Libraries have many roles, but almost none of those roles are exclusive. The three most important ones are collecting, sharing, and archiving for posterity, but these roles are also performed within society as a whole
The british library is a national library, and also a legal deposit library (one of only six) under the 2003 Act, and its job is to perform those roles. Others may perform the roles, but it isn't their stated purpose enshrined in law to do so.
This does make them exceptional. There is a huge difference between saying "DRM is stopping me doing Y" and saying "DRM is stopping me doing Y, which I am required to do by law".
The US is simply saying if you want to come here you must provide these bits of information. And by indications the US isnt saying stay home, though I think they should, they are saying if you dont provide the information you will have a delay while you talk to us about it.
Actually, the problem is they aren't asking the passenger (who is free to provide their info to a non-EU entity not covered by EU law), they are asking the airlines.
The airlines got the data from the passenger in the EU and have to abide by EU laws on that data. The could give it to the US (data export is possible under the directive) provided whoever they give it to agrees to keep the data protected under the EU laws. The US won't agree to that - they want the data with no restrictions, which EU law doesn't allow the airline to do.
My understanding (IANAL etc.) is that if the US asked the passengers directly for the data, there would not be a problem and the passengers could not expect that data to be subject to EU DP laws.
I would expect no less traveling to Europe. I have been detained by Customs in Europe and I followed the procedures without complaint because I was in someone elses country and they have a right to run it the way they want
Except, they (the US) aren't doing this in the US, they are requiring data to be sent in the EU, before the plane leaves the EU, otherwise they will refuse the plane landing or fine the airline.
A better comparison would be the EU customs detaining you in the US before you left, under EU procedures, but on US soil.
There is an established way to do (roughly - without the detention) this - simply require a visa for entry, interview required at US embassy (effectively US soil, US procedures). It's even already established that airlines can be fined for bringing a passenger without the right visa.
Firefox replaced it (from part of the same codebase - and note the JWZ doesn't say the code was bad) arguably precisely because they did exactly what JWZ says mozilla should have done and didn't - ie. shipped product early and often.
I think the real question is, as an open-source developer, why wouldn't you choose GPLv3 over v2? Because you want some company to use your program and then sue you because you made use of their patents?
GPLv3 will not stop that, all it does is revoke their licence if they do. Of course, you'll have to sue them to enforce that. If you've any cash left after fighting the patent case (maybe they'll take some IP in settlement, now lets see, what IP have you got?).
Or you want your software to make DRM devices cheaper to create?
DRM is a technology, so I don't care. I might care about how people _use_ that technology, but that is outside the scope of the licence. Technologies aren't good or evil - people use them for good or evil, and trying to ban a technology because some use it for evil is silly.
s/DRM devices/nuclear bombs/
Do we want to make nukes easier to create ? Hell no - quick let's add a "no using this software to make nukes" clause.
Then, in 30yrs time, when Bruce Willis fails to nuke the asteroid because of inferior proprietary software, we can all congratulate ourselves on how much "evil technology" we made harder to create.
You understand the difference between not being able to do something because it's physically impossible and being intentionally disallowed from doing it, right?
Yes, I think, but I don't know which applies in the flying case, since it is physically possible for me to fly, given wings / jetpack, but was I intentionally designed not to have them or was it evolutionary accident.
Maybe it would be clearer if we just stuck to devices with GPL software in.
If the device was intentionally designed with soldered eprom instead of socketed (or some more modern equivalents), is that "physically impossible" or "intentionally disallowed" ? Does it matter if the intent was to save costs or to prevent modification ? Does the GPLv3 mandate sockets ?
If the device was socketed but in a locked case (maybe with anti-tamper self-destruct), so the mfr. can upgrade it using the correct key, but I can't, is that "physically impossible" or "intentionally disallowed" ? Does the GPLv3 require I be given the key / opening tools / etc. ?
Does it matter what type of lock / key it is ? If so, why, since the effect is the same ?
What is stopping you compiling it and running it ?
Your hardware doesn't run (by design) programs that you compile ?
Well, get some that does then. Why did you get non-user-modifiable hardware if you wanted to modify it ?
I can't modify the software on (for one example) my phone, do I care ? no. Because if I _wanted_ to modify it, I'd have bought a phone which supported me modifying it.
You can take every modification Tivo has made to GPL software and use it in your own PVR (or in something completely different) - even compete with Tivo if you wish, and even if you eat their lunch in the marketplace, they _still_ have to give you every improvement they make to the GPL software. You can benefit from _all_ their GPL software R&D, for free.
Provided, of course, that you reciprocate, so they benefit from all yours too (_that_ is the core of the GPL - nothing to do with hardware).
So, why is this code not useful ? It isn't useful to me, as I don't want to build a PVR, but there's loads of GPL code that I have no use for - that doesn't mean it isn't useful.
No. He thinks that the people who BOUGHT AND LEGALLY OWN THE HARDWARE deserve to be "root".
Well, why doesn't he _say_ that in the GPLv3 then ?
The word OWN does not appear in it except referring to "their own keys" and "ther own removal", nor does "owner" or "bought" "sell" or "purchase".
What it does say is this:
Some computers are designed to deny users access to install or run modified versions of the software inside them. This is fundamentally incompatible with the purpose of the GPL, which is to protect users' freedom to change the software. Therefore, the GPL ensures that the software it covers will not be restricted in this way.
