IANAMA (I'm not a materials engineer) but to my best understanding carbon nanotubes come in single walled (SWNT) and multiwalled (MWNT) flavors. The former are what you want for the elevator because they have extraordinary tensile strength and are very light (worthy of noting is that while their *theoretical* tensile strength is 5 times what you need for an elevator - 300GPa - and you need a safety factor of about 2 to actually make one - ~110GPa - the strongest single SWNT made to date is somewhere around 60GPa. I *think*.)
The latter - multiwalled - are much more dense and so will not be fit for an elevator - too heavy. These might actually be of use where strong rigid materials are required, such as construction. Just remember that we construct not out of what is strong but of what is cheap and readily available, hence some places use more wood and others use more concrete, and nobody uses steel except where local cheap materials don't cut it (lile.. skyscrapers).
Would be nice to have someone who has up-to-date info clear this up.
Rocket-based technology is still eons away from being anywhere near the cheap and safe ballpark. At the current rate it is evolving as a commercial service (which it is), it will take forever to realize your dream. Even were we to design a 2005-era rocket from scratch using all the newest and latest tech, It's still *barely* acheiving its goal using insanely sophisticated and insanely expensive equipment.
What I hope we will see in the next 20 years is something that will make space economically feasible, to several magnitudes of scale more entities than it is feasible for today. Be it elevators (which at this point looks most likely), rotovators or any other technology with the ability to scale. Switching to such a technology offers to drastically improve the economic and safety standing of the system, will make it attractive for any business that has a reason to poke a finger in space (0-grav manufacturing of pharmaceuticals and silicone, remote-area power delivery via tourism, mining solar system, inhabiting solar system, space-tourism, etc) and we'll live to see a gold rush bigger than dot.com, going straight up.
It's gonna be fun to watch (especially if you buy into this industry early;-)
I believe what grandparent wanted to say is that materials will be light enough to coat a 5-story-tall mecha in armor that could withstand 2005-era firepower as presented by the very informative long post above (never mind that penetration power will probbably evolve no less than material science and by the time we have the stuff, we'll need 4 meters equivalent of solid steel on the front).
Mecha could then be [a] made of stuff light enough that it can actually both be built to move and be able to withstand modern firepower. [b] still very very heavy, enough to remove a tank from business by stepping on it.
The tearing-up-the-terrain could be overlooked by a fictional scenario of battles on some space-rock with no roads or other terrain we care to preserve.
It's the scaling of mechanics that we're not going to overcome so easily. The infrastructural material strength required to withstand a joint flexing during walking (never mind leaping), in a 5- or 6-figure-ton mecha, would require the supporting skeleton to be made of something much much much stronger than anything we know (including nanotubes).
I want to see real Atlas-class mechs. And ones that really walk, not roll.
Lift foot. Lightly position foot on tank. Shift weight on foot. Note audible Shzplat-a-*poof*. Lift foot. Take two steps back. Observe 60-Ton flat metal plate. Repeat.
Then again I never exactly got the math of the weight thing figured out. A modern MBT (think M1) weighs in around the 60 Ton ballpark, whereas an Atlas which dwarfs it quite considerably (enough to quite literally smudge it by stepping on it) said "100 Ton". Maybe I forgot to multiply by 1000 or something? Judging by its size, it should weigh closer to the several (if not several dozen) kiloton range.
1. You buy an appliance that can manufacture a hell of a lot of stuff you use on a daily basis.
Plus it can manufacture itself, so if you really want to, you can make a couple more. Personally, I think it'll take a while before a commercial entity will offer a product that will compete with it. It makes no financial sense. They'd sooner make an appliance that can manufacture anything *but* itself. But let's skip to the point where someone built a self-replicating machine, and much like gmail accounts, sooner than not everybody's got one for himself and 6 or 50 to give out to his friends.
2. You stop buying a major percentage of the stuff you use on a daily basis.
3. Some companies, from many sectors, who make stuff that you've been using go out of business. Not a bad thing in itself. Many whip-makers went out of business when cars were invented and horses stopped being the preferred means of travel. It's the natural course of technology.
4. People make more and more stuff that they need out of raw materials.
5. Raw materials become more scarce. No more plastics in the bin. Plastic Recycling plants get no more plastic cuz nobody's throwing it away anymore. The price on anything made from materials useable at home rises dramatically.
6. Things we cannot manufacture at home yet which we still buy and are made of said materials, say your car, go up in price due to higher material costs.
So you pay less for your plates and but more for your car.
7. New content market: Designs. Expect the DIAA (you heard it here first). We'll be pirating our dishwasher plans from P2P, paying for raw materials, making it at home. Designers will be watermarking (plasticmarking) their designs. Ripping groups will be removing the marks. Machines whose designs we steal and which we built will call home over the net to activate unless we download a hacked version.
This time around, it won't be just the music industry clinging to their antiquated business model. Suddenly every company that sold you a plate, cup, PDA case or pen will want to sell you PENS (where you pay per item) rather than a design of a pen (where you pay once to get the design). First they'll laugh at you, then they'll fight you (and this time around will probbably have a humongous lobby compared to what the RIAA has today), and then you'll win of course, those of the companies that managed to strip off the bulk of the no longer neccesary manufacturing from the price of what you buy making it to the next round. Change indeed. Like software or music, all over again.
Most large P2P content comes in a zip files, which have 40 RAR volumes of 3 Ripped CD ISO's, which, even in the completely unrealistic scenario of being all decipherable by any commercial product, will be stripped of all CSS, Macrovision and future watermarking or any other annoyance bits in the first place by the guy who ripped them.
