stakeholders saying, "Give us our money back before you crater so we can cut our losses." There are fundamental structural flaws here that make this company a prime candidate for darwination.
In short, the company failed to earn the right to live.
And while that might be a tragedy to you and anyone else looking at the situation from a business standpoint and hoping to someday profit from it, the fact of the matter is that the contributions that the company made to the free software world (like internationalization work TL did) aren't going away.
To me, this sad event underscores one of the virtues of Open Source Software. Even though the profit-driven organization goes away, the code and contributions it made remains.
Kashif, I'm sincerely sorry to hear about your brother, but he's not the first person at TL who's gotten the shaft for what is basically leadership ineptitude. Hopefully, he'll be one of the last, though.
What strikes me as ironic, though, is that I had a conversation about this back in November of 1999 (and I left the company the next day for this and some other reasons) with the (then) CEO where I warned him about this. My immediate supe warned both him and the board this was going to happen and had his foresight rewarded by being marginalized in the company's decision making structure until he finally got sick of it and left a few months after I did. If I see an article anywhere claiming that the company was blind sided by this or that they blame it all on "market conditions", I won't know whether to laugh or cry.
In the popular street vernacular (at least for 1992:-), let's kick the ballistics. And keep in mind that the costs of development and number of units shipped are just theoretical here and used only to illustrate a point. I didn't have any access to any hard numbers since TL was most definitely not an "openbook" shop and absolutely loved to keep secrets.
Any rate.
You are a small start up company, and you have three (arguably two) products. The first one is a Linux distribution you sell as a the "desktop" version for $50 a shot. To date, in the US, you can claim about 100,000 sales of this particular product. If you focused on this product and this product only, you'd spend about $400,000 to produce it (salaries+benefits, cost of printing the cds, advertising).
Your second product, you bill as a "server". You charge $700 for this product, which is the same as your desktop product except you strip out stuff like XF86, GNOME, kde, pcmcia support, and other things that don't make sense to have on a server. You set up contracts to bundle third party software (like say a commercial mail system, database, or whatever) as a bundle. Since you're not actually developing a whole lot of new stuff here, you can piggy back most of the costs (salaries+benefits) on to the cost of the first product... so the actual cost to produce is around $200K. And today, you've sold a couple hundred units, so an expectation of shipping 1000 units isn't too far beyond the scope of believability (we're keeping numbers round here to make the math easier).
Since there's a lot of overlap, both of those products can arguably be considered the same thing (but from a sales/revenue standpoint, they're distinct).
And now you have the last product. A load balancing product (clustering is beowulf, folks) that can nominally do the same thing as some of the hardware offerings from companies like Cisco. Let's be generous and say that you will need another $200K to develop this product (probably an underestimation) by itself. It's not taking into account that you'll still need the distro to be the vehicle for delivering the product, or a full swing ad campaign. Let's say you've had that on the market for 2 months, and you've only sold 5 copies at about $2K a pop.
So, here are the numbers (sorry about the periods, but there doesn't seem to be any good way to set up a table in a comment with slashcode):
Now, suddenly, the dot-com era begins. VCs are throwing money at you like a Div I schools throws hookers at an all-star quarterback. You get a nice chunk of money, and you have to decide where you invest it. Do you:
a) invest more in the desktop/server product by
hiring more developers and try to increase
those sales number by improving your product
and going head to head with other commercial
distros who are doing the same thing?
OR....
b) invest money into building up a sales and
marketing brigade dedicated solely to the
cluster product in the hopes that you some day
will be able to ship 100,000 units of that at
$2K?
If you chose "a", then you're not qualified to be a TL executive. The answer is "b". And not only do we choose "b", but we start giving away our desktop product for free by reducing the price to $40, giving retail stores a $20 rebate on top of the $10 discount we already give them, and then a $10 rebate inside the box for the customer. That way, we can eliminate that troublesome "revenue" crap and turn our only source of real income into a cost center (because it's about $5 to actually make a boxed product).
And, oh yeah... let's piss off the community by trying to close source everything we can lay a claim to that isn't GPL'd (like our cluster product), release a marketing announcement for the most pedestrian of accomplishments, and generally try avoid supporting our customers with stuff like security updates.
I brought up these concerns and said, "I think it's a mistake to focus so many resources on cluster and ignore the base distro". I was told, "I don't see us being a billion dollar company without doing it."
