Remember, the "equivalent access " clause is referring to "offering access to copy from a designated place", such as downloading (though not necessarily so). I think you're putting too much weight on equivalent access. My making the copies and offering them to you does not have anything to do with "equivalent access".
Hmm, I'm not sure I follow you. And I dont feel i'm putting too much weight on equivalent access. I get the impression you're interpreting the GPL as:
"copy from designated place" == "equivalent access" and hence fulfilling former -> "3a is fulfilled".
However that is not the way the GPL words it, the GPL is worded as:
"if copy from designated place" AND "equivalent access" then "3a is fullfilled".
So "equivalent access" is a critical determining test as to whether you fulfill 3a when distributing via "copy from designated place". And the GPL does not define or further qualify "equivalent access", that's left for professional interpretation, possibly judicial interpretation as a result of litigation. The exact wording of the GPL is:
If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code
Ie: if (copy binary from designated place), then if (equivalent-access to source at same place) -> 3a considered fulfilled.
3a doesn't say anything about equivalent access, 3a doesn't say anything about source only, 3a doesn't say anything about written offers. If 3a can be interpreted to mean you CAN add an additional charge for the act of distributing the sources (plus the binary), as opposed to just distributing the binary, then it isn't really requiring the sources to be distributed.
You are absolutely correct, bar in one critical detail. Yes, you can charge whatever the hell you like for source, or for binary+source. The detail you are missing though or the fallacy if you wish, I believe, is that you seem to think you can distribute binaries and only *offer* source and still fulfill 3a somehow.
You can not. As you point out yourself, there is no talk of offers in 3a. However, it is impossible to avail of 3a if you distribute binary only - the *offer* of source is not enough. Specifically, in order to distribute binaries under 3a you must:
Accompany it with the complete corresponding machine-readable source code
I fail to see by what reading you think you can ship only binaries and yet fulfill the above. If you distribute binaries without source commercially then you *must* be distributing under 3b, and hence you *must* provide an offer of source at cost, valid for 3 years.
There is simply nothing else in the GPL that allows you to distribute binaries commercially.:)
Either you ship your binaries under 3a:
- source accompanying the binaries or equivalent-access to binaries and source from a designated place (eg a password protected FTP site, or a special one-time, customer-specific HTTP URL).
Or you ship under 3b:
- ship binaries, offer the customer the right to obtain source at cost.
Again, you can *not* avail of 3a if you give your customer a binary-only CD for $30 and *offer* source @ whatever price. Only 3b allows you to not ship source, but then you can only charge cost. (and yes, if the media for the source genuinely costs $900, you can charge that. And no, you can not pad it with 0s - at least you cant charge for it, it's not part of the cost of copying the source.).
If 3a can be interpreted to mean you CAN add an additional charge for the act of distributing the sources (plus the binary), as opposed to just distributing the binary, then it isn't really requiring the sources to be distributed.
How? 3a *unambigiously* requires you to ship the source (on physical media, or same place + equivalent access). There
Actually it is a product of the language spoken by Friesians and other tribes of Angles who settled in Britannia, mixed with latin. Then later the Vikings came along and settled in Angleland, mostly in the north and again influenced the language. Then a bunch of accountants, sorry Normans, (descended from the Vikings who invaded northern Gaul and settled with the local Gauls) invaded from France, and again influenced the language and culture.
After that, England was never invaded again, and the language never again underwent radical changes, just slow evolution, through immigration, exploration and colonisation. Which is why Chaucer's Canterbury tales, 700 hundred odd years old, are still readable, even to a modern english speaker.
I cant think of any differences in modern english that are a product of the USA, other than some trivial spelling and usage differences.
I don't keep track, but if there are any IBM-Sun negotiations or lawsuits going on, this comment probably has to do with that.
IBM are one of Suns' main competitors. Both sell PC servers, RISC servers, Linux, Unix®, Java, software support, a range of "middleware" software and IT consultancy services.
Sun, like all companies that actually have competition, will jostle for position with, make statements about and generally try to best and outsmart their competition. That's how competition works...
Since sun owns unix source (via the license with old ATT), they can always sell Linux. Now, if they can prevent others, well, then they have a case about solaris.
I really cant even imagine Sun trying to do something like that though.
I'm reading the comments on this story and I'm just amazed at how many comments are so hostile to Sun - I just dont understand where this hostility comes from. Sure I can understand people being critical of Sun, and criticicism is good, but this outright hatred is just weird.
Now, I'm a (recent) Sun employee[2], so maybe I'm blinded by my paycheck, but it seems to me that to consider a company that:
Paid for a lot of the HID and developer time that was put into GNOME, and continues to pay people to work on GNOME
Bought StarOffice, open-sourced it, continues to fund the development of OpenOffice
Have stated they are working out the details of open-sourcing its own Solaris Unix and will be doing so.
Is possibly the most active promoter of Linux on the corporate desktop by way of JDS. (remember, both IBM and RedHat execs last year made comments about Linux not being ready for desktop).
Is a long standing and ongoing developer of and contributor to Unix technologies and software (including Linux).
Walks the talk when it comes to pervasise deployment of Unix and Linux within corporate IT. The Sun corporate network is possibly one of the worlds largest cohesive Unix networks, and that includes Linux, not least by way of JDS.
as being a reasonable pariah for the Linux community is just strange.
So Sun still push Solaris over Linux, well why wouldnt Sun? Sun have spent a long time working on it, the people at Sun are proud of Solaris. Surely they have as much right to be proud of their (their, cause I havnt contributed to Solaris) work as the "Linux" developers[1] have to be of theirs? And even so, Sun still do spend money on technologies that are of benefit to Unix in general, be it Solaris, Linux, BSD, whatever.. and spend money marketing what is effectively Linux.
So Sun bought out licence rights from SCO, how evil of them, but if you're responsible for Sun and you have a chance to fully secure your "IP" (yuk) rights wouldn't it be corporate irresponsibility to not do so? Remember, you can be sued by shareholders for your inactions as much as your actions.
So Sun settled a long-running dispute with MS, how evil of them. But MS infringed on Suns' rights, is Sun not allowed to get a fat cheque from MS for MSs' wrongdoing, should Sun instead have continued litigating the matter at great expense and uncertainty? Would Sun maybe then later being awarded a fat cheque from MS by court order have then *not* been evil? The settlement recompenses Sun for wrong done to it and lets Sun get on with things, why is that evil?
At the end of the day, Sun are a Unix company. Sun are not perfect, no entity is, and Sun will have to adapt to changing market conditions, as all companies do, but they're the only big company who are and have been 100% committed to Unix from day one of their existence. Sure, Sun would prefer to sell you Solaris, and why not, Solaris is still Unix, and work on any one Unix ultimately benefits all unixes, be it directly or by virtue of competition. Never mind that Sun also directly contribute to technologies/projects that are key to Linux, as well as many other cross-platform projects, and also market Linux in one segment of the market.
The irony of course is that most of these/. weenies who like to spout this ill-informed "Sun is evil, they hate Linux!!!" clap-trap are likely doing so from the "comfort" of their Win32 PCs.
Linux, Solaris, BSD, etc.. So what, they're all Unix. Unix in part draws it's strength and health from diversity, from not being a heterogenous meritocracy, not a homogenous monopoly. Sun has long been a valuable contributor to that meritocracy of ideas.
Vive la difference!