All that, of course, on P2P protocols such as donkey and torrent, that most commercial products such as Checkpoint FW-1 don't yet scan on application level (and I'd wager new protocols will arrive sooner than app-level scanning implementations for existing ones in these products). P2P does NOT use HTTP and FTP.
*WHAT* precisely are the ISP's expected to scan for that will successfully differ legal from illegal content on any modern network, and using what tools?
These attempts to control information by dinkhead suits who watch too many episodes of NCIS is amusing in a pathetic way.
They just don't get it, you cannot control information with the Internet around. No more than you can control the consumption of oxygen on the planet. It is completely unenforceable.
And even if all of this was remotely and conceivably close to somehow being financially, politically and legally feasible to enforce (which it is a long long long way from), even if encryption was somehow moved out of the way, FreeNet is always lurking in the darker shadows underneath the mainstream P2P networks, always waiting for people to be forced to use it...
I live in Australia, where the police quite recently started spontaneously checking drivers for traces of drugs.
Due to the inaccuracy of the tests (and one too many highly-publicized false positives), people (who didn't take drugs) sued the shit out of them for determining that they did. Bluntly put, they had their asses handed over to them tied with a cute red ribbon, and the program that was receiving a lot of publicity and marketed with a hugh media campaign was practically scrapped.
Same with comms sniffing. Try and do that, get whatever results you get, but act on it and the amount of false positives will make sure your ass is handed over to you.
Besides, that'll just push ip-layer encryption (IPSec) into production faster.
On one hand, they keep mentioning SLI, SLI, SLI. On the other hand, These mobos are server mobos, loaded with stuff I frankly could do without, like SATA2 (IIRC, the fastest hard-drives out there are barely 50% of the way saturating a SATA1 link), Firewire, 8 memory slots and PCI-X.
What SLI croud need is a simple mobo with a simple feature set, a couple of PCIex1 slots, the two full x16's, the USB, audio, double GbE & the works as offerd by the 2200, and a couple of 939-pin sockets coming from a decent mobo maker like Gigabyte that doesn't charge double for it's badge (read: ASUS, TYAN, etc.)
All in a sub-200$ box.
Then all of us who can dish out the cash for an SLI setup accompanied by two (wish nVidia drivers supported more) monitors/projectors, crank up the resolution to max on both and enjoy the scenery.
>>So long as you do not comingle GPL code with your own you are free to make use of it in your business and can even make some serious coin at it.
It's you who is ignorant my friend. What you're talking about is called LGPL, not GPL. GPL applies to linked code as well, which is what makes it unuseable.
Your definition of free is thus so vague as to where it starts and where it ends that it's meaningless.
I could [somewhat sarcastically] argue Microsoft is free because you have the freedom to purchase their products.
In reality freedom (liber, as in speech) is a scalar, not a boolean. It's not so much free/unfree as it is 'how free?'. GPL offers significantly more freedom than a purchased MS prooduct, while for those who need even more freedom (namely software dev companies) LGPL-licensed-code offers that extra they need in order to be able to use it, and BSD/X11 offers even more, though the bit they need lies in the GPLLGPL delta.
As for where one man's freedom ends and another begins, I'm a bit vague. You're implying the contributer gets more freedom by pegging GPL on his contirbution than by pegging LGPL. I beg to differ. He gets no extra freedom by forcing whoever uses his contribution to publish everything compiled/linked with it. By minimuzing this to 'everything compiled with it' he'd be getting a significantly larger userbase (coming from the corporate side), and raking in more contributions from that side by giving them a realistic price they'd gladly pay (read: contribute back), as opposed to putting an infinite pricetag and scaring most of them away.
GPL is about the freedom to use contributions to wage an ideological war on those who pay programmers. Had the entire GPL codebase been LGPL'd, it would have let the closed-source world both use and contribute back in a non-suicidal way, and IMHO would have served itself better. This war only feeds the ideologists.
No, it was very much to be used in a commercial product distributed for profit. It was also a negligible part of the entire product, and opensourcing the product due to using whatever small GPL modules was unrealistic.
cabextract for a product that needed cab extraction capability.
Was written as a CLI utility, and I'd have gladly wrapped in a library and had my employer contribute the code in that library.
We'd also have a libcab on Freshmeat today had it been LGPL'd.
Re:There's a missing fifth fundamental freedom
on
Being Free is Hard to Do
·
· Score: 2, Insightful
No. First, your point is valid, there is not and cannot be an absolute freedom. The BSD license is also limiting to an extent. I cannot claim I wrote the code for instance. I cannot prevent the author from continuing to give it out for free, even if I relicense the same code and sell it.
For myself, I'm not implying anything other that I'm personally greatly appreciative of having an opensource package called "unzip" with a BSD license, which saved me the need to waste 2 months trying to figure out and implement ZIP compression.
I am implying that for my [now-ex] employer, or the example of Qualystem (which I chose for the exmaple simply because I know their software doesn't have an opensource equivalent) GPL is not free as in beer. For them this is the difference between having something people are willing to pay for and not having it, thus the implications of using GPL cost horrendous amounts of money.
I am using the X11/BSD license as a reference point, as they provide similar opensource contributions, only without the unrealistic pricetag designed to wage war on closed-source companies.
This can, I guess, be somehow regarded as a 'Free as in speech' issue (Is my employer free to use the code for his purpose or not?) as well, but the real issue here is money, so I guess the easy way to look at it is the beer context.
No you don't, and you'd make me very happy if I'm wrong about this and you actually do. Pray tell.