Have you ever been in a conversation with someone and then just suddenly realized that the person you were talking was going to end up drinking the poisoned Kool-Aid and nothing you said or did would ever keep that from happening? It's a fscking eery feeling, let me tell ya.
Any rate.
To John: don't think about it, or you'll just get frustrated. To Cliff: Toldyaso. To Lonn: please stop before this happens again. And to Rok: thanks for dropping my name from the CREDITS file even though 30% of TL 7 uses RPMs with my name in the changelog, dork.
To everyone else involved in the debacle, best of luck to you and I hope things work out for you.
Christof Pohl was actually distributing an "improved" mod_frontpage apache module. Basically, it did the same thing as the crap that MS/RTR have wedged into the actual apache binary, but it compartmentalized permissions for dealing with the subwebs through the fpexec user (kind of like suexec). I felt a lot safer, and it provided a nice solution for my customers where I could include support for FP on our servers without having to fsck up the apache binary. I have asked RTR to look into making a DSO, but it seems like the request has been ignored...
Any rate, mod_frontpage apparently has been orphaned by Christof. FreeBSD seems to be actively maintaining it, and the have a version that works with FP 5.0 (2002) available in their ports tree... Mandrake has built an RPM based off of the FreeBSD code. I was able to take the SRPM from Mandrake, make some edits to the spec file, and get mod_frontpage running on RH 6.2, 7.1, 7.2., and 7.3 systems from my own RPM. Works great with the official RH errata apache RPMs for each platform, as well as the 1.3.26 RPMs I've created.
So, there are solutions out there. But you'll be waiting a long time if you insist that a vendor hand them to you.:-)
I find it hard to accept that he "believes what he is saying", especially since he never bothered to look at what the states were proposing in the first place.
The fact that he sold out for a "well, we'll talk about it" is funny as hell, though.
As for the code already released under the GPL, that's GPL'ed permanently for everyone.
Actually, no, it isn't. The GPL is just a license, and the copyright holder is free to change the license at any point in time he or she wants to. If it was case that Tilly was the original copyright owner and the company has just now popped up and said, "That belongs to us, we want you to assign the copyright over to us immediately and we're not going to release it as GPL any more", then everything he's released up till now will remain copyright'd to him and thusly would remain GPL'd and you could start a fork right there (like openssh and ssh). No problems, no worries.
The problem, and this is the really insidious part, is that the company is saying that Tilly never EVER held the copyright to any of the work he's done, therefore he didn't have the authority to license the code, therefore anything he's ever published has to come down. And even worse, Tilly finds himself in the very unenviable and legally actionable position of having "leaked" intellectual property to the rest of the world. Even assigning the copyright to the FSF or someone else wouldn't have worked since he didn't have the authority to do it.
That's the part that really, really sucks
Your idea of "underground coding" (ie, writing code and releasing it under an alias or anonymously) is very interesting, though. Technically, you're still not protected, but if the employer isn't able to firmly establish a connection between you and the code, there's no practical way to stop it. I just pity you if you get caught.:-)
Another method that I've heard that I thought was interesting to start each and every project you intend to GPL off with the source code from an already existing GPL project. Ie, grab the source code for GNU ls and start hacking rather than starting from scratch so that even if the company does claim ownership of the copyright they're still bound by the GPL.:-)
The company can't stop it from being distributed, and can't stop others from redistributing. When WE download a GPL'ed code, there's an implicit understanding on OUR part that we have the freedom to redistribute according to the terms of the license; irrelevant of anything else, we have the right to demand our side of the legal bargain.
Again, from the company's standpoint, permission was never given to license the code in the first place, therefore the license is invalid for that code.
Oh well. Like I said in another post, kharma's a bitch.:-) Either that or a common sign of a doomed company is that it tends to start getting pennywise and pound foolish and trying to pull stupid stunts like this. Or maybe both.
Hahahahahahah..... I used to think that section of the CLC afforded me some protection, too.
See, I used to work for a company in Cali. In fact, it was a company that was (and is still) producing a Linux distribution. Despite having profited from Open Source for years, the DotCom boom and all the money it brought to the table encouraged management to start making the Great Intellectual Property Land Grab. One of my co-workers came in one day to discover that the company wanted to claim a small perl he wrote on his laptop in his apartment at 2am as vital Intellectual Property. The company's position was that the "relates to company business" exception could be excercised because it was a "software company" (and we were a "linux distributor" when the contract was signed a few months earlier) and his perl script was obviously software.