1. What is a Linux developer exactly, aside from Linux kernel developers? I work on stuff at Sun that runs on Linux and Solaris. It's all Unix to me..
2. NB: I do not speak for Sun, opinions in this post are my own. Statemen
This is pretty cool, I think we're converging on agreement.
Indeed;)
I see now we simply failed to communicate, as what I was referring to was "here's the binary for $20, do you want the source CD for another $20" - that's "an additional charge for the source code" under 3a.
Hmmm.. possible, presuming it counts as equivalent access. I'm not sure it does though, because in order to be able to say "For an extra $20 you can download the source" implies that source and binary acess are *not* equivalent - otherwise how do you enforce that people must pay the $20 extra for the source?
If it does not fall under equivalent-access, then one is required to extend a 3b offer to anyone who downloads only the binary, hence rendering the "want source for $20 extra?" useless (unless of course the $20 is at cost).
Btw, I dont agree with the GPL FAQ answer you quoted, it seems simplistic at best and dangerously misleading at worst, if not downright wrong:). The GPL uses the term "same place" to clarify the meaning of equivalent-access, so any scheme that seperates the two in a significant way (eg $20 to get access to the binaries, $20 to get access to the source) does not sound like equivalent-access to me.
But it isnt wrong though, you *are* allowed to charge whatever you like for source. Just anyone who downloads *only* the binaries under your scheme is then entitled to a 3b offer, which presumably will be much less that $20.
In other words, if my "fee for downloading source" is the same as my fee for downloading binaries, which is $5/megabyte for a 4 megabyte binary, and I pack the source tar file with the source packages for gcc, glibc, make, bash, texinfo, binutils, textutils, fileutils, shellutils, binutils, and emacs,
You can do that, but you're not likely to stay in business for long. Also, you may be open to false-advertising suits if you advertise "100s of MBs of free software!" and the bulk of it is 0-padding.
I'm going to be doing exactly what you said: binary for $REASONABLE_AMOUNT, source for $TOTALLY_RIDICULOUS_AMOUNT_THAT_I_CAN_RETIRE_ON,
Yes, you can do this, there is no restriction on how much charge, except for 3b source. So there is no problem in charging $TOTALLY_RIDICULOUS_AMOUNT_THAT_I_CAN_RETIRE_ON for the source. The only thing is that any sensible consumer will *not* buy the source CD, they'll buy the binary CD and then come back to you and ask for the source at cost price under 3b. So the best thing to do is to *not* provide binaries at all, and hence avoid having to make a 3b offer altogether. If the source is valuable enough you will get your OTALLY_RIDICULOUS_AMOUNT_THAT_I_CAN_RETIRE_ON, market will decide whether it's worth it. If it isnt, you might need to lower your price.
(this presumes your scheme is not considered equivalent access, which i dont think it can be reasonably considered to be so.)
effectively making a binary-only distribution.
Nope, cause I just ignore your $HUGE_AMOUNT source CD, buy the binary CD for $REASONABLE_AMOUNT and then avail of the 3b offer to get the source at cost.
The purpose I was referring to is the GPL's, which is to make sure that anyone who receives a copy of the binary can get a copy of the source for reasonable terms. I see a conflict with your responses; on the one hand, your explanation of "No you cant, because then you're not distributing the source with the binaries obviously. " is pretty much exactly what I was trying to get at, but then you also agree that you can charge whatever you want for physical media.
Yes, you can charge as much as you want, either for source, for binaries, or for source+binaries. Distributing source does not imply distributing under 3b. You can distribute source and charge $HUGE_AMOUNT.
Only the 3b offer of source is restricted in cost, and you must make it if you distribute binaries without equivalent-access to source.
Well, "the information". Just guessing that they have it and asking for it doesn't seem to be sufficient, at least based on the two references in the GPL FAQ that I gave you.
Well, the GPL FAQ is not infallible, as we both seem agreed on and anyway irrelevant. Alternatively, look on it that the GPL FAQ restricts itself to giving answers that are almost certainly correct, or more likely to hold up in court, but that this need not mean that other alternative answers would not also hold true (whether alongside the GPL FAQ answer or not). Eg, having a binary copy of the work and a copy of the 3c offer almost certainly entitles one to avail of the original 3b offer, which is the answer the GPL FAQ gives, yet this need not preclude the possibility that literally any 3rd party could avail of the offer, be that party a holder of a 3c offer and/or the binaries or not.
Ultimately only the opinion of a judge will matter. And it seems to me that an offer open "to any 3rd party" leaves a lot of scope for a lawyer to argue that this does not require the offeree to have the full details of the offer, written or not or to have the binaries, in order to be able to avail of the offer. Remember, if one knows the offer must exist, one already will know the general form of the offer, as specified by 3b. Whether that is enough, only a judge could decide, and it might even depend on other context.
The counter-argument is that any 3rd party could only receive the offer because of 3c and hence should have the details. However, what of a 3rd party who is given the binaries but not the full details of the 3b offer? Are they excluded from availing of the original 3b offer? What if they have the offer, but not the binaries? If the offer only applies to those who receive it via a 3c offer, that would indeed require one to have the binaries.
But then that is obviously a tighter definition than "any 3rd party". If the GPL had intended the offer to be restricted in such ways, why doesnt it then have language to qualify "any 3rd party" thus?
Anyway, one for a judge to decide:)
Regarding 3a, what I meant was that if I sell you a binary release on CD for $20, can I also charge you an additional $20 for the source CD (which you can decline if you wish)?
No you cant, because then you're not distributing the source with the binaries obviously. It's quite simple as I see it:
- distribute the work in full (ie binaries and source) and you can charge whatever you like (3a) and you have no further obligations (ie no need to make a 3b offer).
- distribute only binary versions of the work, then you can charge whatever you like, but you must additionally make a 3b offer and provide source at reasonable cost for performing the copy, no more.
- distribute source, and you can charge whatever you like.
Now, for this last case you'd probably want to make sure to distribute *only* source, or at least, do not ever distribute binary only because then you'd have to make a 3b offer and people wouldnt need to buy the source, they could just get at cost via the 3b offer, presuming my interpretation of "any 3rd party" is correct, and your interpretation of "must have the binaries" or "must have copy of offer" is not.
The FAQ says you can charge a fee for downloading, why wouldn't that apply to charging a fee for physical media?
You can yes. You can charge whatever you want for a GPL work, be it source only, binary only or binary+source. The only place where are you restricted is in distributing source as part of a 3b offer, which you must make if you distribute binaries and charge more than cost for it (3c being noncommercial, for whatever definition of noncommercial).
If I can charge you a different amount for "binary + source" than "binary (but I offered you source and you didn't want it)", and both fall under 3a (thus no written offer, no obligation to provide source in the future), then that seems to defeat the purpo
Regarding who can execute the "written offer", from the GPL FAQ (also see this later question) seems to clearly indicate that you must have a copy of the written offer
No, you must have the offer, written or not. Simple knowledge that the offer must exist could be sufficient.
The second question clarifies that if you have the binaries, you automatically have the offer. However, the inverse, that without the binaries you do not have the offer, is not neccessarily true. Commonly, the reason to take up the offer is because you have the binaries, obviously.
But the GPL offer, AFAICT, does not restrict the offer to either those who have it received the binaries or the written offer, simply "any 3rd party".
Again from the GPL FAQ, the mail order interpretation (you must, if requested, provide the requested source on physical media posted through the mail) seems at odds with the actual language of the GPL which merely requires the source to be provided on "customary media".