It is very clearly about USERS and not OWNERS.
The rest of the licence backs this up - anyone you "convey" the work to has to have full source including all keys, and "convey" is defined in terms of third parties making or receiving copies of the work, nothing about ownership vs. rental.
So, either 1. RMS can't get what he thinks expressed properly in the GPL or 2. he doesn't think what you think he does.
Given his past writings on the subject of (non owner) users having root access (info su, or google "GNU su support wheel" or similar), (2) is most likely by far.
He could say that all **future** Linux code becomes GPL3 otherwise it does not get gitted, but that cannot be retospectively applied to existing code and would mean that the majority of the Linux code would remain GPL2 for a long time.
I don't think he can, because he couldn't mix the two together - as far as I can see, GPLv3 adds additional restrictions, so you can't link it with GPLv2.
Two situations. First, one can have the ability to see and modify source code, but not run the program. Case in
Been that way since the 80s. Nothing new. Nothing crumbled as a result.
Second, ability to run the program, but not see the source code.
Ditto.
Back almost two decades ago in college I used plenty of GNU software.
In some cases, we had access to the source, but on the machines on which that software ran, I had nowhere near enough disk quota to rebuild a modified version, let alone install and run it. In some cases the programs we had access to were modified from the GNU source, and the full modified source was not made available.
From my experience at that time, this sort of setup was very common in academia, which was typically where you found GNU software then.
The GPL didn't noticably crumble as a result, and in fact its use has expanded massively since then.
Why ? Because we still had the freedom to look at the source and learn from it, take the source (get the original unmodified source in case 2), modify it to our hearts content, and run it somewhere else. So we did.
I was able in '92 to take a whole set of development tools and applications off a big proprietary Unix box and build/port them onto a Linux PC, which I then used as primary PC for almost ten years. That is the freedom GPL gives, No way could I have done that with proprietary apps.
To follow your logic, any system on which GPL software is installed must grant all users full admin rights to allow them to modify it _in_ _place_ (and therefore you could never use GPL software burned into ROM).
RMS might think that giving everyone root everywhere is the right thing to do, but outside of MIT, in the real world, it is totally impractical. Lots of people definitely _won't_ use GPLv3 software if that is what it means.
bad assumption, you either are free to do that (under the laws where you are) or you aren't - but it's entirely unlikely that the availability of copying hardware has anything to do with it.
copying consumer hardware has been sold (and promoted) for years in the UK ("high speed dubbing" anyone?) but it is NOT and never has been legal to copy tapes / lps / cds without copyright holder's explicit permission.
similarly I can buy cars where the top gear isn't usable below 70mph, so can I assume I'm free to go faster than that (since "that is the whole reason for its existence") ? - no, the legal use of that top gear is restricted to private roads/tracks where I've got permission - NOT on normal roads.
similarly, at least here, the legal use of a CD recorder is restricted to copying CDs (eg. stuff you've recorded yourself), where you have explicit copyright permission - NOT normal purchased CDs.
It's time people realised that when you buy RIAA music on CD etc. you pay a lot for not much at all. [In other parts of the world we don't even have the fair use / format shift rights that the US (sort of, maybe) consumer has].
When you buy music on CD you are buying the following:
The limited right (no public performance etc.) to listen to the music on that CD by playing that CD in a device that will play it.
Thats it.
CD doesn't play in your player ? Too bad - use one that plays it.
But I broke / scratched/the CD ? Too bad, stuff breaks, buy another.
But the CD got nicked ? Too bad, stuff gets nicked, talk to your insurer, buy another.
But what if you don't sell it anymore ? Well, sorry we lost a sale but hey, lots of things are irreplacable. Jaguar probably can't replace your E-Type either, but they'll have some nice new models. We have some nice new music, wanna buy ?
But what if I want it in MP3 for my MP3 player ? Well why didn't you buy it in MP3 ?
No, I mean I want it in MP3 _as well_ as CD ? So buy both (when did you last see "buy the hardcover get the softcover free") ?
But you don't sell it in MP3 ? - We're very sorry we've missed a sale, see above....and so on
Maybe if we educated people as to what (little) to RIAA are actually selling, rather than pretend they're selling something much more useful, people would wake up and buy from independents who can/do offer far more and not from RIAA.
The "just default" option was in response to a question on what to do if you "hadn't got the money" to pay.
In that case "just default" _is_ probably your best option - you haven't got any money anyway, so what are they going to take from you anyway ?
If the default judgement is in the millions it just means you go bankrupt for more and it might even help - if you owe so much, then maybe no one's going to question you going bankrupt.
Defending yourself might be more satisfying if you can make the other guys waste more of their money on lawyers to get nothing at the end of it, but it might be more risky. I don't think the RIAA can put you in jail from a civil suit, wheras I'm pretty sure that in a lot of places the judge can, for contempt, if he/she things you are taking the p***.
I'll clarify too - it's the FSF/GNU maintainers I think behaved poorly over BSD/GPL (the original BSD which is not GPL compatible), not Debian.
Debian doesn't get it right all the time, but they don't tend to completely ignore the problem and they do take difficult decisions like pulling code from the distribution.