If he'd been willing to fight, I'm pretty sure that he could have won as I'm pretty sure that the shady stuff they were trying to do wouldn't stand up in front of a judge... but in the end it was easier just to let the company have the damn script and quit the job to get away from the contract terms rather than to fight the good fight in court and try to pretend that management wasn't going to hold a grudge and try to fsck us later.
On a positive note, the company in question has been relegated to irrelevance by the marketplace and is currently struggling to find something to sell that will actually make enough money that they might one day dare to dream of profitability. And all that despite the fact that they own the IP rights to a small perl script.:-)
Kharma's a bitch.
Any rate, point is... Even if it looks like the law protects you, get them to admit that you're protected in your contract as well. "Is this your signature?" is such a damning question that really undermines your argument in court... I'd much rather the other party be the ones that have to answer it than myself.
Slashdot Mirror
Not quite. It was mentioned in the series (sorry, not enough of a fan to quote the episode title) that the silver droids with the sweeping red LED eye were built by the Cylons, a reptilian race.
The Cylons were fighting some war, so they built themselves a nice little robot army and gave them a simple command: "Go blow stuff up and kill things." So the robots go out and do it and wipe out the Cylons' enemies. Then they start looking for new targets, and, well... the Cylons happen to be the closest civilization, so the bots wipe them out. And then they start wandering the universe looking for stuff to kill.
It's basically the same premise as the Berserker novels by Saberhagen.
Moral of the story: Automating killing machines with just enough intelligence to distinguish between a lifeform and a rock and giving them power supplies that last for hundreds of thousands of years is a really, really, monumentally stupid idea.
And, just for the record, if a huge armada of homicidal robots showed up and started razing my world, I really wouldn't give a crap where they came from until *AFTER* I'd managed to stop 'em.
I've recently been sucked into the competitive gameplay world (where teams organize into divisions, leagues, etc, tournaments are held periodically for cash and prizes, and all that good stuff). As much as I used to chuckle at the thought of "pro gamers", it turns out that there can be just as much nuance to strategy and execution to appreciate in watching a multiplayer video game as there is in watching say a football game. At least to my mind.
One cool thing about mods is that they can be used to improve games to a point where they're suitable for competition. The ETpro mod by bani for the game Wolfenstein: Enemy Territory alters some aspects of gameplay to make it more suitable stopwatch competitions.
The other thing mods can do, and this is kinda neat, is actually add in features to accomodate game spectators. Again, using ETpro as an example, bani included some small changes to help shoutcasters quickly identify players and get stats during the match. A multiview feature was also added so that a spectator could watch the game from several different points of view with a Picture-in-Picture style setup.
In the future, I see mods stepping up to fill in the roles that the original game developers either couldn't think of or didn't want to address because the competition world wasn't their target audience. I can see a mod coming out that can not only handle broadcasting video of the match, but offers optional commentary via an mp3/ogg stream from a caster and presents information kind of in the same way FOX does for football games (current scores, tickers for other matches, league stats for players, etc).
Yeah. Mods are crucial if you want to let your users take your software places you'd never even thought of before.
And I'll say it again:
.... to the employer's business.
:(
Section 2870 does *NOT* protect you if you're a software developer. Check out sub section 1. It includes an exemption for the company if your invention relates at the time of conception
The company's case against you is pretty straighforward: "We're in the business of making and selling software. Your invention is software. Hand it over."
And they don't even have to pat you on the head for writing it.
I hope homeboy has more luck trying to exert his rights under the CLC than other people who've fallen for it have had.
The thing is (and I'm going to try avoid spoiling the plotline for those who have never read the books) that the inevitable showdown with Sauron and his minions isn't the end of the story. Tolkien went all out in the crafting of the story, and that means he covered not only Big Battles, but also what happens when it's all over and our heroes are trying to get back to their old, normal lives.
And Sauruman being defeated but not eliminated is influential (but by no means necessary) to that part of the story.
To my mind, Jackson can be forgiven for skipping over some stuff from the book to accomodate the movie. Except for leaving out Tom Bombadil. I'LL NEVER FORGIVE THAT! TO ARMS! TO ARMS! REMEMBER THE BOMBADIL!!