Agreed. I dont think the GPL FAQ here reflects the reality of the GPL, the internet these days is a "customary medium". Maybe the GPL FAQ is out of date, who knows, anyway for whatever reason, yes, agreed, its interpretation seems incorrect.
The GPL does say that access to the source when downloading a binary must be "the same place" and "equivalent access"
No, it doesnt say that. It simply says that if you distribute binaries via giving access to a place for copying from (eg FTP server), and if you make the source available at the same place, then that counts as having distributed the source with the binaries. Ie you then do not need to provide the "3 year offer".
(latter link is also the one that claims that you can't distribute under 3b via anonymous FTP).
Yes, that GPL FAQ entry seems dodgy too, agreed. I dont see why one could not distribute binaries via FTP and source via a 3b offer.
However, I had always interpreted the 3a rule as not allowing an additional charge for the source code
I dont see how you could make such an interpretation. You may charge what you like for the binaries. (whether or not you provide source with the binaries).
You may only charge cost of physical distribution for source, if you have not provided source with the binaries.
while the interpretation seems to be that you can not charge more for the source than you did for the binary
Well, that question refers specifically to FTP (or other online access) access and rests on the assumption the GPL has that one can not avail of 3b if one provides net access to binaries. Anyway, the answer is valid: if one offers online access and wishes to avoid having to make a 3b offer then one must provide equivalent access, hence clearly one can not charge more for the source.
Or do they mean that if you charged $20 for the binary, you can't charge more than $20 for the sources, even if the sources take up 100 times as much space).
Correct, remember, one can charge what one likes for the binaries. So if one provides source with the binaries, to avoid obligations of having to make a 3b offer, then one would have to incredibly stupid to charge a price that does not cover ones costs (including the bandwidth needed for the source download). The GPL does try to protect redistributors of GPL software from their own stupidity;).
I'm looking at this from the viewpoint of someone trying to sell GPL software as if it was proprietary, making it as inconvenient and expensive for the average person to download the sources and pass them on, and attempt to prevent copying and redistribution of the binary as well, and see if the GPL properly blocks such attempts or if there are loopholes that should be fixed.
Sigh, but that applies to noncommercial *redistribution*, ie you must pass the offer along. Now, while this *should* mean that a 3rd party looking to take up a 3b offer will have the details of the offer in writing, as they will it received via 3c, there is nothing in GPL that limits the offer to those who have it in writing. Again, the pertinent phrase in the GPL as to whom the offer applies to is "any 3rd party", no more, no less, and certainly no requirement that a 3rd party who wishes to avail of the offer must have a copy of the offer in writing.
Further, the offeree is not bound by the GPL. See below. The only thing that matters is what the GPL requires of the offerer, they are the only ones bound to make an offer that meets the criteria defined in the GPL. Only the text of 3b is of interest in determining what the offer must consist of. (and hence what "any 3rd party" can accept). Note also that "offer" has a meaning in contract law, at least in the British legal system (from which many other legal systems across the world derive, including US).
I'm pretty sure I've seen an opinion from the FSF that the written offer is only valid for people who have received a copy of that written offer (or, as they put it "the information you received as to the offer"),
Well, only a judge could decide it for sure. "any 3rd party", to my thinking at least, means "anybody" particularly as 3b does *not* qualify it in any way (eg "any 3rd party" + "who posseses a copy of the work" or + "who was given a copy of the offer via 3c", etc.).
I'd be interested if you could find reference of this FSF opinion you say to remember that defines "any 3rd party" in such an unusual way.
so it would seem legitimate to require confirmation that you have said information, such as a password, or maybe an MD5 hash of the binary that the source corresponds to.
But the GPL 3b offer does *not* require that the offeree have a copy of the work, hence if someone were to try restrict the offer in such a way they would be in contravention of making the offer open to "any 3rd party". Ie any such requirement itself would need to be accessible to any 3rd party (ie publically available).
Also, the GPL does not apply to the offeree, as the offeree is not redistributing the work, only seeking to take up the offer made by someone who did redistribute the work and made a 3b offer. Only the offerer (as the reason the offer was made must be because the offerer distributed the work under the GPL within the last 3 years) is bound by the GPL.
I do think that the "written offer" part isn't very well defined. I know that the GPL FAQ seems to think it means "mail order", but I don't see any mention of that in the GPL.
The GPL does now refer to online redistribution. The offer is reasonably well specified AFAICT, if one presumes one may not add conditions to the offer. (see below).
I said Can they require you to have registered the binary program with them previously (along with that agreement earlier that you'll pay $150)? and you responded No, that would be an additional restriction.
Hmm, right that answer was hasty. Your hypothetical $150 charge referred to the source. And yes, you're right, it wouldnt be an additional restriction, however it would be wrong because it would be not an acceptable 3b offer (cant charge $150 for source). The registration requirement for binary is fine, but see my ansewr before that, there will be nothing to stop anyone removing the registration requirement and redistributing the modified registration-free binaries.
They only say you can't charge more than your cost for the written copy, they don't say you can't require anything else, which was the point of my last few hypotheticals.
That's an interesting question. However, nor does the GPL allow one to add additional requirements to the offer. Remember, one has *0* rights without the GPL, a
Actually, to request the sources, you must have the information from the written offer they gave to the person to whom they didn't provide source code.
Where on earth did you get this idea from?
All you need to know is that the GPL offer was made within the last 3 years. Now, it would help to know the details of the offer, and if the offerer refuses to honour the offer, having some kind of evidence, be it written or not, may help to convince them to reconsider. If they still refuse, I'm not sure if any 3rd party can take action or whether only the copyright holder can do so, you'd need to ask a lawyer (certainly the copyright holder *can* sue).
However you absolutely do not need a copy of the original offer in order to avail of it.
That written offer could, for instance, contain a password that is required to get the source, and that password could uniquely identify the person who passed it along.
Absolute tossology. To require some kind of password in order to avail of the offer is incompatible with the language of who the offer applies to in the GPL: "any 3rd party". The offerer would be in breach of the GPL.
Where are you getting mad ideas from?
However, I believe, as others have stated, that cutting off a subscription, or taking any other negative action against someone for the mere act of requesting sources or redistributing the source, the written offer, MD5 checksums, or the binaries is a "further restriction" that the GPL doesn't allow.
Correct.
Sell for really cheap a CD with a Linux installer on it for a nice distribution. The CD is in a case that has a seal saying "this is GPL software - you may request the source code by for this product by using the information on the enclosed written offer - you agree to pay $150 for each such copy requested with the enclosed offer." When you run the installer, it connects to your server and puts up a click-through license that takes the user's name, address, credit card information and requires the user to authorize a $5 charge to install Linux, and $150 for each use of the enclosed "written offer" for the source code
Firstly, how does this circumvent the GPL? Secondly, you could not charge $150 to download the source, as $150 is not a reasonable charge to perform distribution. Given that Apple charge 99c for multi-MB downloads, and that a significant part of that 99c is RIAA licencing costs, I'd suggest you could not charge more than 50c/10MB.
If the user, or anyone that gets the activation key, requests the source code, you start raking in the $150 charges (so much for "only the cost of copying").
Again you're mistaken, you can not charge $150 for access to source code distributed to the user seperately. Further, you *can* charge $150 for the binaries, you can charge $1000 or more if you want. So why charge for the source, especially when you say most people wont bother asking for the source?
