"The patent system is similar in many respects to totalitarian or facist governments that censor their people and kill millions of jews. I forgot about that."
Now you really got it.
It *really* censors truly innovative companies out of fear of a too much expensive trial (who is not violating Ms patent on double click, for instance?) And it *really* kills millions of people both directly and indirectly since they can't pay for terribly overprized medicines due to patent strongholding.
"OK... didn't know that. But then how does it resolve that dependency if it is missing?"
Just the same way its deb counterpart (dpkg is its name) does: not my job; do nothing.
There's the extended misconception that rpm is something equivalent to apt. Surprise: it is not.
rpm (the program) is the package manager for rpm (the package format). dpkg (the program) is the package manager for deb (the package format).
And both of them have more or less the same functionality. Of course, neither of them will manage dependency resolution and package aquisition.
Now: apt (or aptitude, or dselect) is the dependecy resolution manager and package aquisitor for the Debian distribution and derivatives. It doesn't install packages, but binds to the package manager and its able to order dependency trees and knows (when properly configured) about how and where get the packages it will pass to the package manager for their installation.
There are, of course, tools with that kind of functionality for the rpm package manager like yum, dpkg-rpm, up2date, etc.
".deb files can include multiple dependency resolutions while.rpm can only have one. I was using original poster's language."
Which is not a problem since rpm packages depend on files while deb depend on other packages so if an rpm says it depends on libfoo.so.1 it's because it depends on libfoo.so.1 and nothing else would do. On the other hand a deb can depend on anyone out of a dozen packages that provide libfoo.so.1.
"I disagree. I have never had a pencil give me a blue screen of death."
That's because it's under the "minimum technology" level.
While a pencil is unable to give you one of those technically advanced "blue screen of death", a more evolutioned fountain pen can (over your shirt, even).
"Do new features need to be implemented, or is it alright if the community is only able to step up to provide bug fixes?"
Quite a good point.
Closed source, by its own nature is ill of featuritis. No matter how good the software is, it *needs* new features in order for the next release to be marketeed and "stay in the wave", so to say.
Open source, on the other hand *can* suffer featuritis too, but it's not a *must*. If the codebase already fits a nice niche, there's no need to touch it (bugfixing apart). A paradigmatic example: CVS. Of course there are different points of view about how an SCM tool has to be; of course different products for different niches can be developed (and *are* developed). But it fills its niche quite good and there's no pressure to develop CVS 12 with new and astounding features. Other products with similar "phylosophy" can arise (say Subversion); different products for the same niche from a different point of view can arise (say arch and derivatives), but there's no new CVS version "with e-mail capabilities", and there's no need for it, either. And as long as it fits a niche for somebody with the ability to hunt bugs on it, it will be used and it will be maintained (how many people miss privative software from the 80's and 90's "oh, I enjoyed [some software from Qualcomm, or the Borland from old, or your favourite here]", I'd be happy if I could use something like that now... well with open source you still could use it -vi, emacs or cvs' code base are not younger, but still they're there).
Take a mid-range desktop with enough hard disk space for your needs, and hire a proven record free-lancer for a day for the set up and about one hour a week for maintenance.
Provided you hire the proper person, all your email-related problems will fade away.
"The majority of businesses are small businesses lf less than 50 employees. If they have to have 33 "computer people" because they do all their own stuff internally, they're less competitive than their competitor, who has one "local geek" and hires everyone else on an as-needed basis."
But they both are less competitive than the competitor that has one "local geek" that knows his job and can do all their own stuff internally without resorting (except, maybe, some coding, from time to time) to externalities.
Seriously, where is the IT world going if a 50-less company can't do all their stuff with only one (two max: one sysadmin and one helpdesk/hardware) IT people?
"They'll go for this because it makes sense for them."
"instead of allowing the greedy cheats to downgrade my service"
You rose up a good comparation. What I can't understand are the conclussions you extract out of it. Instead of concluding that crappy "unlimited" broadband are fraudulent too and a clause auction is proper, you accept the 'statu quo' from the movie renter too!