Nope, I haven't seen yellow dog (I'm a tightwad and since Macs are waaaaaay more expensive than PCs... just kidding :). What I've seen of other arch support has been the code included in RH's anaconda (and I admit I haven't been spelunking in anaconda for about 6 months or so).
But it's good to know that I've got the option to change arch's in midstream if I wanna.
Thanks also to madmonkey for the info about Alpha support... my XL300 is still running RH 5.2, so I've only ever seen anaconda in action on x86 archs.
I don't know if you've ever actually worked with anaconda, but (like other open source software) it's possible to hammer it do whatever you need. There is support in anaconda for non i386 archs (s390, sparc, and IIRC, vestigial traces of the alpha installer). Yes, it's going to be a pain to implement code to handle new archs (like the PPC), but there are enough examples of how to do it that it should be possible.
:) It's going to enable some cool stuff to be done with Debian.
The one thing that makes me downright ecstatic in all this is the prospect of being able to use the "kickstart" feature of anaconda for Debian. RH's kickstart is pretty damn flexible (as opposed to FAI, FreeBSD's unattended install mode, Solaris's jumpstart, and even the Winders solutions that are available). With the kickstart, it's possible to build and install a customized system from modular parts (instead of having to rely on image based installs)... and that makes it easy to slide in updates or quickly implement new install types.
Hardware autodetection is abstracted out via kudzu (yes, it's a pain after the OS is installed, but at install time it's a godsend and makes probing hardware programmatically much easier).
On top of that, you can hack up anaconda to do some other "interesting kickstartish type stuff" (in the words of Matt Wilson).
Kudos for the Progeny boys for making this available.
Erm.
/usr/lib/ and see what's there instead of just checking it's own DB and just blankly assuming that if something wasn't installed via RPM it must not exist, then you'd be absolutely right and this would be a solution without a problem.
The whole point of this is to make it so that the RPM database can be aware of what software is installed on your system... not to simplify installation.
If RPM was smart enough to actually go look in
And let me know if/when you figure that problem out. You could do some cool stuff and make Linux (heck, and *nix-ish OS) more admin friendly almost overnight.
Fair enough... but tell me exactly how RedHat is forcing you to upgrade?
The thing that makes RedHat a million times better than Microsoft is the simple fact that I've got the source code to every product from RedHat that I or the company I work has ever purchased. That means that both using RHN and purchasing a copy of each distro is entirely optional to me. I only have to do that if you want the service, not the software.
To the original poster: Someone else mentioned checking out Debian... this is a valid answer to the problem and one a few folks I know have chosen to go with. Another solution your might want to consider is what the costs of hiring some developers to maintain RH 7.[2|3] beyond it's EOL date would be, or possibly to fork maintain a distro that's based on RHEL (either of those should end up costing less than $350K).
You misunderstand the algorithm. Unlike traditonal lie detector tests where stress indicates an attempt at deciet, in this case I'd say it's almost a requirement for validating the story.
I would expect this to be a stop-gap solution until the technology to tell whether or not the caller has dropped a load in his/her shorts is in place (the smellaphone, maybe?).
Looks like a flub in the wording... the definition is for "Installed System" in section 5:
So s/Server/System/g and it should make more sense.
IANAL, so I don't know if that invalidates the contract or what... but it seems like a fair deal. They just want to get paid for each server that is using their premium service. Just wish whoever wrote up the contracts had been a little more on the ball in making sure the semantics were straight.
Glad you could finally make it. :-)
:-)
On the upside, the 5 year life cycle is a good thing from an administration point of view because you don't have to worry about the product changing its feature-set on you. RH has done a good job of trying to backport security fixes to whatever version they released with the distro (not always possible, though). This is useful because most open source projects reccomend upgrading to the latest version of their software whenever a bug is found and fixed, and this can cause problems. One example was bind 8.2.2... there was a remote exploit in there that was fixed in 8.3.X, but the 8.3.X changed the requirements for how the named.conf was supposed to be configured (I think you had to add in a default TTL or something). With 2000 servers online, rolling our new 8.3.X self-made RPMs actually caused problems since our support staff had to go in and manually update all the customers' named.conf to get it up to snuff. That's what the longer lifecycle is protecting you from.