Show me ONE piece of open-source that offers a WinNT storage driver (the kind you put in when pressing F6 while installing Win2K/XP), which maps the system drive on the MS box to a UNIX SAMBA share, TFTP server, NFS server, or any other network-shared resource. In fact, the only competition they've got is BXP, also a commercial product, only they use a generic MS-only fileserver, whereas Qualystem LAN-PC3 work on any UNIX as their server.
The only thing you might be aware of is doing so using DOS SMB drivers for Win9x, much like old Novell systems. As you might or might not be aware, NT does not run over DOS.
You're right not to trust vendors, but my information comes from my own [5-month-old] research, not from vendor marketing.
Not at all, my flamebaiting friend. A. The business in question pays quite a few coders. It is nothing even remotely close to the parasite you're implying. It is attempting to achieve the best product offering by tapping into external already-invented wheels there's no need to reinvent, whether through purchased SDKs or open source where licensing allows, and internal coding for the rest (which promptly constitutes about 90% of the product anyway). This is much smarter than writing everything yourself, and had you inherited this business tomorrow, you would be making exactly the same decisions. The particular product in question does not allow for selling service, and this software is the source of this business's revenue. I doubt whether you'd shut down your own successful (albeit due to not being opensource) business. Don't expect others to.
B. "You take the bread out of their mouths..." That's a totally idiotic claim to make. Said BSD coders willingly contributed their code for this exact kind of use by having the license they chose allow for this.
C. "...and then whine and moan that more people ought to be donating all their hard work to the purpose of your profitability" I don't believe I ever said anybody 'aught' to be donating anything. I believe in people getting paid, and I greatly respect people who give (because they want to, not because someone thinks they 'aught' to), whether it's code to GNU or money to a charity. When I donate money, the thing that really annoys me is people who think I 'aught' to. I merely pointed out that what is marketed as Free by the GPL is quite unfree to a specific group of people (which you seem religiously keen on demonizing) whereas an otherwise compatible license such as the X11/BSD one *is* indeed free to them. I also pointed out these people are not bad people, and their profit is crucial to programmers (including said contributers) getting paid hard cash at the end of the month. By clearly being ignorant to this connection, I'd probbably be safe to assume you don't get paid to code. That, in turn, gives you very little moral right to pass judgement the way you do.
D. "I've got news for you: Your company is dead and doesn't know it yet." You're probbably right. But hery, you and I will someday both be dead as well. This does NOT, however, imply you should run to the kitchen, stick that double-barrel in your mouth and pump your skull full of lead now, should it? Neither sholuld a closed-source company that is currently making a profit rid itself of its ability to make money just because there of a possible impending doom somewhere along its future. For many companies, the one in question among them, there is yet plenty of time to adjust to new business models.
About the scumbag bit, I'd say you're the biggest unemployed hypocrite I've met on slashdot yet. Had you been employed (and looking after your employers interests), you'd be singing a very different song.
Re:There's a missing fifth fundamental freedom
on
Being Free is Hard to Do
·
· Score: 3, Insightful
Here's an example for you.
A [quite real] company called Qualystem makes a network-boot suite that allows you to boot Windows XP clients off a UNIX/Linux-based server.
There is no opensource alternative for their product. This is said to emphasize there is neither something wrong with their business model, nor are they on the verge of drowning under free&better OS alternatives like traditional UNIX operating systems or products with opensource counterparts are. Their business relies SOLELY on selling software. Not service. Not hardware. Just software.
Such a company cannot use so much as a line of GPL code without legally binding themselves to commit suicide, hence for them GPL != free. For them, GPL is the most expensive thing in the world.
Taking a competitive/unique product from a closed-source business model to an open source one != free.
Yes, the organizations you mentioned do this. Some, like RH, built around it from day one, others, like all the UNIX vendors you mentioned, did it out of a lack of choice, where continuing to compete with the OS world simply started costing them too much, so they went to "If you can't beat them, join them". All the ones you mentioned are the ones who found alternate sources of income. Don't forget the ones that know they don't (but for whose competing products the OS community still has no equivalent, thus their existing and only income is not in any immediate danger. GPL is only available to them through suicide.
RH sells service. IBM sell HW and service, as do SUN, SGI, HP etc. NOKIA sells phones. Not all companies, and not all software, can be marketed as a service like a UNIX system can. Not all sell hardware like Nokia. Without the ability to market the software itself, what do you expect a software company to sell?
Re:There's a missing fifth fundamental freedom
on
Being Free is Hard to Do
·
· Score: 3, Insightful
0. Calm down. Your contribution shouldn't cost you your health. 1. Your contribution is greatly appreciated. 2. I [as a coder who is a paid representative of my employer] would greatly like to use your contribution, free, as you intended. 3. As far as my employer is concerned, your contribution, as useful as it is, is not worth GPL-stamping [e.g. opensourcing] 2 million lines of source, millions of dollars he invested into of engineer time, into a product with a competing edge, which is keeping his business afloat. Asking this of my employer, be it fair or not in your view, is unrealistic. He will not commit financial suicide for you. Don't take that as an insult, you wouldn't commit suicide for me either, even if I donated something to you, and I'm perfectly okay with that. 4. The LGPL allows me to take your code, wrap it in a library, and opensource the code of that library alone, not everything [i.e. my employer's entire product] that links to it. My employer uses, My employer gives something [albeit not his head on a silver platter] back to the community. This is what I referred to as fair and compromise. >> "What do "fair" and "compromise" mean on your planet?" Does your planet have this thing called "Economy", "Businesses" and "Programmers who have children to feed and thus need an employer to pay them a salary"? Mine does. 5. Being able to use your contribution with no strings attached is what makes your contribution free to him, as opposed to costing him something (even if you view that as free because what he pays never reaches your pocket). The term free [as in beer] is not a result of what you get. It's a result of what it costs him.