It is also reasonable for Debian to expect that core GNU packages actually developed and distributed by the FSF would be ok by the GPL.
The FSF on the other hand just seemed to ignore (and deny) the problem (that they were shipping GPL mixed with BSD(original) in violation of the GPL) for years. Worse still, as the copyright holder and maintainer of the gpl, they could have actually fixed it (something Debian usually can't do). Simply add a special exception to allow BSD(original) to the GPL on the code in question or on the GPL itself, but (usenet allegedly) the advertising clause was considered too "obnoxious" to be legitemized in this way. Not IMO a great example for the originators of the GPL to ahve set the rest of the world.
The proprietor takes your credit card and address, and so knows where you are a
Major flaw in argument right there.
He knows a) where your credit card is issues and b) the address to which the credit card is registered.
He doesn't have a clue where you actually _are_.
1. If a US citizen on holiday in the UK walks into a casino with their US credit card and US address and gambles, legally in the UK, no US law broken, right ?
2. If they place a bet over the phone (while in the UK), which is legal in the UK, no US law broken, right ?
3. If they place a bet over the internet (while in the UK), which is legal in the UK, no US law broken, right ?
4. If they place another bet over the internet, when they get back home, which is legal in the UK where the gambling business is, oops, US law broken, right ?
Then consider same 1-4 applying to US citizen who has a UK address and credit card, then apply to a UK citizen (with same) who acesses the gambling site while in the US.
Now explain how the site operater can distinguish between 3 & 4 in each case to stay within US law.
If you can't, then you effectively saying that this US law shuts down online (and telephone, which can be trivially routed anywhere with voip) gambling worldwide, just because you don't like it. Right ?
How about you use a half decent source and half decent math, then face reality.
Your little list only has about 50 incidents (if 911 is four hijackings), of which 14 alone are FALN ("police tied 13 other bombings to the group.").
So how you get to even 99% (on a list of 50 that would mean they have to be _all_ muslim / arab) is beyond me.
Take a half-decent source like tkb http://www.tkb.org/ and you might get somewhere (approx 30000 incidents, covering past 40 years).
Unfortunately they don't look to have an easy way of breaking out attacks against Americans (or by muslim / non).
So lets work it the other way round, take a few of the major known non-muslim groups who target Americans and look at the known events (American targets):
[ Note that I've left out "The Aliens of America" incidents, since he now claims to have been AQ (back when Bin Laden was a small child) - but general consensus seems to be that he was just a nutter on a vengeance trip. ]
If that 150 is 1%, then you need to find 15,000 events by muslims / arabs attacking Americans.
Good luck doing that, TKB (as above) only has 12k total, (worldwide, all groups, all targets), that are by a known group (yep, that means 18k, ie. more than half, are unknown group). So you might be able to get some from there, but you'll still need to find a whole bunch of others.
Or maybe you could just realise you are plain wrong.
Since the statement questioned was "99% of all terrorists acts" (my empahasis) you can hardly expect the response to exclude the rest of the world and the American-friendly/funded terrorists who obviously aren't going to bite the hand that feeds...
Oh, and if you are in the US, while you're working on your "we never did that / don't do that anymore" response, would you mind attending to the little matter of the "anti-terrorist" extradition treaty which you won't ratify because apparently the UK might actually want to use it to extradite some terrorists (as opposed to bankers, which is what you use it for). Obviously, if you don't support / protect / harbour terrorists, there won't be anyone you need to protect from the "anti-terrorist" extradition treaty. Please think about it and get back to the rest of the world when you've decided which it is.
Oh, and also, if we were looking at American targets we'd be looking at
Oaklahoma - not muslim, therefore must be 1% - 168.
That means that American victims of Mulsim terror must be _at least_ 99 * 168 = 16638.
Ok, so 3-4k on 9/11, a few hundred in embasy bombings and Beirut barracks, a couple of thousand soldiers fighting a war of occupation, and er... must be one hell of a lot of incidents I've missed. Perhaps you could fil me in on the other 9-10k.
Or maybe the 99% is just miles out, even if you only count American victims.
Ok, so maybe we should give the guy a break since he's 70-odd, but really if you got that far mightn't you just want to do that one last big thing whether it killed you or not?
Re:Wonder what grade Tanenbaum would give them ?
on
FreeDOS 1.0 Released
·
· Score: 1
Tanenbaum wasn't saying a microkernel was the "right" way. He was saying it was the modern way.
If he didn't say microkernel was "right", he definitely said Linux / monolithic was "wrong". His acutal words were:
"a truly poor idea" "a gross error" "Not the way to go" "a fundamental error"
Wright Flyer [...] Concorde. Which design do you think an avionics professor would consider more modern?
Linux was nowhere near as primitive in computing terms.
OS/360 -> Wright Flyer Linux -> 747 (throwback design - SSTs were already in dev and thought to be the future) Concorde -> Hurd / microkernels
Yep, you're right, your 1960/70s avionics professor would say that Concorder was more modern, and would be the future, and the 747 was obsolete.
History has proved them wrong, along with Tanenbaum. Like the microkernels, Concorde was beautifully engineered, technically advanced, a masterpiece. It was also massively late, over budget and a complete failure at actually doing the job on a large/commercial scale. Meanwhile the "obsolete" design (747, Linux) is the most successful aircraft ever and is still leading the game after the "modern" design (Concorde, microkernels) has failed.
he failure of Hurd versus the success of Linux has little or nothing to do with the kernel architecture and everything to do with the respective project leaderships.