OTOH, your hands are kind of tied. There is some cool stuff coming down the pipe, and RH will be absolutely stubborn about putting it into their older distros. The line will be "that feature is going into our next version, please upgrade when it's available".
So it all gets back to what you need the distro to do. If you'd rather just have admins on the payroll to make sure the software on the server is up and running instead of keeping a few developers on the payroll and maintaining your own in-house distro (and there are a number of companies that have chosen to take the latter route), then knuckle down, sign up for RHN, and start shelling out for entitlements. Or you can find another distro to use... Debian's stable usually has a long lifetime and they do an excellent job of keeping it patched up.
Just remember that whichever option you choose, there will be much pain and expense, and before AS is EOL'd you will probably be replaced by some young gimp who calls you a "crack head" for making whichever choice it is that you will make.
Have a nice day.
Tell you what, folks. If no one comes down here to Crazy Darl's Unix Emporium and buys a license in the next hour, I'm gonna club this baby seal. That's right, I'd club a baby seal to make a better deal. And I'll do it, too... cuz I'm Craaaaazy Darl.
I'd just like to point out a couple of things:
:-)
1. SCO's stock is currently trading at $3.55 / share
2. SCO's Annual Stockholder Meeting is Friday, May 16 at 9am at their headquarters in Lindon, UT.
I just keep getting this really amusing image of a horde of Linux users showing up for the Suits' Fandango and asking annoying questions like:
"I appreciate your concern for protecting my investment by defending the company's IP and all, but I have to ask.... before dumping all this money into lawyers and court costs, did you even bother to send out an email (which would have been free) to the people who are allegedly infringing on our IP and simply asking them to stop?"
"Does that not hurt our chances of winning any legal decisions since we failed to excercise even a minimal amount of prevention?"
"Does releasing our own version of Linux using the same infringing IP after filing the lawsuit against IBM not hurt our chances of winning?"
"Can you think of one good reason why I, as a shareholder, shouldn't demand Darl's head on a silver platter right now for gross incompetence in how he has handled this situation? Can I call for a vote on firing him, or do I need to sue the Board or something?"
That's a pretty big oversimplification of the matter. Basically, David Cutler left Digital after DEC decided to ax a project he was working on to build a new chip and an OS to go along with it, Microsoft offered him a job to work on the chip and the OS, and Cutler managed to recruit most of his dev team from DEC for the new MS squad.
:-)
So saying that NT is just VMS part II isn't really accurate, but the same guilty parties are involved. If you can find it, there was a book called _ShowStopper! The Breakneck Race to Create Windows NT_ that does a pretty good job of chronicling the history of NT during its early days.
Know thy enemy, and all that.
I don't think changing it is really going to help. The guy's main complaint is that students aren't aware of the realities of working in the Real World. Better PR isn't going to solve that situation, but giving students a preview of Hell^Wlife in the Real World(tm) will.
Making students work on an open source project (preferrably for longer than just 3 months) would be a great way to do this. I can't think of a better introduction to working on a team, code maintenance, dealing with users and their vague requirements, bug reports consisting of "It doesn't work," and how good of an idea code reuse can be than that. Or at the very least make internships mandatory for the degree.
The bottom line is that you can't tell people some things... you just have to drop them in the tank with the sharks and let them figure it out on their own. Having more realistic fictional characters in mainstream pop culture media isn't going to do that.
Another is the Borda count, a point system devised by the 18th-century French mathematician Jean Charles Borda, which is now used to rank college football and basketball teams.
Oh, that's just brilliant.
Because if there's one thing that everyone can agree on in the US of A, it's who is the top ranked college football team. Yep... that'd remove all ambiguity and uncertainty from the whole election process for sure and we could always be certain of a clear winner.
Outstanding.
I still say that we should get rid of elections and just have the candidates fight for their position in a no-holds-barred cage match... "Let combat be the judge" and all that. Heck, at least it would boost the CSPAN ratings.
A couple of years ago I heard a story on All Things Considered that was about a gentleman who had opened what is basically a free/open repository of genome info on families with autism.