Which brings me back to my original post - GPL is free to any members of the OS club. X11/BSD is free to everyone, closed-source software businesses (with which one or two slashdotters may have crossed paths during their careers) as well. And it's you, the contributer, who decides who you want to share with.
Think of a TV ad saying "X if FREE, if only you join our club (which promptly costs money to anyone who's, say, asian)". Free, eh?
There's a missing fifth fundamental freedom
on
Being Free is Hard to Do
·
· Score: 4, Insightful
At least as considered by any business who'd want to ingegrate anything, even as miniscule as a c file with 3 functions that calculate CRC.
What's missing is just like "The freedom to improve the program, and release your improvements to the public" only the opposite - "The freedom to improve the program, and not release your improvements to the public" (or sell said improvements to the public for profit)"
This is the issue commonly called copylefting. What it comes down to is "Free for anyone who's part of our [opensource] club" as set forth by the GPL (If you're a Checkpoint dev, a legal obligation to release all/parts-of the source code of the product makes whatever ran you into that obligation anything but free), or "Free to anyone. Period." as set forth by X11/modified-BSD licenses. The latter offer the fifth freedom.
The obligation [e.g. lack of freedom] to integrate GPL code with [often immense] business-owned closed code serves on one hand to spur [few, IMHO] businesses to go opensource, while keeping a dark "obligation" cloud over Open Source that scares the rest away. I personally ran into this dillema at my former workplace. The result was us using BSD-licensed and commercial solutions, while [to my great dismay] avoiding GPL-code like the plague. The LGPL is a fair compromise, unfortunately few projects use it. Sometimes you need code from a GPL app, and you're willing to wrap it in a library yourself (and offer that library's code to the public) but since the original dev never considered this and just slapped the GPL on his work, and you can't use it (whereas had he done so with LGPL, you would be able to do so).
The conclusion (which promptly earned me two flamebait mods last time I said this unliked piece of truth here) is that everything GPL is quite unfree to those [nice, evil, fill your own description] people who pay us coders our salaries and feed our families.
I, personally, as a coder who wants to tap open source where I work, would definitely like it to be otherwise. For the GNU codebase to be as legal-obligation-free and accessible as the X11-ilcensed or mod-BSD-licensed codebase (and a big thank you to anyone altruistic enough to use those licenses on his donated code).
Copylefting is a mixed bag. On one hand it forces (Forcing != "Free as in speech") users of GPL code into GPL'ing their contributions, while on the other hand it scares them away for fear the code they invested money in developing is opened up, a mental barrier they're not always keen on crossing. Sometimes this seems moral (if you fixed a bug in open source, share it and let everyone benefit) whereas sometimes it doesn't (Should I legally have to open-source 2 million lines of closed code of a product with a unique business advantage for using an MD5 routine that was under GPL? Not very attractive, nor very free (AsInSpeech) )
What this iRiver issue speaks of is very scary, as it makes the GPL even scarier for the corporate market decision-makers (who control the big bucks yet who don't always share our techie affinity for Free/Open stuff), and makes a lot of MS FUD come true.
GPL applies its "viral" nature (i.e. license "spreads" to your work if you use anything that has it) once you either incorporate GPL source or link against GPL libraries.
LGPL is more convenient, as it applies its viral nature to source code alone, not linked libraries, so if you want to "contain" what your closed-source company needs to share and yet use open source, you use LGPL libraries or wrap LGPL code in your own library and open-source that library. The rest of the product is not subject to opening that way.
What this article implies is that by running under a GPL kernel with the GPL'd multipurpose busybox binary, all software on the OS immediately becomes GPL. Whoa. Going there is BAD BAD BAD. That's exactly that silver bullet the MS-FUD department is looking so desperately for.
IANAL, but does GPL provide for this? If so, it's even less attaractive to the business environment than the way MS FUD lawyers described the "viral nature" of GPL to date.
You just led me to another simlpe observation I haven't made before:
The second *big* difference in this respect between IE and FF is the goal of the project.
IE is written by a commercial entity. Their goal is to get maximal revenue for minimum investment. This is not a bad thing, it's the underlying principle of what we call an economy and the presence of which differs us from Afghanistan. If they recon adding certain features to a product will not gain them anything substancial ($$$), they will not allocate the resources to do so. Period. The way I see it, it's totally understandable. I perform the same decision with my money every day.
FF is written by a group of volunteer engineers. Their goal, at this stage at least, is their product. Making it stick out due to is superb engineering.
For me, as an engineer, this definitely makes FF preferrable.
The point however is, this advantage also drops off once a certain critical-mass has been reached - only this time it's MS who closes the gap by becoming better rather than FF by becoming less secure.
Once enough people leave IE due to it lacking whatever it is they want, MS will reprioritize adding said features and IE will catch up, or, if we look at what happened in the past and the fact that the open community has more creativity and less red tape than MS does, they will probbably wait for FF to set the new requirements, then implement them in a robust way everybody from poweruser to 'Joe Sixpack' can use, surpass FF by a couple of steps, and FF will fall back to being in the same 2nd place it is in now. Then they'll lay back, and FF or its future counterpart will redefine 'cutting-edge' again. Ad infinum.
In short, this second advantage of FF is just as circumstantial as the first.
Slashdot Mirror
IANAMA (I'm not a materials engineer) but to my best understanding carbon nanotubes come in single walled (SWNT) and multiwalled (MWNT) flavors.
The former are what you want for the elevator because they have extraordinary tensile strength and are very light (worthy of noting is that while their *theoretical* tensile strength is 5 times what you need for an elevator - 300GPa - and you need a safety factor of about 2 to actually make one - ~110GPa - the strongest single SWNT made to date is somewhere around 60GPa. I *think*.)