If they advertise "2MB unlimited", the only way I can be a "greedy cheater" is if I manage to get 3MB unlimited. If they promise "2MB unlimited" but then they make provisions (be it the "short letter" in the contract, a "fairness algorith" or anything else) so I can't really get the promised "2MB unlimited", the "greedy cheaters" can't be anyone but them!
Of course the same is valid when they promise "no-limits movies" but then they make provisions for "no-limits" being "five movies a month" instead.
"You sound like the kind of guy that takes a doggie bag to an all-you-can-eat buffet"
No. It is obviously not the same case. It is all you can EAT, not all you can take away.
"and sues the restaurant because the waitress stops clearing your table as quickly after your twentieth plate."
No. It is obviously not the same case. It is all you can EAT, not all we will CLEAN.
The most stupid side of it is that most probably you are not a corporate owner, so you are on the recieving end too, still you defend such unethical practices.
If you go to an "all-you-can-eat" restaurant, pay your bill and then discover you are not hungry at all, you assume you lost and that's all. But then, somehow, there must be some protection for the bussiness if a gargantuan fellow arrives and eats the restaurant's week benefit! A bussiness is a bet: you offer a service and bet you'll make more money than the running costs. But then, you can win your bet or you can loose it; you just don't break the laws so you always win. If you don't feel comfortable with a particular bet (like no-limits movies at your home, even when you know that you will loose money if people ends up asking for too many films), then you should make a more conservative bet (like "up to six movies a week"): fraud cannot be an option.
"As long as a contract is well understood by both parties and voluntarilty entered into, it's fine.
NF have discovered they can't operate an entirely flat-rate service and have modified their behaviour accordingly."
But the question here is that most probably the corporation knew that from the very beginning and has maliciously made false promises to their clients. That's fraud.
I can almost imagine the corporate meeting:
-Hey! we can promise "ALL THE MOVIES YOU CAN SEE" -But we can't promise that: we don't take benefit on each movie, but a flat rate; the most films we send, the less we gain, and beyond certain point (pretty excel graph here) we loose money! -C'mon... most of the "hardcore" pirates already have wideband connection to Internet, with emule and whatnot: we can go with it
[six months later]
-See? we are loosing money, as I told you. We must change our advertisements rigth now. Maybe we can tell "up to X films a month" or maybe we can have different plans so the more you get, the more you pay... -And loose the marketing advantage of an "all you can eat" motto? You must be crazy! No: what we will do is maintain our policy and unilaterally break our contract with those that take the most of us... on a not so obvious way, of course. -But, but... that's fraud! -(Homer Simpson-like accent)Only if they get us, old chap, only if they get us...
"But NetFlix, clearly the biggest player in that market, has already realized that they can't make it worth being in that business with people like that"
So, instead of changing their advertisements (something, you know, like "up to six movies a week to your home"), they LAY.
"How would you build an Egyptian tomb with today's knowledge?"
By not building it, of course.
"There's still no shortage of thieves who would be willing to break in and steal any valuable items"
Still, I'd bet Lincoln's tomb remains undisturbed doesn't it?
But I disgress; I'm Spanish, and we (disgracefully) have one of the best current tries for a "modern pyramid": Franco's burial monument at "Valle de Los Caídos" (Valley of the Fallen) "the most colossal architecture work built in Europe in the 20th century" (well, it was built as a perennial monument for the Civil War fallen -of course, the victorious party fallen). From the wikipedia: rests of 40.000 causalties of the Spanish Civil War (mostly, the winning party); build on slavery by 12.000 prisioners from the losing party, too many of them dead on the works, was designed by Muguruza to equal "the grandeur of the monuments of old, which defy time and forgetfulness" (so it is indeed a "modern piramyd"); it's cross is 152m high (the biggest of the world).