Also, once someone has the source they can remove the activation key code and legally distribute the activation-key-free version.
So your idea simply will never fly.
Another loophole is what the "written offer" can require of you. It seems legitimate to require you to call a telephone number, go to a Web or FTP site, mail a letter - but in what form can they require payment?
Any normal payment mechanism one would presume. The GPL doesnt specify.
What if the telephone number you're required to call is a 900 number?
"900 number"? I presume these are premium rate numbers? Then that would count towards the cost of distribution, and could not exceed "reasonable".
Can they require you to enter a password, as I asserted earlier?
If the password is publically available, yes. Otherwise no, as it would be in conflict with "any third party".
Can they require you to prove Fermat's Last Theorem or factor a 4096-bit number as part of the process of g
Sveasoft cannot refuse source access FOR THOSE WHO HAVE OBTAINED A BINARY FROM THEM.
This is incorrect. If Sveasoft have ever, in the last 3 years, provided binaries to a customer without providing source, then they are bound by 3b of the GPL (emphasis is mine):
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange;
If Sveasoft have made use of 3b by not shipping source with their binaries, then they are obligated to provide the sources for those particular binaries to anyone who asks for it, for 3 years after shipping the GPLed software in its binary form.
Hmmm, ik ben niet met je eens.. ok, in nederland kan bijna iedereen die daar is opgegroeid engels spreken, maar nederland is een van de weinige landen waar TV programmas subtitled zijn in plaats van dubbed. Dus van een jonge leeftijd zit je bijna elke dag voor de kast een uurtje of meer engels te leren;). Mijn ervaring van niveau van engels in de rest van europa is dat het veel lager zit dan in nederland.
A single CPU computer can execute ONE instruction at the time.
Incorrect, a modern superscalar CPU can execute several instructions at the same time potentially. The pentium was the first Intel CPU able (very crudely) to do this, the P6 was 3-way superscalar (iirc - there was an article linked to on/. about it recently), able to retire (ie execute) 3 instructions per clock cycle. This implies some kind of pipeline (ie the processor must fetch several instructions at the same time from RAM and examine them and decide how to schedule them), which implies that actually such a CPU at any given time has a whole bunch of instructions in different stages of execution.
Meaning one program thread running at the time.
It does not mean that at all.
Most CPUs only support a single context of execution, however some CPUs support multiple execution contexts, intel "HyperThreading" would be one example. So a superscalar CPU with multiple execution contexts could have many instructions in several stages of execution from multiple programme contexts at any given point in time.
When it has switched to a program all the other programs are effectevily at the the mercy of the program now running INCLUDING the OS. Wich is why DOS and Windows and Linux and Mac OS and all the others had "hangups".
You're describing co-operatively multi-tasking operating systems, which linux is not. Ie systems like Windows 3 and MacOS 9 and earlier.
Linux is a preemptive multi-tasking OS, as is MacOS X, WinNT/2k and (partly) Win9x. Under such a system programmes are given only limited periods of time to run, a programme which does not yield control of the system by itself will be suspended eventually. Typically this is done by the OS setting a hardware timer, upon the expiration of which the hardware forcibly returns control to the OS, where upon it can elect to give control of the system to another process (setting that timer again, if needs be). On most hardware this is done with a timer interrupt, eg IRQ 0 on PC class machines, which fires at a preprogrammed interval (100 (older linux) or 1000HZ (2.6) or 1024HZ (Linux or Digital Unix on Alpha)), when the interrupt goes off, the CPU saves the state of the process (as it always does to handle interrupts) and runs the appropriate interrupt vector as installed by the operating system, which can then elect to run another process (usually requiring the OS to save some state of current running process which CPU hadnt saved or which is OS dependent and then restore state of another process).
Anyway, a process on Linux can *NOT* "hang" the system by refusing to yield control. The OS (with help from hardware) will intervene.
still remains a case that all the programs and the OS are fighting for time on 1 single cpu
This isnt really a good mental image to have of a modern OS. The OS does not "fight for time". The OS only ever runs because:
1. A process calls the OS to perform some service on that processes behalf.
Eg, to do work on that processes behalf such as IO (read/write from disk/network/whatever or IPC IO and deliver it to process/destination), or to setup the OS abstractions needed for IO (filehandles typically on Unix) or to interact with OS abstractions, eg to list a directory or running processes or send a signal to a process, etc.
1a. A subset of 1, where a process calls the OS to voluntarily yield control of the CPU it is executing on. The OS potentially can do some housekeeping here before restoring state of another process and allowing it to run.
2. The hardware directly intervenes and executes OS installed functions, typically in response to an interrupt generated by a timer or other hardware or else some exceptional event (typically a memory fault where a memory address is referenced that does not "exist").
Operating systems will typically try to do as little as possible work in the latter case and will try defer as much as p
You might have had a point if English wasn't the accepted norm for international communications and just something the States imposed on everyone.
Engels als internationaale taal is eigenlijk *wel* een geimposeerde taal. Alleen door het Britse koningrijk, niet America. Een grotere deel van het wereld zat, op een punt in tijd of 'n ander, onder het bestuur van Engeland. Engels was het taal van de overheid en van het handel, en daarom ook nu 'swerelds grootse bijtaal. Dat een van haar voormalig kolonies gegroeid is tot werelds meest machtig staat heeft engels wel geholpen maar heeft niks te maken met reden waarom engels nu zo ruim gesproken is.
(excuses als mijn nederlands vraai slecht is, woon al lang in buitenland).
Re:Performance is pretty reasonable
on
XORP 1.0 Released
·
· Score: 2, Interesting
The problem, IMHO, is that ALL high end routers use HARDWARE routing (see: flow/fast switching in 7500/12000s) instead of software routing. Unless you 're building ASICs to handle stuff in the data plane (VIPs or whatever the 12ks use for dCEF and the like), you're not really in any danger of becoming used by the higher end routing equipment manufacturers.
However, they still run their protocols, control "plane", etc. in software on a commodity general purpose CPU, which is what the likes of XORP, GNU Zebra and Quagga cover. Indeed, the Juniper routing engines are literally PC's running some flavour of BSD off of flash. There is nothing stopping one implementing off-board forwarding cards for a PC - you just end up with Juniper's architecture. Intel for example have ASICs targeted toward the building such boards, the Intel Network Processor range, customised Xscale CPUs with PCI interfaces designed for offloading packet-forwarding.
Still, a PC is *more* than capable of replacing any low-end Cisco, eg 26xx, which btw use software forwarding, not hardware, and even mid-range, provided one is careful to match the PC hardware to the requirements.
Cisco at least notified their large carriers before specific details leaked onto the net - I shudder to think of someone posting 0day exploit code for something like this on Full-Disclosure.
There was a Cisco BGP DoS vulnerability announced recently, GNU Zebra and Quagga were not vulnerable to the DoS. Also, why do you think white hats would leak a DoS for an open project but not for IOS? Or why do you think CERT, would not co-ordinate with an open project for vulnerabilities, when they already do so?
There's also Quagga, a fork of the GNU Zebra (thanks Kunihiro), which is further along, more mature, in much wider use than XORP (I've not heard of anyone actually using XORP in production, while GNU Zebra and Quagga most definitely are) and, most importantly, not written in C++;).
No, it's about "one" particular root nameserver, F-root, which is the root ISC operate. It's one IPv4 address, but actually a whole bunch of machines located across the world.