The reason he'd done this is because his son was autistic, and he discovered that biomed research firms were fscking patenting autistic genes to ensure that if a cure was ever found they'd reap huge profits from the treatment. This landrush for intellectual property also had the nasty side effect of killing just about any collaborative research effort for a cure or treatment... no one could get samples to work with. And the companies doing this weren't even trying to protect their own research projects.... they were just sitting on the damn knowledge (or rather, rights to the knowledge).
Now we've got stuff like the CAN Foundation and AGRE, so hopefully there's a shot at developing an answer other than, "Uhhhh.... just stick Rain Man in the looney bin and get on with your life." Hopefully a cure can be found, and failing that a treatment or at least we can figure out what the fsck causes it.
But I can't help but wonder if this "epidemic" might not have been preventable (or at least mitigated) if some greedy bastards had actually used their talents to help other people instead of making the downpayment on the Lexus.
I disagree.
:-)
While RedHat and Debian and Caldera and Mandrake were indeed competitors, they weren't our major ones. These weren't the people we had to beat. It was Microsoft, Novell, and Sun (on the low-end). So it was worse than just taking on RedHat.
Also at that time (1999), Linux was starting to be taken seriously by the IT world and you were starting to get a lot of folks from those companies looking in the general direction of Linux. None of the commercial distros had the resources to pick up all of the experimenters and there was more than enough to go around and this is why we got the whole "Big Four" thing (RH, Caldera, TL, and SuSE).
The problem is, though, that you had to get your numbers up and keep them up, or else people like Oracle, IBM, and SGI didn't want to talk to you. Yeah, it was hard in the beginning since RH had the advantage of being a first mover and therefore a larger market, but they weren't invulnerable at that time. But the weak spot was something that could be quickly reinforced if you didn't move fast enough.
So market share, even if it isn't enough to make you number one in the market, is important. You get the revenue benefits (as pointed out in my first post) plus you get the clout with other vendors you need to have on your side.
The second thing is cluster. Even though no one else in *LINUX* space was offering a loadbalanced kit, it wasn't the only load balancer solution out there. And hardware load balancers end up doing a better job and being more reliable than the software based ones... that's just a fact. And if you've got enough money and the need to make sure that your website is up 24/7, you aren't going to cheap out and spend $2K on a software solution that doesn't meet all your needs when for $5K you can get a switch that does.
The one smart thing TL did to salvage the "cluster" situation was to actually produce a real clustering (beowulf, not load balancing) solution with EnFuzion later on. I still don't think it made a lot of money, but I would guess that it did better than TLCluster did.
IMO, the "opportunity" in the load balancing world was a mirage, and I don't really know of any other way to make a buck in any business other than competing head to head with a market leader (save being the first mover in that market).... if you've got some ideas though, I'm all ears.
This sad event neatly illustrates one of the problems with OSS--when a company gets in trouble there's very little incentive for someone like IBM to ride in on their white horse and rescue the company.
Eh? So?
The company was not genetically viable. If it was, it wouldn't have its stakeholders saying, "Give us our money back before you crater so we can cut our losses." There are fundamental structural flaws here that make this company a prime candidate for darwination.
In short, the company failed to earn the right to live.
And while that might be a tragedy to you and anyone else looking at the situation from a business standpoint and hoping to someday profit from it, the fact of the matter is that the contributions that the company made to the free software world (like internationalization work TL did) aren't going away.
To me, this sad event underscores one of the virtues of Open Source Software. Even though the profit-driven organization goes away, the code and contributions it made remains.
HAH.
:-), let's kick the ballistics. And keep in mind that the costs of development and number of units shipped are just theoretical here and used only to illustrate a point. I didn't have any access to any hard numbers since TL was most definitely not an "openbook" shop and absolutely loved to keep secrets.
0 0)....4, 600,0000 0,000)... ..500,000. .(200,000). ...(190,000)
Kashif, I'm sincerely sorry to hear about your brother, but he's not the first person at TL who's gotten the shaft for what is basically leadership ineptitude. Hopefully, he'll be one of the last, though.
What strikes me as ironic, though, is that I had a conversation about this back in November of 1999 (and I left the company the next day for this and some other reasons) with the (then) CEO where I warned him about this. My immediate supe warned both him and the board this was going to happen and had his foresight rewarded by being marginalized in the company's decision making structure until he finally got sick of it and left a few months after I did. If I see an article anywhere claiming that the company was blind sided by this or that they blame it all on "market conditions", I won't know whether to laugh or cry.