"I don't think convincing the PHBs to use free software is as big a problem as a few years back; just show them the number (money saved)"
True.
"A bigger problem I can see is the management do not want to give anything back to the community [...] where I work, we are not allowed to contribute the local mod we made to some free software back to the community."
You should read yourself. Just show them the numbers. Show a guesstimate about how much money will cost you maintain your private mods as the "base" software evolutions, and how much will be the savings on returning the changes to "the community" so you no longer have to maintain them.
"You can retro-fit LFS with package management easily"
Thus, it is LFS no more.
On the other hand, let's assume it is easy indeed to add package management to LFS. Then, "add" package management to a distribution that *already* has package management in place must be even *easier*. So, again, no place for LFS.
Still, where "stuff that works" really matters (banks, nuclear plants, air traffic control...) too much of them still stand over the "thingies" those "expensive graduate educated Unix admins" created back in the 70's and 80's, and multimillion and even multibillion efforts to substitute them with those modern embelled & enwhisteled systems from the generation of the microsoftian "cheaper spotty kids certified on Windows" have been terribly sounded fiascos.
Maybe it's time for management and beancounters to rethink the situation a bit, don't you think so? (cheaper too usually means more expensive in the long run)
PHB's question: "OK, so if it's free, how do the people who build the distro make money?"
Techie's correct answer: That's *their* problem, not mine or my employer's. All I'm interested in is it's technical suitability. For "enterprise-related questions" all I can say is that if it is good enough for Google to rely their multibillion dollar income cow on it, it surely is good enough for us.
You don't want the company you play with to provide you the "secure token", or else you fastly will end with a ton o'such tokens: you want to be the one to provide *them* the tokens so, somehow, you can provide them all using the same one.
"the hard bit is the politics of it all"
Which, down to the bare bones reads like this:
There's the perception of quite a lot of money around this, so quite a lot of companies want to be the ones taking their piece of the cake (or the biggest piece of the cake if allowed). That you will end up with a worse, uncomfortable and more expensive service is just neglegible "collateral damage".
"Someone can sign their build of a kernel. There's no GPLv3 violation in doing so and distributing that binary, because the signature isn't required to actually use the signed kernel."
Again, that's thin ice.
GPL clearly states that I have the right to access the source code that made up the binary you got me, right?
Now, how can I be really sure that the binaries you passed me are truly build from the source you get me, unless I'm able to follow your procedures, that is, gain access to your secret key, sign the code with it by myself and contrast the results? Maybe a tribunal would have enough by asserting that "operatively" the program signed with your key is equivalent to the program signed with my key... or maybe not.
"Since the FSF is the only entity capable of modifying or releasing future versions of the GPL (as stated in the GPL), we have to trust they will not tamper with the spirit of the license."
No, we don't need to trust this. You can't take a paragraph from a contract and say "that's all with it". You need to see it in context.
Now: what does point 9 say? "9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns."
You see, if the FSF publishes a new version of dissimilar spirit (of course, that should have to be decieded under a tribunal on the worst case) that would be license violation, thus void and unforceable. So if you say "this is distributed under the GPLv2 or later" that's not what it seems, a blank note, since any later GPL version can't be "just anything" but only something similar in spirit (which is quite clearly stated in the preamble) or it won't be a valid relicenseable candidate; at the very least this reasonement would give you a solid position in court, wich is neither more neither less than if you had to fight against any inane corporation that just took your code away.
Slashdot Mirror
"The patent system is similar in many respects to totalitarian or facist governments that censor their people and kill millions of jews. I forgot about that."
Now you really got it.
It *really* censors truly innovative companies out of fear of a too much expensive trial (who is not violating Ms patent on double click, for instance?)
And it *really* kills millions of people both directly and indirectly since they can't pay for terribly overprized medicines due to patent strongholding.
Sixth rule: So what? Now you are dead.
Forth rule: And now, I'll have to kill you!