I'm replying to this, rather than the actual (AC posted) post I should be replying to, so that you get to read it;)
The Netherlands/low-countries did once upon a very long time ago consist of more than it does now. However, the French spun off several provinces into the seperate states of Belgium and Luxembourg. See the wikipedia entry for the seventeen provinces. As that entry notes, one must be careful if one wishes to distinguish between the Netherlands as it was then and the Netherlands as it is now. However, this problem applies to nearly all European countries, as, other than countries whose borders are immutably defined by geography (eg island nations such as Ireland and Britain, and even these two have different borders and state of nationship at differing points in history.), borders, indeed entire states, have waxed and waned through history, including Spain.
However, the fact remains that there is no such country as Holland or Holanda and that the name of the country is Nederland (in dutch), the Netherlands in english and, most definitely, los Paises Bajos in Spanish:). Just to emphasise the point, please visit the website for the website of the dutch embassy in madrid, http://www.embajadapaisesbajos.es/.
The funny thing is that, if Islamic extremism is the biggest threat to the US's security these days, Iraq under Saddam was the far more benign than what Iraq is today and, even more so, than what Iraq is likely to become.
Iraq under Saddam was very secular, adherence to even basic Islamic principles was not much of a concern, one of the top men in Saddam's regime, Tariq Aziz, was a christian. However, ow that Iraq has been utterly destabilised and the US military is likely to have a presence there for quite some time, Iraq has become a magnet for Islamic fundamentalists. Quite ironic that the US invasion of Iraq is what has allowed "al Queda" to really get hold in Iraq - Saddam would not have allowed it;). And whatever stable government that will eventually emerge from the power vacuum will very likely be quite Islamist in nature and anti-american. The majority of the population in Iraq, repressed by Saddam, are Shi'a, as in Iran, and are generally "stricter" muslims. The chaos and security problems are not conducive to having moderates gain power.
Israel has to be the primary reason why one would invade Iraq, given the idealogy of the Bush administration. There was just nothing in the way of any credible security threat presented by Iraq to the USA to justify the invasion. However, Saddam was a threat to Israel, not least because he funded suicide bombers, also because he had bombed Israel in the first Gulf War, and might possibly have tried to do so again if he had ever managed to reacquire the capability to do so.
Also, this Bush administration has well-documented links with the military-industrial complex in the US. War, occupation and rebuilding requires significant spending. And what better favour to do for your old boardroom buddies than by opening the petcocks of the Federal purse and pouring money into Halliburton, et al by way of contracts awarded without tender, never mind the military supply spending? Kind of like how Saddam would divert a lot of the UN Oil-for-Food money into building nice palaces, etc.. for himself and his buddies, except instead of Saddam thieving money that should have been spent on food and medicine, the Bush administration is thieving its own nation's taxpayers money, who get body bags in return (and a skyrocketing deficit).:(
France had major interests in Iraq. ELF, the state-owned French petro-chemical giant, and Total[1], had major interests in Iraq, see eg: http://www.payk.net/mailingLists/iran-news/html/ 19 97/msg01212.html or just google for "elf iraq". There are pictures of Jacques Chirac with Saddam from the 1970s. Of course, there are more recent pictures of Donald Rumsfeld with Saddam.
The Mujahideen (NB: there are a variety of english spellings for the arabic word, as with most arabic words.) btw does not equal the taliban. See the wikipedia entry for Mujahideen. It's a general word. In the afghani case, the taliban were but one faction of the collective resistance movement known as the mujahideen. After the war with the Russians, there was civil war between the Taliban and the other factions, the taliban gaining control of most, but not all, of Afghanistan.
As for motives. Let's be honest, every major power which takes an interest in the middle-east does so because of oil. Additionally, the US has a strong political affiliation with Israel, and has long been very involved in assuring Israeli security. The current administration in particular is quite interested in Israel. See Project for a New American Century (PNAC), there are papers there dating to before the present administration gained power making the case for taking out Iraq, reasoned by way of taking out a potential threat of WMD proliferation and stabilising the middle-east and gaining security for Israel. So taking out Iraq is something the the people behind the Bush administration have had as a goal since long before 20010911.
1. Elf, Total and the belgian PetroFina have all since merged together into TotalFinaElf. Total bought Petrofina at some point and then TotalFina merged with Elf in 2000.
Slashdot Mirror
Remember, the "equivalent access " clause is referring to "offering access to copy from a designated place", such as downloading (though not necessarily so). I think you're putting too much weight on equivalent access. My making the copies and offering them to you does not have anything to do with "equivalent access".
:)
Hmm, I'm not sure I follow you. And I dont feel i'm putting too much weight on equivalent access. I get the impression you're interpreting the GPL as:
"copy from designated place" == "equivalent access" and hence fulfilling former -> "3a is fulfilled".
However that is not the way the GPL words it, the GPL is worded as:
"if copy from designated place" AND "equivalent access" then "3a is fullfilled".
So "equivalent access" is a critical determining test as to whether you fulfill 3a when distributing via "copy from designated place". And the GPL does not define or further qualify "equivalent access", that's left for professional interpretation, possibly judicial interpretation as a result of litigation. The exact wording of the GPL is:
If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code
Ie: if (copy binary from designated place), then if (equivalent-access to source at same place) -> 3a considered fulfilled.
3a doesn't say anything about equivalent access, 3a doesn't say anything about source only, 3a doesn't say anything about written offers. If 3a can be interpreted to mean you CAN add an additional charge for the act of distributing the sources (plus the binary), as opposed to just distributing the binary, then it isn't really requiring the sources to be distributed.
You are absolutely correct, bar in one critical detail. Yes, you can charge whatever the hell you like for source, or for binary+source. The detail you are missing though or the fallacy if you wish, I believe, is that you seem to think you can distribute binaries and only *offer* source and still fulfill 3a somehow.
You can not. As you point out yourself, there is no talk of offers in 3a. However, it is impossible to avail of 3a if you distribute binary only - the *offer* of source is not enough. Specifically, in order to distribute binaries under 3a you must:
Accompany it with the complete corresponding machine-readable source code
I fail to see by what reading you think you can ship only binaries and yet fulfill the above. If you distribute binaries without source commercially then you *must* be distributing under 3b, and hence you *must* provide an offer of source at cost, valid for 3 years.
There is simply nothing else in the GPL that allows you to distribute binaries commercially.
Either you ship your binaries under 3a:
- source accompanying the binaries or equivalent-access to binaries and source from a designated place (eg a password protected FTP site, or a special one-time, customer-specific HTTP URL).
Or you ship under 3b:
- ship binaries, offer the customer the right to obtain source at cost.
Again, you can *not* avail of 3a if you give your customer a binary-only CD for $30 and *offer* source @ whatever price. Only 3b allows you to not ship source, but then you can only charge cost. (and yes, if the media for the source genuinely costs $900, you can charge that. And no, you can not pad it with 0s - at least you cant charge for it, it's not part of the cost of copying the source.).
If 3a can be interpreted to mean you CAN add an additional charge for the act of distributing the sources (plus the binary), as opposed to just distributing the binary, then it isn't really requiring the sources to be distributed.
How? 3a *unambigiously* requires you to ship the source (on physical media, or same place + equivalent access). There
You must have the world's worst teeth if you can fit a DIMM in between them...