Agreed entirely on the project leaderships, but if you look wider, the microkernels have _all_ failed (restricted to niches at best) and the major successful OSes are all monolithic. Is that really down to project leadership in every case ?
I'll help you with some bits of it: Apache is GPL incompatible because of patent clause. As is CDDL. Artistic licence is not even "free software" let alone GPL compatible.
As for BSD... where do we even start....
Well, let's try. Once upon a time, some years ago, some people thought it was compatible - despite an obvious "additional restriction". They thought the restriction should be ignored because it wasn't enforceable or something (although the GPL doesn't say you can ignore restrictions for that reason). They mixed GPL and BSD code merrily and distributed it everywhere.
Then they got some legal advice - no, BSD (orginal) and GPL were (are) NOT compatible.
Oops, wonder what they did then ? Did they remove the offending code, did they get kicked from Debian, did they stop distributing ?
Er, no. They:
Denied it. Ignored it. Disagreed on it. Denied it again (usenet archives remember even better than I do). Hoped it would go away....and carried on shipping the illegal software.
Some years later, they got the BSD licence changed to be GPL compatible (remove advertising clause), and went round changing all the licence notices on other people's code to the new version. Presumably they got explicit permission to do that...
They still distribute the old illegal packages with the original incompatible licences, even today.
So, can we kick these licence cowboys off Debian for this outrageous disregard of the GPL ?
Probably not - after all, they are/were the GNU maintainers...
No, you said that they resolved issues, in reply to the comment (which you quoted):
t's funny because when the Apache Software Foundation has a license that is incompatible with the GPL, no one gave them grief, but SUN moves to one and suddenly they're evil..
where the only issue is _having_ a GPL incompatible licence, so I assumed that's what you claim they resolved.
So, to recap, the issue was (paraphrased):
we have Apache and SUN both creating and using deliberately GPL-incompatible licences for their code.
SUN gets called evil for this, Apache doesn't.
your response as we now understand it is (paraphrased):
Apache were obviously much more cooperative about fixing their licence-mixing screw-up (SUN, on the other hand, didn't screw up and mix licences in the first place)
which is either just a non-sequitur or just plain wierd (repented sinners are better than non-sinners ?)
Wonder what grade Tanenbaum would give them ?
on
FreeDOS 1.0 Released
·
· Score: 3, Interesting
After this:
I still maintain the point that designing a monolithic kernel in 1991 is a fundamental error. Be thankful you are not my student. You would not get a high grade for such a design:-)
what grade would you get for rewriting DOS 15 yrs later, and would it be higher or lower than the Hurd guys get for taking 20+yrs to get to 0.2 (but doing it the "right" way, with a microkernel) ?
"5 years from now everyone will be running free GNU" - Andy Tanenbaum, 1992
no offence intended, you may be a lawyer etc., but I trust the FSF website on this a lot more than someone posting on/. after all, part of the problem here is that Jörg Schilling has been going with his own thoughts on which licences are GPL (in)compatible instead of listening to the relevant experts.
so, until someone credible says otherwise, the GP is right, the Apache Software Foundation does have a license that is incompatible with the GPL. furthermore, since it's been so, and been known to be so, for a number of versions, it is unlikely that this incompatibility is accidental.
on that basis they deserve at least as much grief about it as Sun.
With a closed-source app, you tyically can't even *look* at the code.
Surely with a "closed-source" app you can't look at the code by definition? On the other hand there is lots of proprietary (non-free) code around to look at. Proprietary Source code gets licensed all the time, including for modification and redistribution.
Look at (for one example) the open-sourcing projects at Sun for Solaris and now Java - major headaches (plenty of press on the fact) with "third-party code" that they didn't have / couldn't-get the rights to re-licence as open source. So, it wasn't Sun's code, and it wasn't free/open-source, but nevertheless Sun were reusing it and redistributing it. Legally.
It can happen, it does happen, and in fact IME it is typical within the software industry.
With a GPL app, you can *always* look at the code, and you can *always* reuse the code.
Wrong. You can *only* reuse the code in with other GPL code. If the other code isn't GPL, then it you have to be able to put it under GPL. If you can't do that, then you *can't* reuse the GPL code - period. Furthermore, if you are going to do a clean re-implementation... then you can't even *look* at the GPL code.
Note that it might not be proprietary code you want to mix it with - GPL is incompatible with plenty of free / open-source licences. If you work on Apache (or Mozilla, or LaTeX, or *BSD, or...) you can *not* reuse GPL code - even though you (and the licences of those projects) are perfectly willing to "allow your users the same freedom".
Bottom line: re-using third-party code is a legal PITA (often to the extent that it's easier to re-write than re-use) because of licencing issues, and that is true *regardless* of whether your code or the other code is open-source or proprietary. GPL code is not better (in general) than other code in this regard - in fact it is arguably worse as it is incompatible with so many other licences.
Slashdot Mirror
This process allows you to copy quantum information from one set of atoms to another without measuring it, and thereby destroying it.