"OK... didn't know that. But then how does it resolve that dependency if it is missing?"
Just the same way its deb counterpart (dpkg is its name) does: not my job; do nothing.
There's the extended misconception that rpm is something equivalent to apt. Surprise: it is not.
rpm (the program) is the package manager for rpm (the package format).
dpkg (the program) is the package manager for deb (the package format).
And both of them have more or less the same functionality. Of course, neither of them will manage dependency resolution and package aquisition.
Now:
apt (or aptitude, or dselect) is the dependecy resolution manager and package aquisitor for the Debian distribution and derivatives. It doesn't install packages, but binds to the package manager and its able to order dependency trees and knows (when properly configured) about how and where get the packages it will pass to the package manager for their installation.
There are, of course, tools with that kind of functionality for the rpm package manager like yum, dpkg-rpm, up2date, etc.
Next time, please, compare apples to apples.
".deb files can include multiple dependency resolutions while .rpm can only have one. I was using original poster's language."
Which is not a problem since rpm packages depend on files while deb depend on other packages so if an rpm says it depends on libfoo.so.1 it's because it depends on libfoo.so.1 and nothing else would do. On the other hand a deb can depend on anyone out of a dozen packages that provide libfoo.so.1.
Of course, *specially* here this had to come out.
Thanks for your dedication, Mr Anonymous Coward.
"I disagree. I have never had a pencil give me a blue screen of death."
That's because it's under the "minimum technology" level.
While a pencil is unable to give you one of those technically advanced "blue screen of death", a more evolutioned fountain pen can (over your shirt, even).
"I would DEFINITELY be a little bit concerned about the open source enterprise that had just been bought"
And that's different from any other software company acquisition/absortion exactly how?
Well, at least, for an open sourced product you know you are in exact the same position PLUS the source code being freely avaliable.
"Do new features need to be implemented, or is it alright if the community is only able to step up to provide bug fixes?"
Quite a good point.
Closed source, by its own nature is ill of featuritis. No matter how good the software is, it *needs* new features in order for the next release to be marketeed and "stay in the wave", so to say.
Open source, on the other hand *can* suffer featuritis too, but it's not a *must*. If the codebase already fits a nice niche, there's no need to touch it (bugfixing apart). A paradigmatic example: CVS. Of course there are different points of view about how an SCM tool has to be; of course different products for different niches can be developed (and *are* developed). But it fills its niche quite good and there's no pressure to develop CVS 12 with new and astounding features. Other products with similar "phylosophy" can arise (say Subversion); different products for the same niche from a different point of view can arise (say arch and derivatives), but there's no new CVS version "with e-mail capabilities", and there's no need for it, either. And as long as it fits a niche for somebody with the ability to hunt bugs on it, it will be used and it will be maintained (how many people miss privative software from the 80's and 90's "oh, I enjoyed [some software from Qualcomm, or the Borland from old, or your favourite here]", I'd be happy if I could use something like that now... well with open source you still could use it -vi, emacs or cvs' code base are not younger, but still they're there).
"any recommendations or endorsements?"
Take a mid-range desktop with enough hard disk space for your needs, and hire a proven record free-lancer for a day for the set up and about one hour a week for maintenance.
Provided you hire the proper person, all your email-related problems will fade away.
"The majority of businesses are small businesses lf less than 50 employees. If they have to have 33 "computer people" because they do all their own stuff internally, they're less competitive than their competitor, who has one "local geek" and hires everyone else on an as-needed basis."
But they both are less competitive than the competitor that has one "local geek" that knows his job and can do all their own stuff internally without resorting (except, maybe, some coding, from time to time) to externalities.
Seriously, where is the IT world going if a 50-less company can't do all their stuff with only one (two max: one sysadmin and one helpdesk/hardware) IT people?
"They'll go for this because it makes sense for them."
Even if it doesn't make sense at all.
"They should do what they do with the rack mounted search appliance"
They'll do, no doubt about it.