Sigh, I wish one could edit one's own posts. To fix the most glaring mistake in my post:
;) )
Unix in part draws it's strength and health from diversity, from not being a heterogenous meritocracy, not a homogenous monopoly.
Should of course read:
Unix draws its strength and health from diversity, from being a heterogenous meritocracy, not a homogenous monopoly.
This:
Sure, Sun would prefer to sell you Solaris, and why not,
Should read:
Sure, Sun would prefer to sell you Solaris, and why not?
Though, even with change, that paragraph still is not structured terribly well. Ah well.. (and yes, I did preview.
Actually it is a product of the language spoken by Friesians and other tribes of Angles who settled in Britannia, mixed with latin. Then later the Vikings came along and settled in Angleland, mostly in the north and again influenced the language. Then a bunch of accountants, sorry Normans, (descended from the Vikings who invaded northern Gaul and settled with the local Gauls) invaded from France, and again influenced the language and culture.
After that, England was never invaded again, and the language never again underwent radical changes, just slow evolution, through immigration, exploration and colonisation. Which is why Chaucer's Canterbury tales, 700 hundred odd years old, are still readable, even to a modern english speaker.
I cant think of any differences in modern english that are a product of the USA, other than some trivial spelling and usage differences.
I don't keep track, but if there are any IBM-Sun negotiations or lawsuits going on, this comment probably has to do with that.
IBM are one of Suns' main competitors. Both sell PC servers, RISC servers, Linux, Unix®, Java, software support, a range of "middleware" software and IT consultancy services.
Sun, like all companies that actually have competition, will jostle for position with, make statements about and generally try to best and outsmart their competition. That's how competition works...
Since sun owns unix source (via the license with old ATT), they can always sell Linux. Now, if they can prevent others, well, then they have a case about solaris.
I really cant even imagine Sun trying to do something like that though.
I'm reading the comments on this story and I'm just amazed at how many comments are so hostile to Sun - I just dont understand where this hostility comes from. Sure I can understand people being critical of Sun, and criticicism is good, but this outright hatred is just weird.
Now, I'm a (recent) Sun employee[2], so maybe I'm blinded by my paycheck, but it seems to me that to consider a company that:
as being a reasonable pariah for the Linux community is just strange.
So Sun still push Solaris over Linux, well why wouldnt Sun? Sun have spent a long time working on it, the people at Sun are proud of Solaris. Surely they have as much right to be proud of their (their, cause I havnt contributed to Solaris) work as the "Linux" developers[1] have to be of theirs? And even so, Sun still do spend money on technologies that are of benefit to Unix in general, be it Solaris, Linux, BSD, whatever.. and spend money marketing what is effectively Linux.
So Sun bought out licence rights from SCO, how evil of them, but if you're responsible for Sun and you have a chance to fully secure your "IP" (yuk) rights wouldn't it be corporate irresponsibility to not do so? Remember, you can be sued by shareholders for your inactions as much as your actions.
So Sun settled a long-running dispute with MS, how evil of them. But MS infringed on Suns' rights, is Sun not allowed to get a fat cheque from MS for MSs' wrongdoing, should Sun instead have continued litigating the matter at great expense and uncertainty? Would Sun maybe then later being awarded a fat cheque from MS by court order have then *not* been evil? The settlement recompenses Sun for wrong done to it and lets Sun get on with things, why is that evil?
At the end of the day, Sun are a Unix company. Sun are not perfect, no entity is, and Sun will have to adapt to changing market conditions, as all companies do, but they're the only big company who are and have been 100% committed to Unix from day one of their existence. Sure, Sun would prefer to sell you Solaris, and why not, Solaris is still Unix, and work on any one Unix ultimately benefits all unixes, be it directly or by virtue of competition. Never mind that Sun also directly contribute to technologies/projects that are key to Linux, as well as many other cross-platform projects, and also market Linux in one segment of the market.
The irony of course is that most of these /. weenies who like to spout this ill-informed "Sun is evil, they hate Linux!!!" clap-trap are likely doing so from the "comfort" of their Win32 PCs.
Linux, Solaris, BSD, etc.. So what, they're all Unix. Unix in part draws it's strength and health from diversity, from not being a heterogenous meritocracy, not a homogenous monopoly. Sun has long been a valuable contributor to that meritocracy of ideas.
Vive la difference!
1. What is a Linux developer exactly, aside from Linux kernel developers? I work on stuff at Sun that runs on Linux and Solaris. It's all Unix to me..
2. NB: I do not speak for Sun, opinions in this post are my own. Statemen
It is possible that Sun would give them to SCO since it is almost certain that SCO has no real case at this point.
Why would Sun do that?
--paulj
(A sun employee, *not* speaking, or asking questions in this case, for or on behalf of Sun).
This is pretty cool, I think we're converging on agreement.
;)
:). The GPL uses the term "same place" to clarify the meaning of equivalent-access, so any scheme that seperates the two in a significant way (eg $20 to get access to the binaries, $20 to get access to the source) does not sound like equivalent-access to me.
Indeed
I see now we simply failed to communicate, as what I was referring to was "here's the binary for $20, do you want the source CD for another $20" - that's "an additional charge for the source code" under 3a.
Hmmm.. possible, presuming it counts as equivalent access. I'm not sure it does though, because in order to be able to say "For an extra $20 you can download the source" implies that source and binary acess are *not* equivalent - otherwise how do you enforce that people must pay the $20 extra for the source?
If it does not fall under equivalent-access, then one is required to extend a 3b offer to anyone who downloads only the binary, hence rendering the "want source for $20 extra?" useless (unless of course the $20 is at cost).
Btw, I dont agree with the GPL FAQ answer you quoted, it seems simplistic at best and dangerously misleading at worst, if not downright wrong
But it isnt wrong though, you *are* allowed to charge whatever you like for source. Just anyone who downloads *only* the binaries under your scheme is then entitled to a 3b offer, which presumably will be much less that $20.
In other words, if my "fee for downloading source" is the same as my fee for downloading binaries, which is $5/megabyte for a 4 megabyte binary, and I pack the source tar file with the source packages for gcc, glibc, make, bash, texinfo, binutils, textutils, fileutils, shellutils, binutils, and emacs,
You can do that, but you're not likely to stay in business for long. Also, you may be open to false-advertising suits if you advertise "100s of MBs of free software!" and the bulk of it is 0-padding.
I'm going to be doing exactly what you said: binary for $REASONABLE_AMOUNT, source for $TOTALLY_RIDICULOUS_AMOUNT_THAT_I_CAN_RETIRE_ON,
Yes, you can do this, there is no restriction on how much charge, except for 3b source. So there is no problem in charging $TOTALLY_RIDICULOUS_AMOUNT_THAT_I_CAN_RETIRE_ON for the source. The only thing is that any sensible consumer will *not* buy the source CD, they'll buy the binary CD and then come back to you and ask for the source at cost price under 3b. So the best thing to do is to *not* provide binaries at all, and hence avoid having to make a 3b offer altogether. If the source is valuable enough you will get your OTALLY_RIDICULOUS_AMOUNT_THAT_I_CAN_RETIRE_ON, market will decide whether it's worth it. If it isnt, you might need to lower your price.
(this presumes your scheme is not considered equivalent access, which i dont think it can be reasonably considered to be so.)
effectively making a binary-only distribution.
Nope, cause I just ignore your $HUGE_AMOUNT source CD, buy the binary CD for $REASONABLE_AMOUNT and then avail of the 3b offer to get the source at cost.