The latter - multiwalled - are much more dense and so will not be fit for an elevator - too heavy. These might actually be of use where strong rigid materials are required, such as construction. Just remember that we construct not out of what is strong but of what is cheap and readily available, hence some places use more wood and others use more concrete, and nobody uses steel except where local cheap materials don't cut it (lile.. skyscrapers).
Would be nice to have someone who has up-to-date info clear this up.
If I remember correctly, Trusted Computing is baaad, at least as far as we /.'ers are concerned.
Why is everyone bashing Microsoft for dropping it?
Rejoyce!
Rocket-based technology is still eons away from being anywhere near the cheap and safe ballpark. At the current rate it is evolving as a commercial service (which it is), it will take forever to realize your dream. Even were we to design a 2005-era rocket from scratch using all the newest and latest tech, It's still *barely* acheiving its goal using insanely sophisticated and insanely expensive equipment.
;-)
What I hope we will see in the next 20 years is something that will make space economically feasible, to several magnitudes of scale more entities than it is feasible for today. Be it elevators (which at this point looks most likely), rotovators or any other technology with the ability to scale. Switching to such a technology offers to drastically improve the economic and safety standing of the system, will make it attractive for any business that has a reason to poke a finger in space (0-grav manufacturing of pharmaceuticals and silicone, remote-area power delivery via tourism, mining solar system, inhabiting solar system, space-tourism, etc) and we'll live to see a gold rush bigger than dot.com, going straight up.
It's gonna be fun to watch (especially if you buy into this industry early
A hollywood studio is a commercial entity. It sells a commodity for profit.
What you're charged for covers making the film. Let's call that X. Add to that the cost of the distribution.
If they distribute it using Bittorrent, distribution costs, say, n.
If they distribute it via their own server, distributiion costs 2n.
What would you rather do, pay X+n and share or X+2n and download direct? I assume with time both options will be viable.
Of course, for a negligible n (in relation to X), you'd expect a company that cares about its clients to provide you with the X+2n option.
Quad-yearly national survey concludes americans support George Bush.
What's new?
>> and too small to easily hit with an aircraft
Q: Do you know how snipers are eliminated in modern warfare?
A: Artillery, Artillery and yet more artillery.
I believe what grandparent wanted to say is that materials will be light enough to coat a 5-story-tall mecha in armor that could withstand 2005-era firepower as presented by the very informative long post above (never mind that penetration power will probbably evolve no less than material science and by the time we have the stuff, we'll need 4 meters equivalent of solid steel on the front).
Mecha could then be
[a] made of stuff light enough that it can actually both be built to move and be able to withstand modern firepower.
[b] still very very heavy, enough to remove a tank from business by stepping on it.
The tearing-up-the-terrain could be overlooked by a fictional scenario of battles on some space-rock with no roads or other terrain we care to preserve.
It's the scaling of mechanics that we're not going to overcome so easily. The infrastructural material strength required to withstand a joint flexing during walking (never mind leaping), in a 5- or 6-figure-ton mecha, would require the supporting skeleton to be made of something much much much stronger than anything we know (including nanotubes).
I want to see real Atlas-class mechs. And ones that really walk, not roll.
Lift foot.
Lightly position foot on tank.
Shift weight on foot.
Note audible Shzplat-a-*poof*.
Lift foot.
Take two steps back.
Observe 60-Ton flat metal plate.
Repeat.
Then again I never exactly got the math of the weight thing figured out. A modern MBT (think M1) weighs in around the 60 Ton ballpark, whereas an Atlas which dwarfs it quite considerably (enough to quite literally smudge it by stepping on it) said "100 Ton". Maybe I forgot to multiply by 1000 or something? Judging by its size, it should weigh closer to the several (if not several dozen) kiloton range.
1. You buy an appliance that can manufacture a hell of a lot of stuff you use on a daily basis.
Plus it can manufacture itself, so if you really want to, you can make a couple more. Personally, I think it'll take a while before a commercial entity will offer a product that will compete with it. It makes no financial sense. They'd sooner make an appliance that can manufacture anything *but* itself. But let's skip to the point where someone built a self-replicating machine, and much like gmail accounts, sooner than not everybody's got one for himself and 6 or 50 to give out to his friends.
2. You stop buying a major percentage of the stuff you use on a daily basis.
3. Some companies, from many sectors, who make stuff that you've been using go out of business. Not a bad thing in itself. Many whip-makers went out of business when cars were invented and horses stopped being the preferred means of travel. It's the natural course of technology.
4. People make more and more stuff that they need out of raw materials.
5. Raw materials become more scarce. No more plastics in the bin. Plastic Recycling plants get no more plastic cuz nobody's throwing it away anymore. The price on anything made from materials useable at home rises dramatically.
6. Things we cannot manufacture at home yet which we still buy and are made of said materials, say your car, go up in price due to higher material costs.
So you pay less for your plates and but more for your car.
7. New content market: Designs. Expect the DIAA (you heard it here first).
We'll be pirating our dishwasher plans from P2P, paying for raw materials, making it at home. Designers will be watermarking (plasticmarking) their designs. Ripping groups will be removing the marks. Machines whose designs we steal and which we built will call home over the net to activate unless we download a hacked version.
This time around, it won't be just the music industry clinging to their antiquated business model. Suddenly every company that sold you a plate, cup, PDA case or pen will want to sell you PENS (where you pay per item) rather than a design of a pen (where you pay once to get the design). First they'll laugh at you, then they'll fight you (and this time around will probbably have a humongous lobby compared to what the RIAA has today), and then you'll win of course, those of the companies that managed to strip off the bulk of the no longer neccesary manufacturing from the price of what you buy making it to the next round.