If you can't measure it in the first place (and the original gets destroyed in the process), how do you know that what you end up with is a copy ?
Libraries have many roles, but almost none of those roles are exclusive. The three most important ones are collecting, sharing, and archiving for posterity, but these roles are also performed within society as a whole
The british library is a national library, and also a legal deposit library (one of only six) under the 2003 Act, and its job is to perform those roles. Others may perform the roles, but it isn't their stated purpose enshrined in law to do so.
This does make them exceptional. There is a huge difference between saying "DRM is stopping me doing Y" and saying "DRM is stopping me doing Y, which I am required to do by law".
The US is simply saying if you want to come here you must provide these bits of information. And by indications the US isnt saying stay home, though I think they should, they are saying if you dont provide the information you will have a delay while you talk to us about it.
Actually, the problem is they aren't asking the passenger (who is free to provide their info to a non-EU entity not covered by EU law), they are asking the airlines.
The airlines got the data from the passenger in the EU and have to abide by EU laws on that data. The could give it to the US (data export is possible under the directive) provided whoever they give it to agrees to keep the data protected under the EU laws. The US won't agree to that - they want the data with no restrictions, which EU law doesn't allow the airline to do.
My understanding (IANAL etc.) is that if the US asked the passengers directly for the data, there would not be a problem and the passengers could not expect that data to be subject to EU DP laws.
I would expect no less traveling to Europe. I have been detained by Customs in Europe and I followed the procedures without complaint because I was in someone elses country and they have a right to run it the way they want
Except, they (the US) aren't doing this in the US, they are requiring data to be sent in the EU, before the plane leaves the EU, otherwise they will refuse the plane landing or fine the airline.
A better comparison would be the EU customs detaining you in the US before you left, under EU procedures, but on US soil.
There is an established way to do (roughly - without the detention) this - simply require a visa for entry, interview required at US embassy (effectively US soil, US procedures). It's even already established that airlines can be fined for bringing a passenger without the right visa.
So, they should have just retired the product, right....?
0 6/04/12/sunset-announcement-for-fxtb-10x-and-mozil la-suite-17x/
They did. Mozilla the product is dead.
http://developer.mozilla.org/devnews/index.php/20
Firefox replaced it (from part of the same codebase - and note the JWZ doesn't say the code was bad) arguably precisely because they did exactly what JWZ says mozilla should have done and didn't - ie. shipped product early and often.
GPLv3 will not stop that, all it does is revoke their licence if they do. Of course, you'll have to sue them to enforce that. If you've any cash left after fighting the patent case (maybe they'll take some IP in settlement, now lets see, what IP have you got?).
DRM is a technology, so I don't care. I might care about how people _use_ that technology, but that is outside the scope of the licence. Technologies aren't good or evil - people use them for good or evil, and trying to ban a technology because some use it for evil is silly.
s/DRM devices/nuclear bombs/
Do we want to make nukes easier to create ? Hell no - quick let's add a "no using this software to make nukes" clause.
Then, in 30yrs time, when Bruce Willis fails to nuke the asteroid because of inferior proprietary software, we can all congratulate ourselves on how much "evil technology" we made harder to create.
You understand the difference between not being able to do something because it's physically impossible and being intentionally disallowed from doing it, right?
Yes, I think, but I don't know which applies in the flying case, since it is physically possible for me to fly, given wings / jetpack, but was I intentionally designed not to have them or was it evolutionary accident.
Maybe it would be clearer if we just stuck to devices with GPL software in.
If the device was intentionally designed with soldered eprom instead of socketed (or some more modern equivalents), is that "physically impossible" or "intentionally disallowed" ? Does it matter if the intent was to save costs or to prevent modification ? Does the GPLv3 mandate sockets ?
If the device was socketed but in a locked case (maybe with anti-tamper self-destruct), so the mfr. can upgrade it using the correct key, but I can't, is that "physically impossible" or "intentionally disallowed" ? Does the GPLv3 require I be given the key / opening tools / etc. ?
Does it matter what type of lock / key it is ? If so, why, since the effect is the same ?
What is stopping you compiling it and running it ?
Your hardware doesn't run (by design) programs that you compile ?
Well, get some that does then. Why did you get non-user-modifiable hardware if you wanted to modify it ?
I can't modify the software on (for one example) my phone, do I care ? no. Because if I _wanted_ to modify it, I'd have bought a phone which supported me modifying it.
You can take every modification Tivo has made to GPL software and use it in your own PVR (or in something completely different) - even compete with Tivo if you wish, and even if you eat their lunch in the marketplace, they _still_ have to give you every improvement they make to the GPL software. You can benefit from _all_ their GPL software R&D, for free.
Provided, of course, that you reciprocate, so they benefit from all yours too (_that_ is the core of the GPL - nothing to do with hardware).
So, why is this code not useful ? It isn't useful to me, as I don't want to build a PVR, but there's loads of GPL code that I have no use for - that doesn't mean it isn't useful.
Well, why doesn't he _say_ that in the GPLv3 then ?
The word OWN does not appear in it except referring to "their own keys" and "ther own removal", nor does "owner" or "bought" "sell" or "purchase".
What it does say is this:
It is very clearly about USERS and not OWNERS.
The rest of the licence backs this up - anyone you "convey" the work to has to have full source including all keys, and "convey" is defined in terms of third parties making or receiving copies of the work, nothing about ownership vs. rental.