The purpose I was referring to is the GPL's, which is to make sure that anyone who receives a copy of the binary can get a copy of the source for reasonable terms. I see a conflict with your responses; on the one hand, your explanation of "No you cant, because then you're not distributing the source with the binaries obviously. " is pretty much exactly what I was trying to get at, but then you also agree that you can charge whatever you want for physical media.
Yes, you can charge as much as you want, either for source, for binaries, or for source+binaries.
Distributing source does not imply distributing under 3b. You can distribute source and charge $HUGE_AMOUNT.
Only the 3b offer of source is restricted in cost, and you must make it if you distribute binaries without equivalent-access to source.
Well, "the information". Just guessing that they have it and asking for it doesn't seem to be sufficient, at least based on the two references in the GPL FAQ that I gave you.
:)
Well, the GPL FAQ is not infallible, as we both seem agreed on and anyway irrelevant. Alternatively, look on it that the GPL FAQ restricts itself to giving answers that are almost certainly correct, or more likely to hold up in court, but that this need not mean that other alternative answers would not also hold true (whether alongside the GPL FAQ answer or not). Eg, having a binary copy of the work and a copy of the 3c offer almost certainly entitles one to avail of the original 3b offer, which is the answer the GPL FAQ gives, yet this need not preclude the possibility that literally any 3rd party could avail of the offer, be that party a holder of a 3c offer and/or the binaries or not.
Ultimately only the opinion of a judge will matter. And it seems to me that an offer open "to any 3rd party" leaves a lot of scope for a lawyer to argue that this does not require the offeree to have the full details of the offer, written or not or to have the binaries, in order to be able to avail of the offer. Remember, if one knows the offer must exist, one already will know the general form of the offer, as specified by 3b. Whether that is enough, only a judge could decide, and it might even depend on other context.
The counter-argument is that any 3rd party could only receive the offer because of 3c and hence should have the details. However, what of a 3rd party who is given the binaries but not the full details of the 3b offer? Are they excluded from availing of the original 3b offer? What if they have the offer, but not the binaries? If the offer only applies to those who receive it via a 3c offer, that would indeed require one to have the binaries.
But then that is obviously a tighter definition than "any 3rd party". If the GPL had intended the offer to be restricted in such ways, why doesnt it then have language to qualify "any 3rd party" thus?
Anyway, one for a judge to decide
Regarding 3a, what I meant was that if I sell you a binary release on CD for $20, can I also charge you an additional $20 for the source CD (which you can decline if you wish)?
No you cant, because then you're not distributing the source with the binaries obviously. It's quite simple as I see it:
- distribute the work in full (ie binaries and source) and you can charge whatever you like (3a) and you have no further obligations (ie no need to make a 3b offer).
- distribute only binary versions of the work, then you can charge whatever you like, but you must additionally make a 3b offer and provide source at reasonable cost for performing the copy, no more.
- distribute source, and you can charge whatever you like.
Now, for this last case you'd probably want to make sure to distribute *only* source, or at least, do not ever distribute binary only because then you'd have to make a 3b offer and people wouldnt need to buy the source, they could just get at cost via the 3b offer, presuming my interpretation of "any 3rd party" is correct, and your interpretation of "must have the binaries" or "must have copy of offer" is not.
The FAQ says you can charge a fee for downloading, why wouldn't that apply to charging a fee for physical media?
You can yes. You can charge whatever you want for a GPL work, be it source only, binary only or binary+source. The only place where are you restricted is in distributing source as part of a 3b offer, which you must make if you distribute binaries and charge more than cost for it (3c being noncommercial, for whatever definition of noncommercial).
If I can charge you a different amount for "binary + source" than "binary (but I offered you source and you didn't want it)", and both fall under 3a (thus no written offer, no obligation to provide source in the future), then that seems to defeat the purpo
Regarding who can execute the "written offer", from the GPL FAQ (also see this later question) seems to clearly indicate that you must have a copy of the written offer
;).
No, you must have the offer, written or not. Simple knowledge that the offer must exist could be sufficient.
The second question clarifies that if you have the binaries, you automatically have the offer. However, the inverse, that without the binaries you do not have the offer, is not neccessarily true. Commonly, the reason to take up the offer is because you have the binaries, obviously.
But the GPL offer, AFAICT, does not restrict the offer to either those who have it received the binaries or the written offer, simply "any 3rd party".
Again from the GPL FAQ, the mail order interpretation (you must, if requested, provide the requested source on physical media posted through the mail) seems at odds with the actual language of the GPL which merely requires the source to be provided on "customary media".
Agreed. I dont think the GPL FAQ here reflects the reality of the GPL, the internet these days is a "customary medium". Maybe the GPL FAQ is out of date, who knows, anyway for whatever reason, yes, agreed, its interpretation seems incorrect.
The GPL does say that access to the source when downloading a binary must be "the same place" and "equivalent access"
No, it doesnt say that. It simply says that if you distribute binaries via giving access to a place for copying from (eg FTP server), and if you make the source available at the same place, then that counts as having distributed the source with the binaries. Ie you then do not need to provide the "3 year offer".
(latter link is also the one that claims that you can't distribute under 3b via anonymous FTP).
Yes, that GPL FAQ entry seems dodgy too, agreed. I dont see why one could not distribute binaries via FTP and source via a 3b offer.
However, I had always interpreted the 3a rule as not allowing an additional charge for the source code
I dont see how you could make such an interpretation. You may charge what you like for the binaries. (whether or not you provide source with the binaries).
You may only charge cost of physical distribution for source, if you have not provided source with the binaries.
while the interpretation seems to be that you can not charge more for the source than you did for the binary
Well, that question refers specifically to FTP (or other online access) access and rests on the assumption the GPL has that one can not avail of 3b if one provides net access to binaries. Anyway, the answer is valid: if one offers online access and wishes to avoid having to make a 3b offer then one must provide equivalent access, hence clearly one can not charge more for the source.
Or do they mean that if you charged $20 for the binary, you can't charge more than $20 for the sources, even if the sources take up 100 times as much space).
Correct, remember, one can charge what one likes for the binaries. So if one provides source with the binaries, to avoid obligations of having to make a 3b offer, then one would have to incredibly stupid to charge a price that does not cover ones costs (including the bandwidth needed for the source download). The GPL does try to protect redistributors of GPL software from their own stupidity
I'm looking at this from the viewpoint of someone trying to sell GPL software as if it was proprietary, making it as inconvenient and expensive for the average person to download the sources and pass them on, and attempt to prevent copying and redistribution of the binary as well, and see if the GPL properly blocks such attempts or if there are loopholes that should be fixed.
That's a good idea I guess.
Why, from the GPL itself. 3 c:
Sigh, but that applies to noncommercial *redistribution*, ie you must pass the offer along. Now, while this *should* mean that a 3rd party looking to take up a 3b offer will have the details of the offer in writing, as they will it received via 3c, there is nothing in GPL that limits the offer to those who have it in writing. Again, the pertinent phrase in the GPL as to whom the offer applies to is "any 3rd party", no more, no less, and certainly no requirement that a 3rd party who wishes to avail of the offer must have a copy of the offer in writing.