Change indeed.
Like software or music, all over again.
They can't do it with what's in place today.
Most large P2P content comes in a zip files, which have 40 RAR volumes of 3 Ripped CD ISO's, which, even in the completely unrealistic scenario of being all decipherable by any commercial product, will be stripped of all CSS, Macrovision and future watermarking or any other annoyance bits in the first place by the guy who ripped them.
All that, of course, on P2P protocols such as donkey and torrent, that most commercial products such as Checkpoint FW-1 don't yet scan on application level (and I'd wager new protocols will arrive sooner than app-level scanning implementations for existing ones in these products). P2P does NOT use HTTP and FTP.
*WHAT* precisely are the ISP's expected to scan for that will successfully differ legal from illegal content on any modern network, and using what tools?
These attempts to control information by dinkhead suits who watch too many episodes of NCIS is amusing in a pathetic way.
They just don't get it, you cannot control information with the Internet around. No more than you can control the consumption of oxygen on the planet. It is completely unenforceable.
And even if all of this was remotely and conceivably close to somehow being financially, politically and legally feasible to enforce (which it is a long long long way from), even if encryption was somehow moved out of the way, FreeNet is always lurking in the darker shadows underneath the mainstream P2P networks, always waiting for people to be forced to use it...
Actually, they did, and they had.
I live in Australia, where the police quite recently started spontaneously checking drivers for traces of drugs.
Due to the inaccuracy of the tests (and one too many highly-publicized false positives), people (who didn't take drugs) sued the shit out of them for determining that they did. Bluntly put, they had their asses handed over to them tied with a cute red ribbon, and the program that was receiving a lot of publicity and marketed with a hugh media campaign was practically scrapped.
Same with comms sniffing. Try and do that, get whatever results you get, but act on it and the amount of false positives will make sure your ass is handed over to you.
Besides, that'll just push ip-layer encryption (IPSec) into production faster.
On one hand, they keep mentioning SLI, SLI, SLI.
On the other hand, These mobos are server mobos, loaded with stuff I frankly could do without, like SATA2 (IIRC, the fastest hard-drives out there are barely 50% of the way saturating a SATA1 link), Firewire, 8 memory slots and PCI-X.
What SLI croud need is a simple mobo with a simple feature set, a couple of PCIex1 slots, the two full x16's, the USB, audio, double GbE & the works as offerd by the 2200, and a couple of 939-pin sockets coming from a decent mobo maker like Gigabyte that doesn't charge double for it's badge (read: ASUS, TYAN, etc.)
All in a sub-200$ box.
Then all of us who can dish out the cash for an SLI setup accompanied by two (wish nVidia drivers supported more) monitors/projectors, crank up the resolution to max on both and enjoy the scenery.
That'll be the day.
>>So long as you do not comingle GPL code with your own you are free to make use of it in your business and can even make some serious coin at it.
It's you who is ignorant my friend. What you're talking about is called LGPL, not GPL. GPL applies to linked code as well, which is what makes it unuseable.
I suggest you take the time to read up.
Your definition of free is thus so vague as to where it starts and where it ends that it's meaningless.
I could [somewhat sarcastically] argue Microsoft is free because you have the freedom to purchase their products.
In reality freedom (liber, as in speech) is a scalar, not a boolean. It's not so much free/unfree as it is 'how free?'.
GPL offers significantly more freedom than a purchased MS prooduct, while for those who need even more freedom (namely software dev companies) LGPL-licensed-code offers that extra they need in order to be able to use it, and BSD/X11 offers even more, though the bit they need lies in the GPLLGPL delta.
As for where one man's freedom ends and another begins, I'm a bit vague. You're implying the contributer gets more freedom by pegging GPL on his contirbution than by pegging LGPL. I beg to differ. He gets no extra freedom by forcing whoever uses his contribution to publish everything compiled/linked with it. By minimuzing this to 'everything compiled with it' he'd be getting a significantly larger userbase (coming from the corporate side), and raking in more contributions from that side by giving them a realistic price they'd gladly pay (read: contribute back), as opposed to putting an infinite pricetag and scaring most of them away.
GPL is about the freedom to use contributions to wage an ideological war on those who pay programmers. Had the entire GPL codebase been LGPL'd, it would have let the closed-source world both use and contribute back in a non-suicidal way, and IMHO would have served itself better. This war only feeds the ideologists.
No, it was very much to be used in a commercial product distributed for profit.
It was also a negligible part of the entire product, and opensourcing the product due to using whatever small GPL modules was unrealistic.
cabextract for a product that needed cab extraction capability.
Was written as a CLI utility, and I'd have gladly wrapped in a library and had my employer contribute the code in that library.
We'd also have a libcab on Freshmeat today had it been LGPL'd.
No.
First, your point is valid, there is not and cannot be an absolute freedom. The BSD license is also limiting to an extent. I cannot claim I wrote the code for instance. I cannot prevent the author from continuing to give it out for free, even if I relicense the same code and sell it.
For myself, I'm not implying anything other that I'm personally greatly appreciative of having an opensource package called "unzip" with a BSD license, which saved me the need to waste 2 months trying to figure out and implement ZIP compression.
I am implying that for my [now-ex] employer, or the example of Qualystem (which I chose for the exmaple simply because I know their software doesn't have an opensource equivalent) GPL is not free as in beer. For them this is the difference between having something people are willing to pay for and not having it, thus the implications of using GPL cost horrendous amounts of money.