That's only a "beta", you know...
"instead of allowing the greedy cheats to downgrade my service"
You rose up a good comparation. What I can't understand are the conclussions you extract out of it. Instead of concluding that crappy "unlimited" broadband are fraudulent too and a clause auction is proper, you accept the 'statu quo' from the movie renter too!
If they advertise "2MB unlimited", the only way I can be a "greedy cheater" is if I manage to get 3MB unlimited. If they promise "2MB unlimited" but then they make provisions (be it the "short letter" in the contract, a "fairness algorith" or anything else) so I can't really get the promised "2MB unlimited", the "greedy cheaters" can't be anyone but them!
Of course the same is valid when they promise "no-limits movies" but then they make provisions for "no-limits" being "five movies a month" instead.
"You sound like the kind of guy that takes a doggie bag to an all-you-can-eat buffet"
No. It is obviously not the same case. It is all you can EAT, not all you can take away.
"and sues the restaurant because the waitress stops clearing your table as quickly after your twentieth plate."
No. It is obviously not the same case. It is all you can EAT, not all we will CLEAN.
The most stupid side of it is that most probably you are not a corporate owner, so you are on the recieving end too, still you defend such unethical practices.
If you go to an "all-you-can-eat" restaurant, pay your bill and then discover you are not hungry at all, you assume you lost and that's all. But then, somehow, there must be some protection for the bussiness if a gargantuan fellow arrives and eats the restaurant's week benefit! A bussiness is a bet: you offer a service and bet you'll make more money than the running costs. But then, you can win your bet or you can loose it; you just don't break the laws so you always win. If you don't feel comfortable with a particular bet (like no-limits movies at your home, even when you know that you will loose money if people ends up asking for too many films), then you should make a more conservative bet (like "up to six movies a week"): fraud cannot be an option.
"As long as a contract is well understood by both parties and voluntarilty entered into, it's fine.
NF have discovered they can't operate an entirely flat-rate service and have modified their behaviour accordingly."
But the question here is that most probably the corporation knew that from the very beginning and has maliciously made false promises to their clients. That's fraud.
I can almost imagine the corporate meeting:
-Hey! we can promise "ALL THE MOVIES YOU CAN SEE"
-But we can't promise that: we don't take benefit on each movie, but a flat rate; the most films we send, the less we gain, and beyond certain point (pretty excel graph here) we loose money!
-C'mon... most of the "hardcore" pirates already have wideband connection to Internet, with emule and whatnot: we can go with it
[six months later]
-See? we are loosing money, as I told you. We must change our advertisements rigth now. Maybe we can tell "up to X films a month" or maybe we can have different plans so the more you get, the more you pay...
-And loose the marketing advantage of an "all you can eat" motto? You must be crazy! No: what we will do is maintain our policy and unilaterally break our contract with those that take the most of us... on a not so obvious way, of course.
-But, but... that's fraud!
-(Homer Simpson-like accent)Only if they get us, old chap, only if they get us...
"But NetFlix, clearly the biggest player in that market, has already realized that they can't make it worth being in that business with people like that"
So, instead of changing their advertisements (something, you know, like "up to six movies a week to your home"), they LAY.
And still, somehow, you find this reasonably.
By not building it, of course.
"There's still no shortage of thieves who would be willing to break in and steal any valuable items"Still, I'd bet Lincoln's tomb remains undisturbed doesn't it?
But I disgress; I'm Spanish, and we (disgracefully) have one of the best current tries for a "modern pyramid": Franco's burial monument at "Valle de Los Caídos" (Valley of the Fallen) "the most colossal architecture work built in Europe in the 20th century" (well, it was built as a perennial monument for the Civil War fallen -of course, the victorious party fallen). From the wikipedia: rests of 40.000 causalties of the Spanish Civil War (mostly, the winning party); build on slavery by 12.000 prisioners from the losing party, too many of them dead on the works, was designed by Muguruza to equal "the grandeur of the monuments of old, which defy time and forgetfulness" (so it is indeed a "modern piramyd"); it's cross is 152m high (the biggest of the world).