So, either
1. RMS can't get what he thinks expressed properly in the GPL
or
2. he doesn't think what you think he does.
Given his past writings on the subject of (non owner) users having root access (info su, or google "GNU su support wheel" or similar), (2) is most likely by far.
He could say that all **future** Linux code becomes GPL3 otherwise it does not get gitted, but that cannot be retospectively applied to existing code and would mean that the majority of the Linux code would remain GPL2 for a long time.
I don't think he can, because he couldn't mix the two together - as far as I can see, GPLv3 adds additional restrictions, so you can't link it with GPLv2.
Two situations. First, one can have the ability to see and modify source code, but not run the program. Case in
Been that way since the 80s. Nothing new. Nothing crumbled as a result.
Second, ability to run the program, but not see the source code.
Ditto.
Back almost two decades ago in college I used plenty of GNU software.
In some cases, we had access to the source, but on the machines on which that software ran, I had nowhere near enough disk quota to rebuild a modified version, let alone install and run it. In some cases the programs we had access to were modified from the GNU source, and the full modified source was not made available.
From my experience at that time, this sort of setup was very common in academia, which was typically where you found GNU software then.
The GPL didn't noticably crumble as a result, and in fact its use has expanded massively since then.
Why ? Because we still had the freedom to look at the source and learn from it, take the source (get the original unmodified source in case 2), modify it to our hearts content, and run it somewhere else. So we did.
I was able in '92 to take a whole set of development tools and applications off a big proprietary Unix box and build/port them onto a Linux PC, which I then used as primary PC for almost ten years. That is the freedom GPL gives, No way could I have done that with proprietary apps.
To follow your logic, any system on which GPL software is installed must grant all users full admin rights to allow them to modify it _in_ _place_ (and therefore you could never use GPL software burned into ROM).
RMS might think that giving everyone root everywhere is the right thing to do, but outside of MIT, in the real world, it is totally impractical. Lots of people definitely _won't_ use GPLv3 software if that is what it means.
so I assume I'm free to duplicate CDs
m oney/2006/05/07/cnbpi07.xml for example.
bad assumption, you either are free to do that (under the laws where you are) or you aren't - but it's entirely unlikely that the availability of copying hardware has anything to do with it.
copying consumer hardware has been sold (and promoted) for years in the UK ("high speed dubbing" anyone?) but it is NOT and never has been legal to copy tapes / lps / cds without copyright holder's explicit permission.
http://www.telegraph.co.uk/money/main.jhtml?xml=/
similarly I can buy cars where the top gear isn't usable below 70mph, so can I assume I'm free to go faster than that (since "that is the whole reason for its existence") ? - no, the legal use of that top gear is restricted to private roads/tracks where I've got permission - NOT on normal roads.
similarly, at least here, the legal use of a CD recorder is restricted to copying CDs (eg. stuff you've recorded yourself), where you have explicit copyright permission - NOT normal purchased CDs.
It's time people realised that when you buy RIAA music on CD etc. you pay a lot for not much at all. [In other parts of the world we don't even have the fair use / format shift rights that the US (sort of, maybe) consumer has].
/the CD ? Too bad, stuff breaks, buy another.
...and so on
When you buy music on CD you are buying the following:
The limited right (no public performance etc.) to listen to the music on that CD by playing that CD in a device that will play it.
Thats it.
CD doesn't play in your player ? Too bad - use one that plays it.
But I broke / scratched
But the CD got nicked ? Too bad, stuff gets nicked, talk to your insurer, buy another.
But what if you don't sell it anymore ? Well, sorry we lost a sale but hey, lots of things are irreplacable. Jaguar probably can't replace your E-Type either, but they'll have some nice new models. We have some nice new music, wanna buy ?
But what if I want it in MP3 for my MP3 player ? Well why didn't you buy it in MP3 ?
No, I mean I want it in MP3 _as well_ as CD ? So buy both (when did you last see "buy the hardcover get the softcover free") ?
But you don't sell it in MP3 ? - We're very sorry we've missed a sale, see above.
Maybe if we educated people as to what (little) to RIAA are actually selling, rather than pretend they're selling something much more useful, people would wake up and buy from independents who can/do offer far more and not from RIAA.
The "just default" option was in response to a question on what to do if you "hadn't got the money" to pay.
In that case "just default" _is_ probably your best option - you haven't got any money anyway, so what are they going to take from you anyway ?
If the default judgement is in the millions it just means you go bankrupt for more and it might even help - if you owe so much, then maybe no one's going to question you going bankrupt.
Defending yourself might be more satisfying if you can make the other guys waste more of their money on lawyers to get nothing at the end of it, but it might be more risky. I don't think the RIAA can put you in jail from a civil suit, wheras I'm pretty sure that in a lot of places the judge can, for contempt, if he/she things you are taking the p***.
I'll clarify too - it's the FSF/GNU maintainers I think behaved poorly over BSD/GPL (the original BSD which is not GPL compatible), not Debian.
Debian doesn't get it right all the time, but they don't tend to completely ignore the problem and they do take difficult decisions like pulling code from the distribution.
It is also reasonable for Debian to expect that core GNU packages actually developed and distributed by the FSF would be ok by the GPL.