In the popular street vernacular (at least for 1992
Any rate.
You are a small start up company, and you have three (arguably two) products. The first one is a Linux distribution you sell as a the "desktop" version for $50 a shot. To date, in the US, you can claim about 100,000 sales of this particular product. If you focused on this product and this product only, you'd spend about $400,000 to produce it (salaries+benefits, cost of printing the cds, advertising).
Your second product, you bill as a "server". You charge $700 for this product, which is the same as your desktop product except you strip out stuff like XF86, GNOME, kde, pcmcia support, and other things that don't make sense to have on a server. You set up contracts to bundle third party software (like say a commercial mail system, database, or whatever) as a bundle. Since you're not actually developing a whole lot of new stuff here, you can piggy back most of the costs (salaries+benefits) on to the cost of the first product... so the actual cost to produce is around $200K. And today, you've sold a couple hundred units, so an expectation of shipping 1000 units isn't too far beyond the scope of believability (we're keeping numbers round here to make the math easier).
Since there's a lot of overlap, both of those products can arguably be considered the same thing (but from a sales/revenue standpoint, they're distinct).
And now you have the last product. A load balancing product (clustering is beowulf, folks) that can nominally do the same thing as some of the hardware offerings from companies like Cisco.
Let's be generous and say that you will need another $200K to develop this product (probably an underestimation) by itself. It's not taking into account that you'll still need the distro to be the vehicle for delivering the product, or a full swing ad campaign. Let's say you've had that on the market for 2 months, and you've only sold 5 copies at about $2K a pop.
So, here are the numbers (sorry about the periods, but there doesn't seem to be any good way to set up a table in a comment with slashcode):
Prod___Units Sold__Price___TtlCost__ Ttl Profit
desktop.....100,000........50......(400,0
server.............1000......700......(2
cluster.................5.....2000....
Now, suddenly, the dot-com era begins. VCs are throwing money at you like a Div I schools throws hookers at an all-star quarterback. You get a nice chunk of money, and you have to decide where you invest it. Do you:
a) invest more in the desktop/server product by
hiring more developers and try to increase
those sales number by improving your product
and going head to head with other commercial
distros who are doing the same thing?
OR....
b) invest money into building up a sales and
marketing brigade dedicated solely to the
cluster product in the hopes that you some day
will be able to ship 100,000 units of that at
$2K?
If you chose "a", then you're not qualified to be a TL executive. The answer is "b". And not only do we choose "b", but we start giving away our desktop product for free by reducing the price to $40, giving retail stores a $20 rebate on top of the $10 discount we already give them, and then a $10 rebate inside the box for the customer. That way, we can eliminate that troublesome "revenue" crap and turn our only source of real income into a cost center (because it's about $5 to actually make a boxed product).
And, oh yeah... let's piss off the community by trying to close source everything we can lay a claim to that isn't GPL'd (like our cluster product), release a marketing announcement for the most pedestrian of accomplishments, and generally try avoid supporting our customers with stuff like security updates.
I brought up these concerns and said, "I think it's a mistake to focus so many resources on cluster and ignore the base distro". I was told, "I don't see us being a billion dollar company without doing it."
Have you ever been in a conversation with someone and then just suddenly realized that the person you were talking was going to end up drinking the poisoned Kool-Aid and nothing you said or did would ever keep that from happening? It's a fscking eery feeling, let me tell ya.
Any rate.
To John: don't think about it, or you'll just get frustrated. To Cliff: Toldyaso. To Lonn: please stop before this happens again. And to Rok: thanks for dropping my name from the CREDITS file even though 30% of TL 7 uses RPMs with my name in the changelog, dork.
To everyone else involved in the debacle, best of luck to you and I hope things work out for you.
Christof Pohl was actually distributing an "improved" mod_frontpage apache module. Basically, it did the same thing as the crap that MS/RTR have wedged into the actual apache binary, but it compartmentalized permissions for dealing with the subwebs through the fpexec user (kind of like suexec). I felt a lot safer, and it provided a nice solution for my customers where I could include support for FP on our servers without having to fsck up the apache binary. I have asked RTR to look into making a DSO, but it seems like the request has been ignored...