"I don't think convincing the PHBs to use free software is as big a problem as a few years back; just show them the number (money saved)"
True.
"A bigger problem I can see is the management do not want to give anything back to the community [...] where I work, we are not allowed to contribute the local mod we made to some free software back to the community."
You should read yourself. Just show them the numbers. Show a guesstimate about how much money will cost you maintain your private mods as the "base" software evolutions, and how much will be the savings on returning the changes to "the community" so you no longer have to maintain them.
"You can retro-fit LFS with package management easily"
Thus, it is LFS no more.
On the other hand, let's assume it is easy indeed to add package management to LFS. Then, "add" package management to a distribution that *already* has package management in place must be even *easier*. So, again, no place for LFS.
"As far as they know, they got LESS."
Still, where "stuff that works" really matters (banks, nuclear plants, air traffic control...) too much of them still stand over the "thingies" those "expensive graduate educated Unix admins" created back in the 70's and 80's, and multimillion and even multibillion efforts to substitute them with those modern embelled & enwhisteled systems from the generation of the microsoftian "cheaper spotty kids certified on Windows" have been terribly sounded fiascos.
Maybe it's time for management and beancounters to rethink the situation a bit, don't you think so? (cheaper too usually means more expensive in the long run)
PHB's question: "OK, so if it's free, how do the people who build the distro make money?"
Techie's correct answer: That's *their* problem, not mine or my employer's. All I'm interested in is it's technical suitability. For "enterprise-related questions" all I can say is that if it is good enough for Google to rely their multibillion dollar income cow on it, it surely is good enough for us.
"designing the protocol is the easy bit"
Which, down to the bare bones reads like this:
You don't want the company you play with to provide you the "secure token", or else you fastly will end with a ton o'such tokens: you want to be the one to provide *them* the tokens so, somehow, you can provide them all using the same one.
"the hard bit is the politics of it all"
Which, down to the bare bones reads like this:
There's the perception of quite a lot of money around this, so quite a lot of companies want to be the ones taking their piece of the cake (or the biggest piece of the cake if allowed). That you will end up with a worse, uncomfortable and more expensive service is just neglegible "collateral damage".
"FreeBSD ported"
Well, I think I know what you mean, but sorry no, the FreeBSD autoporting daemon should be portd.
(Yes, it was a disgraced attempt to make a joke)
"Someone can sign their build of a kernel. There's no GPLv3 violation in doing so and distributing that binary, because the signature isn't required to actually use the signed kernel."
Again, that's thin ice.
GPL clearly states that I have the right to access the source code that made up the binary you got me, right?
Now, how can I be really sure that the binaries you passed me are truly build from the source you get me, unless I'm able to follow your procedures, that is, gain access to your secret key, sign the code with it by myself and contrast the results? Maybe a tribunal would have enough by asserting that "operatively" the program signed with your key is equivalent to the program signed with my key... or maybe not.
"Since the FSF is the only entity capable of modifying or releasing future versions of the GPL (as stated in the GPL), we have to trust they will not tamper with the spirit of the license."
No, we don't need to trust this. You can't take a paragraph from a contract and say "that's all with it". You need to see it in context.
Now: what does point 9 say?
"9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns."
You see, if the FSF publishes a new version of dissimilar spirit (of course, that should have to be decieded under a tribunal on the worst case) that would be license violation, thus void and unforceable. So if you say "this is distributed under the GPLv2 or later" that's not what it seems, a blank note, since any later GPL version can't be "just anything" but only something similar in spirit (which is quite clearly stated in the preamble) or it won't be a valid relicenseable candidate; at the very least this reasonement would give you a solid position in court, wich is neither more neither less than if you had to fight against any inane corporation that just took your code away.