The FSF on the other hand just seemed to ignore (and deny) the problem (that they were shipping GPL mixed with BSD(original) in violation of the GPL) for years. Worse still, as the copyright holder and maintainer of the gpl, they could have actually fixed it (something Debian usually can't do). Simply add a special exception to allow BSD(original) to the GPL on the code in question or on the GPL itself, but (usenet allegedly) the advertising clause was considered too "obnoxious" to be legitemized in this way. Not IMO a great example for the originators of the GPL to ahve set the rest of the world.
The proprietor takes your credit card and address, and so knows where you are a
Major flaw in argument right there.
He knows a) where your credit card is issues and b) the address to which the credit card is registered.
He doesn't have a clue where you actually _are_.
1. If a US citizen on holiday in the UK walks into a casino with their US credit card and US address and gambles, legally in the UK, no US law broken, right ?
2. If they place a bet over the phone (while in the UK), which is legal in the UK, no US law broken, right ?
3. If they place a bet over the internet (while in the UK), which is legal in the UK, no US law broken, right ?
4. If they place another bet over the internet, when they get back home, which is legal in the UK where the gambling business is, oops, US law broken, right ?
Then consider same 1-4 applying to US citizen who has a UK address and credit card, then apply to a UK citizen (with same) who acesses the gambling site while in the US.
Now explain how the site operater can distinguish between 3 & 4 in each case to stay within US law.
If you can't, then you effectively saying that this US law shuts down online (and telephone, which can be trivially routed anywhere with voip) gambling worldwide, just because you don't like it. Right ?
Do the math, then face reality.
I've done the math.
How about you use a half decent source and half decent math, then face reality.
Your little list only has about 50 incidents (if 911 is four hijackings), of which 14 alone are FALN ("police tied 13 other bombings to the group.").
So how you get to even 99% (on a list of 50 that would mean they have to be _all_ muslim / arab) is beyond me.
Take a half-decent source like tkb http://www.tkb.org/ and you might get somewhere (approx 30000 incidents, covering past 40 years).
Unfortunately they don't look to have an easy way of breaking out attacks against Americans (or by muslim / non).
So lets work it the other way round, take a few of the major known non-muslim groups who target Americans and look at the known events (American targets):
FALN: (120+ bombings)
FARC: 10+
November 17: 10+
Unabomber: 16
ELN: 5+
SO, we are looking at 150+, say 150.
[ Note that I've left out "The Aliens of America" incidents, since he now claims to have been AQ (back when Bin Laden was a small child) - but general consensus seems to be that he was just a nutter on a vengeance trip. ]
If that 150 is 1%, then you need to find 15,000 events by muslims / arabs attacking Americans.
Good luck doing that, TKB (as above) only has 12k total, (worldwide, all groups, all targets), that are by a known group (yep, that means 18k, ie. more than half, are unknown group). So you might be able to get some from there, but you'll still need to find a whole bunch of others.
Or maybe you could just realise you are plain wrong.
LD50 for Ricin on the other hand is less than that for a whole person. A bit of that on a needle is a big problem.
Since the statement questioned was "99% of all terrorists acts" (my empahasis) you can hardly expect the response to exclude the rest of the world and the American-friendly/funded terrorists who obviously aren't going to bite the hand that feeds...
Oh, and if you are in the US, while you're working on your "we never did that / don't do that anymore" response, would you mind attending to the little matter of the "anti-terrorist" extradition treaty which you won't ratify because apparently the UK might actually want to use it to extradite some terrorists (as opposed to bankers, which is what you use it for). Obviously, if you don't support / protect / harbour terrorists, there won't be anyone you need to protect from the "anti-terrorist" extradition treaty. Please think about it and get back to the rest of the world when you've decided which it is.
Oh, and also, if we were looking at American targets we'd be looking at
Oaklahoma - not muslim, therefore must be 1% - 168.
That means that American victims of Mulsim terror must be _at least_ 99 * 168 = 16638.
Ok, so 3-4k on 9/11, a few hundred in embasy bombings and Beirut barracks, a couple of thousand soldiers fighting a war of occupation, and er... must be one hell of a lot of incidents I've missed. Perhaps you could fil me in on the other 9-10k.
Or maybe the 99% is just miles out, even if you only count American victims.
In other news this morning, apparently Branson has offered Wm. Shatner a free trip... but he's too scared to go (unless they pay him...)
0 .html
eg. http://www.thesun.co.uk/article/0,,2-2006410413,0
Ok, so maybe we should give the guy a break since he's 70-odd, but really if you got that far mightn't you just want to do that one last big thing whether it killed you or not?
Tanenbaum wasn't saying a microkernel was the "right" way. He was saying it was the modern way.
If he didn't say microkernel was "right", he definitely said Linux / monolithic was "wrong". His acutal words were:
"a truly poor idea"
"a gross error"
"Not the way to go"
"a fundamental error"
Wright Flyer [...] Concorde. Which design do you think an avionics professor would consider more modern?
Linux was nowhere near as primitive in computing terms.
OS/360 -> Wright Flyer
Linux -> 747 (throwback design - SSTs were already in dev and thought to be the future)
Concorde -> Hurd / microkernels
Yep, you're right, your 1960/70s avionics professor would say that Concorder was more modern, and would be the future, and the 747 was obsolete.