I am using the X11/BSD license as a reference point, as they provide similar opensource contributions, only without the unrealistic pricetag designed to wage war on closed-source companies.
This can, I guess, be somehow regarded as a 'Free as in speech' issue (Is my employer free to use the code for his purpose or not?) as well, but the real issue here is money, so I guess the easy way to look at it is the beer context.
No you don't, and you'd make me very happy if I'm wrong about this and you actually do. Pray tell.
Show me ONE piece of open-source that offers a WinNT storage driver (the kind you put in when pressing F6 while installing Win2K/XP), which maps the system drive on the MS box to a UNIX SAMBA share, TFTP server, NFS server, or any other network-shared resource. In fact, the only competition they've got is BXP, also a commercial product, only they use a generic MS-only fileserver, whereas Qualystem LAN-PC3 work on any UNIX as their server.
The only thing you might be aware of is doing so using DOS SMB drivers for Win9x, much like old Novell systems. As you might or might not be aware, NT does not run over DOS.
You're right not to trust vendors, but my information comes from my own [5-month-old] research, not from vendor marketing.
Not at all, my flamebaiting friend.
A. The business in question pays quite a few coders. It is nothing even remotely close to the parasite you're implying.
It is attempting to achieve the best product offering by tapping into external already-invented wheels there's no need to reinvent, whether through purchased SDKs or open source where licensing allows, and internal coding for the rest (which promptly constitutes about 90% of the product anyway). This is much smarter than writing everything yourself, and had you inherited this business tomorrow, you would be making exactly the same decisions. The particular product in question does not allow for selling service, and this software is the source of this business's revenue. I doubt whether you'd shut down your own successful (albeit due to not being opensource) business. Don't expect others to.
B. "You take the bread out of their mouths..."
That's a totally idiotic claim to make. Said BSD coders willingly contributed their code for this exact kind of use by having the license they chose allow for this.
C. "...and then whine and moan that more people ought to be donating all their hard work to the purpose of your profitability"
I don't believe I ever said anybody 'aught' to be donating anything. I believe in people getting paid, and I greatly respect people who give (because they want to, not because someone thinks they 'aught' to), whether it's code to GNU or money to a charity. When I donate money, the thing that really annoys me is people who think I 'aught' to.
I merely pointed out that what is marketed as Free by the GPL is quite unfree to a specific group of people (which you seem religiously keen on demonizing) whereas an otherwise compatible license such as the X11/BSD one *is* indeed free to them. I also pointed out these people are not bad people, and their profit is crucial to programmers (including said contributers) getting paid hard cash at the end of the month. By clearly being ignorant to this connection, I'd probbably be safe to assume you don't get paid to code. That, in turn, gives you very little moral right to pass judgement the way you do.
D. "I've got news for you: Your company is dead and doesn't know it yet."
You're probbably right. But hery, you and I will someday both be dead as well. This does NOT, however, imply you should run to the kitchen, stick that double-barrel in your mouth and pump your skull full of lead now, should it? Neither sholuld a closed-source company that is currently making a profit rid itself of its ability to make money just because there of a possible impending doom somewhere along its future. For many companies, the one in question among them, there is yet plenty of time to adjust to new business models.
About the scumbag bit, I'd say you're the biggest unemployed hypocrite I've met on slashdot yet.
Had you been employed (and looking after your employers interests), you'd be singing a very different song.
Here's an example for you.
A [quite real] company called Qualystem makes a network-boot suite that allows you to boot Windows XP clients off a UNIX/Linux-based server.
There is no opensource alternative for their product. This is said to emphasize there is neither something wrong with their business model, nor are they on the verge of drowning under free&better OS alternatives like traditional UNIX operating systems or products with opensource counterparts are.
Their business relies SOLELY on selling software. Not service. Not hardware. Just software.
Such a company cannot use so much as a line of GPL code without legally binding themselves to commit suicide, hence for them GPL != free. For them, GPL is the most expensive thing in the world.
Taking a competitive/unique product from a closed-source business model to an open source one != free.
Yes, the organizations you mentioned do this. Some, like RH, built around it from day one, others, like all the UNIX vendors you mentioned, did it out of a lack of choice, where continuing to compete with the OS world simply started costing them too much, so they went to "If you can't beat them, join them". All the ones you mentioned are the ones who found alternate sources of income. Don't forget the ones that know they don't (but for whose competing products the OS community still has no equivalent, thus their existing and only income is not in any immediate danger. GPL is only available to them through suicide.
RH sells service. IBM sell HW and service, as do SUN, SGI, HP etc. NOKIA sells phones.
Not all companies, and not all software, can be marketed as a service like a UNIX system can. Not all sell hardware like Nokia. Without the ability to market the software itself, what do you expect a software company to sell?
0. Calm down. Your contribution shouldn't cost you your health.
1. Your contribution is greatly appreciated.
2. I [as a coder who is a paid representative of my employer] would greatly like to use your contribution, free, as you intended.
3. As far as my employer is concerned, your contribution, as useful as it is, is not worth GPL-stamping [e.g. opensourcing] 2 million lines of source, millions of dollars he invested into of engineer time, into a product with a competing edge, which is keeping his business afloat. Asking this of my employer, be it fair or not in your view, is unrealistic. He will not commit financial suicide for you. Don't take that as an insult, you wouldn't commit suicide for me either, even if I donated something to you, and I'm perfectly okay with that.
4. The LGPL allows me to take your code, wrap it in a library, and opensource the code of that library alone, not everything [i.e. my employer's entire product] that links to it. My employer uses, My employer gives something [albeit not his head on a silver platter] back to the community. This is what I referred to as fair and compromise.