Any rate, mod_frontpage apparently has been orphaned by Christof. FreeBSD seems to be actively maintaining it, and the have a version that works with FP 5.0 (2002) available in their ports tree... Mandrake has built an RPM based off of the FreeBSD code. I was able to take the SRPM from Mandrake, make some edits to the spec file, and get mod_frontpage running on RH 6.2, 7.1, 7.2., and 7.3 systems from my own RPM. Works great with the official RH errata apache RPMs for each platform, as well as the 1.3.26 RPMs I've created.
So, there are solutions out there. But you'll be waiting a long time if you insist that a vendor hand them to you. :-)
Since you are positive, that must mean you have proof. Please provide that proof.
On 4of12's behalf: here ya go, dingleberry.
Check this news.com article on the cross examination out.
I find it hard to accept that he "believes what he is saying", especially since he never bothered to look at what the states were proposing in the first place.
The fact that he sold out for a "well, we'll talk about it" is funny as hell, though.
As for the code already released under the GPL, that's GPL'ed permanently for everyone.
Actually, no, it isn't. The GPL is just a license, and the copyright holder is free to change the license at any point in time he or she wants to. If it was case that Tilly was the original copyright owner and the company has just now popped up and said, "That belongs to us, we want you to assign the copyright over to us immediately and we're not going to release it as GPL any more", then everything he's released up till now will remain copyright'd to him and thusly would remain GPL'd and you could start a fork right there (like openssh and ssh). No problems, no worries.
The problem, and this is the really insidious part, is that the company is saying that Tilly never EVER held the copyright to any of the work he's done, therefore he didn't have the authority to license the code, therefore anything he's ever published has to come down. And even worse, Tilly finds himself in the very unenviable and legally actionable position of having "leaked" intellectual property to the rest of the world. Even assigning the copyright to the FSF or someone else wouldn't have worked since he didn't have the authority to do it.
That's the part that really, really sucks
Your idea of "underground coding" (ie, writing code and releasing it under an alias or anonymously) is very interesting, though. Technically, you're still not protected, but if the employer isn't able to firmly establish a connection between you and the code, there's no practical way to stop it. I just pity you if you get caught. :-)
Another method that I've heard that I thought was interesting to start each and every project you intend to GPL off with the source code from an already existing GPL project. Ie, grab the source code for GNU ls and start hacking rather than starting from scratch so that even if the company does claim ownership of the copyright they're still bound by the GPL. :-)
The company can't stop it from being distributed, and can't stop others from redistributing. When WE download a GPL'ed code, there's an implicit understanding on OUR part that we have the freedom to redistribute according to the terms of the license; irrelevant of anything else, we have the right to demand our side of the legal bargain.
Again, from the company's standpoint, permission was never given to license the code in the first place, therefore the license is invalid for that code.
Oh well. Like I said in another post, kharma's a bitch. :-) Either that or a common sign of a doomed company is that it tends to start getting pennywise and pound foolish and trying to pull stupid stunts like this. Or maybe both.
Hahahahahahah..... I used to think that section of the CLC afforded me some protection, too.
See, I used to work for a company in Cali. In fact, it was a company that was (and is still) producing a Linux distribution. Despite having profited from Open Source for years, the DotCom boom and all the money it brought to the table encouraged management to start making the Great Intellectual Property Land Grab. One of my co-workers came in one day to discover that the company wanted to claim a small perl he wrote on his laptop in his apartment at 2am as vital Intellectual Property. The company's position was that the "relates to company business" exception could be excercised because it was a "software company" (and we were a "linux distributor" when the contract was signed a few months earlier) and his perl script was obviously software.
If he'd been willing to fight, I'm pretty sure that he could have won as I'm pretty sure that the shady stuff they were trying to do wouldn't stand up in front of a judge... but in the end it was easier just to let the company have the damn script and quit the job to get away from the contract terms rather than to fight the good fight in court and try to pretend that management wasn't going to hold a grudge and try to fsck us later.
On a positive note, the company in question has been relegated to irrelevance by the marketplace and is currently struggling to find something to sell that will actually make enough money that they might one day dare to dream of profitability. And all that despite the fact that they own the IP rights to a small perl script. :-)
Kharma's a bitch.
Any rate, point is... Even if it looks like the law protects you, get them to admit that you're protected in your contract as well. "Is this your signature?" is such a damning question that really undermines your argument in court... I'd much rather the other party be the ones that have to answer it than myself.