History has proved them wrong, along with Tanenbaum. Like the microkernels, Concorde was beautifully engineered, technically advanced, a masterpiece. It was also massively late, over budget and a complete failure at actually doing the job on a large/commercial scale. Meanwhile the "obsolete" design (747, Linux) is the most successful aircraft ever and is still leading the game after the "modern" design (Concorde, microkernels) has failed.
he failure of Hurd versus the success of Linux has little or nothing to do with the kernel architecture and everything to do with the respective project leaderships.
Agreed entirely on the project leaderships, but if you look wider, the microkernels have _all_ failed (restricted to niches at best) and the major successful OSes are all monolithic. Is that really down to project leadership in every case ?
It might be well understood _now_ but it took a few years (see below) - and _you_ clearly still haven't got it. Read http://www.gnu.org/licenses/license-list.html#GPL
I'll help you with some bits of it: Apache is GPL incompatible because of patent clause. As is CDDL. Artistic licence is not even "free software" let alone GPL compatible.
As for BSD... where do we even start....
Well, let's try. Once upon a time, some years ago, some people thought it was compatible - despite an obvious "additional restriction".
They thought the restriction should be ignored because it wasn't enforceable or something (although the GPL doesn't say you can ignore restrictions for that reason).
They mixed GPL and BSD code merrily and distributed it everywhere.
Then they got some legal advice - no, BSD (orginal) and GPL were (are) NOT compatible.
Oops, wonder what they did then ? Did they remove the offending code, did they get kicked from Debian, did they stop distributing ?
Er, no. They:
Denied it. Ignored it. Disagreed on it. Denied it again (usenet archives remember even better than I do). Hoped it would go away.
Some years later, they got the BSD licence changed to be GPL compatible (remove advertising clause), and went round changing all the licence notices on other people's code to the new version. Presumably they got explicit permission to do that...
They still distribute the old illegal packages with the original incompatible licences, even today.
So, can we kick these licence cowboys off Debian for this outrageous disregard of the GPL ?
Probably not - after all, they are/were the GNU maintainers...
No, you said that they resolved issues, in reply to the comment (which you quoted):
t's funny because when the Apache Software Foundation has a license that is incompatible with the GPL, no one gave them grief, but SUN moves to one and suddenly they're evil..
where the only issue is _having_ a GPL incompatible licence, so I assumed that's what you claim they resolved.
So, to recap, the issue was (paraphrased):
we have Apache and SUN both creating and using deliberately GPL-incompatible licences for their code.
SUN gets called evil for this, Apache doesn't.
your response as we now understand it is (paraphrased):
Apache were obviously much more cooperative about fixing their licence-mixing screw-up (SUN, on the other hand, didn't screw up and mix licences in the first place)
which is either just a non-sequitur or just plain wierd (repented sinners are better than non-sinners ?)
what grade would you get for rewriting DOS 15 yrs later, and would it be higher or lower than the Hurd guys get for taking 20+yrs to get to 0.2 (but doing it the "right" way, with a microkernel) ?
"5 years from now everyone will be running free GNU" - Andy Tanenbaum, 1992
really ? someone needs to tell the FSF then, because they still list all the apache licenses as incompatible http://www.fsf.org/licensing/licenses/index_html#
no offence intended, you may be a lawyer etc., but I trust the FSF website on this a lot more than someone posting on
so, until someone credible says otherwise, the GP is right, the Apache Software Foundation does have a license that is incompatible with the GPL. furthermore, since it's been so, and been known to be so, for a number of versions, it is unlikely that this incompatibility is accidental.
on that basis they deserve at least as much grief about it as Sun.
With a closed-source app, you tyically can't even *look* at the code.
Surely with a "closed-source" app you can't look at the code by definition? On the other hand there is lots of proprietary (non-free) code around to look at. Proprietary Source code gets licensed all the time, including for modification and redistribution.
Look at (for one example) the open-sourcing projects at Sun for Solaris and now Java - major headaches (plenty of press on the fact) with "third-party code" that they didn't have / couldn't-get the rights to re-licence as open source. So, it wasn't Sun's code, and it wasn't free/open-source, but nevertheless Sun were reusing it and redistributing it. Legally.
It can happen, it does happen, and in fact IME it is typical within the software industry.
With a GPL app, you can *always* look at the code, and you can *always* reuse the code.
Wrong. You can *only* reuse the code in with other GPL code. If the other code isn't GPL, then it you have to be able to put it under GPL. If you can't do that, then you *can't* reuse the GPL code - period. Furthermore, if you are going to do a clean re-implementation... then you can't even *look* at the GPL code.
Note that it might not be proprietary code you want to mix it with - GPL is incompatible with plenty of free / open-source licences. If you work on Apache (or Mozilla, or LaTeX, or *BSD, or...) you can *not* reuse GPL code - even though you (and the licences of those projects) are perfectly willing to "allow your users the same freedom".
Bottom line: re-using third-party code is a legal PITA (often to the extent that it's easier to re-write than re-use) because of licencing issues, and that is true *regardless* of whether your code or the other code is open-source or proprietary. GPL code is not better (in general) than other code in this regard - in fact it is arguably worse as it is incompatible with so many other licences.