>> "What do "fair" and "compromise" mean on your planet?"
Does your planet have this thing called "Economy", "Businesses" and "Programmers who have children to feed and thus need an employer to pay them a salary"? Mine does.
5. Being able to use your contribution with no strings attached is what makes your contribution free to him, as opposed to costing him something (even if you view that as free because what he pays never reaches your pocket). The term free [as in beer] is not a result of what you get. It's a result of what it costs him.
Which brings me back to my original post - GPL is free to any members of the OS club. X11/BSD is free to everyone, closed-source software businesses (with which one or two slashdotters may have crossed paths during their careers) as well. And it's you, the contributer, who decides who you want to share with.
Think of a TV ad saying "X if FREE, if only you join our club (which promptly costs money to anyone who's, say, asian)".
Free, eh?
At least as considered by any business who'd want to ingegrate anything, even as miniscule as a c file with 3 functions that calculate CRC.
What's missing is just like "The freedom to improve the program, and release your improvements to the public" only the opposite -
"The freedom to improve the program, and not release your improvements to the public" (or sell said improvements to the public for profit)"
This is the issue commonly called copylefting.
What it comes down to is "Free for anyone who's part of our [opensource] club" as set forth by the GPL (If you're a Checkpoint dev, a legal obligation to release all/parts-of the source code of the product makes whatever ran you into that obligation anything but free), or "Free to anyone. Period." as set forth by X11/modified-BSD licenses. The latter offer the fifth freedom.
The obligation [e.g. lack of freedom] to integrate GPL code with [often immense] business-owned closed code serves on one hand to spur [few, IMHO] businesses to go opensource, while keeping a dark "obligation" cloud over Open Source that scares the rest away. I personally ran into this dillema at my former workplace. The result was us using BSD-licensed and commercial solutions, while [to my great dismay] avoiding GPL-code like the plague.
The LGPL is a fair compromise, unfortunately few projects use it. Sometimes you need code from a GPL app, and you're willing to wrap it in a library yourself (and offer that library's code to the public) but since the original dev never considered this and just slapped the GPL on his work, and you can't use it (whereas had he done so with LGPL, you would be able to do so).
The conclusion (which promptly earned me two flamebait mods last time I said this unliked piece of truth here) is that everything GPL is quite unfree to those [nice, evil, fill your own description] people who pay us coders our salaries and feed our families.
I, personally, as a coder who wants to tap open source where I work, would definitely like it to be otherwise. For the GNU codebase to be as legal-obligation-free and accessible as the X11-ilcensed or mod-BSD-licensed codebase (and a big thank you to anyone altruistic enough to use those licenses on his donated code).
Wishful thinking I guess...
Copylefting is a mixed bag. On one hand it forces (Forcing != "Free as in speech") users of GPL code into GPL'ing their contributions, while on the other hand it scares them away for fear the code they invested money in developing is opened up, a mental barrier they're not always keen on crossing.
Sometimes this seems moral (if you fixed a bug in open source, share it and let everyone benefit) whereas sometimes it doesn't (Should I legally have to open-source 2 million lines of closed code of a product with a unique business advantage for using an MD5 routine that was under GPL? Not very attractive, nor very free (AsInSpeech) )
What this iRiver issue speaks of is very scary, as it makes the GPL even scarier for the corporate market decision-makers (who control the big bucks yet who don't always share our techie affinity for Free/Open stuff), and makes a lot of MS FUD come true.
GPL applies its "viral" nature (i.e. license "spreads" to your work if you use anything that has it) once you either incorporate GPL source or link against GPL libraries.
LGPL is more convenient, as it applies its viral nature to source code alone, not linked libraries, so if you want to "contain" what your closed-source company needs to share and yet use open source, you use LGPL libraries or wrap LGPL code in your own library and open-source that library. The rest of the product is not subject to opening that way.
What this article implies is that by running under a GPL kernel with the GPL'd multipurpose busybox binary, all software on the OS immediately becomes GPL. Whoa. Going there is BAD BAD BAD.
That's exactly that silver bullet the MS-FUD department is looking so desperately for.
IANAL, but does GPL provide for this? If so, it's even less attaractive to the business environment than the way MS FUD lawyers described the "viral nature" of GPL to date.
Spoken like a true engineer.
You just led me to another simlpe observation I haven't made before:
The second *big* difference in this respect between IE and FF is the goal of the project.
IE is written by a commercial entity. Their goal is to get maximal revenue for minimum investment.
This is not a bad thing, it's the underlying principle of what we call an economy and the presence of which differs us from Afghanistan.
If they recon adding certain features to a product will not gain them anything substancial ($$$), they will not allocate the resources to do so. Period. The way I see it, it's totally understandable. I perform the same decision with my money every day.
FF is written by a group of volunteer engineers. Their goal, at this stage at least, is their product. Making it stick out due to is superb engineering.
For me, as an engineer, this definitely makes FF preferrable.
The point however is, this advantage also drops off once a certain critical-mass has been reached - only this time it's MS who closes the gap by becoming better rather than FF by becoming less secure.
Once enough people leave IE due to it lacking whatever it is they want, MS will reprioritize adding said features and IE will catch up, or, if we look at what happened in the past and the fact that the open community has more creativity and less red tape than MS does, they will probbably wait for FF to set the new requirements, then implement them in a robust way everybody from poweruser to 'Joe Sixpack' can use, surpass FF by a couple of steps, and FF will fall back to being in the same 2nd place it is in now. Then they'll lay back, and FF or its future counterpart will redefine 'cutting-edge' again. Ad infinum.
In short, this second advantage of FF is just as circumstantial as the first.