Further, the offeree is not bound by the GPL. See below. The only thing that matters is what the GPL requires of the offerer, they are the only ones bound to make an offer that meets the criteria defined in the GPL. Only the text of 3b is of interest in determining what the offer must consist of. (and hence what "any 3rd party" can accept). Note also that "offer" has a meaning in contract law, at least in the British legal system (from which many other legal systems across the world derive, including US).
I'm pretty sure I've seen an opinion from the FSF that the written offer is only valid for people who have received a copy of that written offer (or, as they put it "the information you received as to the offer"),
Well, only a judge could decide it for sure. "any 3rd party", to my thinking at least, means "anybody" particularly as 3b does *not* qualify it in any way (eg "any 3rd party" + "who posseses a copy of the work" or + "who was given a copy of the offer via 3c", etc.).
I'd be interested if you could find reference of this FSF opinion you say to remember that defines "any 3rd party" in such an unusual way.
so it would seem legitimate to require confirmation that you have said information, such as a password, or maybe an MD5 hash of the binary that the source corresponds to.
But the GPL 3b offer does *not* require that the offeree have a copy of the work, hence if someone were to try restrict the offer in such a way they would be in contravention of making the offer open to "any 3rd party". Ie any such requirement itself would need to be accessible to any 3rd party (ie publically available).
Also, the GPL does not apply to the offeree, as the offeree is not redistributing the work, only seeking to take up the offer made by someone who did redistribute the work and made a 3b offer. Only the offerer (as the reason the offer was made must be because the offerer distributed the work under the GPL within the last 3 years) is bound by the GPL.
I do think that the "written offer" part isn't very well defined. I know that the GPL FAQ seems to think it means "mail order", but I don't see any mention of that in the GPL.
The GPL does now refer to online redistribution. The offer is reasonably well specified AFAICT, if one presumes one may not add conditions to the offer. (see below).
I said Can they require you to have registered the binary program with them previously (along with that agreement earlier that you'll pay $150)? and you responded No, that would be an additional restriction.
Hmm, right that answer was hasty. Your hypothetical $150 charge referred to the source. And yes, you're right, it wouldnt be an additional restriction, however it would be wrong because it would be not an acceptable 3b offer (cant charge $150 for source). The registration requirement for binary is fine, but see my ansewr before that, there will be nothing to stop anyone removing the registration requirement and redistributing the modified registration-free binaries.
They only say you can't charge more than your cost for the written copy, they don't say you can't require anything else, which was the point of my last few hypotheticals.
That's an interesting question. However, nor does the GPL allow one to add additional requirements to the offer. Remember, one has *0* rights without the GPL, a
Actually, to request the sources, you must have the information from the written offer they gave to the person to whom they didn't provide source code.
Where on earth did you get this idea from?
All you need to know is that the GPL offer was made within the last 3 years. Now, it would help to know the details of the offer, and if the offerer refuses to honour the offer, having some kind of evidence, be it written or not, may help to convince them to reconsider. If they still refuse, I'm not sure if any 3rd party can take action or whether only the copyright holder can do so, you'd need to ask a lawyer (certainly the copyright holder *can* sue).
However you absolutely do not need a copy of the original offer in order to avail of it.
That written offer could, for instance, contain a password that is required to get the source, and that password could uniquely identify the person who passed it along.
Absolute tossology. To require some kind of password in order to avail of the offer is incompatible with the language of who the offer applies to in the GPL: "any 3rd party". The offerer would be in breach of the GPL.
Where are you getting mad ideas from?
However, I believe, as others have stated, that cutting off a subscription, or taking any other negative action against someone for the mere act of requesting sources or redistributing the source, the written offer, MD5 checksums, or the binaries is a "further restriction" that the GPL doesn't allow.
Correct.
Sell for really cheap a CD with a Linux installer on it for a nice distribution. The CD is in a case that has a seal saying "this is GPL software - you may request the source code by for this product by using the information on the enclosed written offer - you agree to pay $150 for each such copy requested with the enclosed offer." When you run the installer, it connects to your server and puts up a click-through license that takes the user's name, address, credit card information and requires the user to authorize a $5 charge to install Linux, and $150 for each use of the enclosed "written offer" for the source code
Firstly, how does this circumvent the GPL? Secondly, you could not charge $150 to download the source, as $150 is not a reasonable charge to perform distribution. Given that Apple charge 99c for multi-MB downloads, and that a significant part of that 99c is RIAA licencing costs, I'd suggest you could not charge more than 50c/10MB.
If the user, or anyone that gets the activation key, requests the source code, you start raking in the $150 charges (so much for "only the cost of copying").
Again you're mistaken, you can not charge $150 for access to source code distributed to the user seperately. Further, you *can* charge $150 for the binaries, you can charge $1000 or more if you want. So why charge for the source, especially when you say most people wont bother asking for the source?
Also, once someone has the source they can remove the activation key code and legally distribute the activation-key-free version.
So your idea simply will never fly.
Another loophole is what the "written offer" can require of you. It seems legitimate to require you to call a telephone number, go to a Web or FTP site, mail a letter - but in what form can they require payment?
Any normal payment mechanism one would presume. The GPL doesnt specify.
What if the telephone number you're required to call is a 900 number?
"900 number"? I presume these are premium rate numbers? Then that would count towards the cost of distribution, and could not exceed "reasonable".
Can they require you to enter a password, as I asserted earlier?
If the password is publically available, yes. Otherwise no, as it would be in conflict with "any third party".
Can they require you to prove Fermat's Last Theorem or factor a 4096-bit number as part of the process of g
Doesn't Solaris 10 have x86-64 support?
It will have yes. According to the Register Sun have Solaris 10 x86-64 booting and multi-user.
Sveasoft cannot refuse source access FOR THOSE WHO HAVE OBTAINED A BINARY FROM THEM.
This is incorrect. If Sveasoft have ever, in the last 3 years, provided binaries to a customer without providing source, then they are bound by 3b of the GPL (emphasis is mine):
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange;
If Sveasoft have made use of 3b by not shipping source with their binaries, then they are obligated to provide the sources for those particular binaries to anyone who asks for it, for 3 years after shipping the GPLed software in its binary form.
Hmmm, ik ben niet met je eens.. ok, in nederland kan bijna iedereen die daar is opgegroeid engels spreken, maar nederland is een van de weinige landen waar TV programmas subtitled zijn in plaats van dubbed. Dus van een jonge leeftijd zit je bijna elke dag voor de kast een uurtje of meer engels te leren ;). Mijn ervaring van niveau van engels in de rest van europa is dat het veel lager zit dan in nederland.
A single CPU computer can execute ONE instruction at the time.
/. about it recently), able to retire (ie execute) 3 instructions per clock cycle. This implies some kind of pipeline (ie the processor must fetch several instructions at the same time from RAM and examine them and decide how to schedule them), which implies that actually such a CPU at any given time has a whole bunch of instructions in different stages of execution.
Incorrect, a modern superscalar CPU can execute several instructions at the same time potentially. The pentium was the first Intel CPU able (very crudely) to do this, the P6 was 3-way superscalar (iirc - there was an article linked to on
Meaning one program thread running at the time.
It does not mean that at all.
Most CPUs only support a single context of execution, however some CPUs support multiple execution contexts, intel "HyperThreading" would be one example. So a superscalar CPU with multiple execution contexts could have many instructions in several stages of execution from multiple programme contexts at any given point in time.
When it has switched to a program all the other programs are effectevily at the the mercy of the program now running INCLUDING the OS. Wich is why DOS and Windows and Linux and Mac OS and all the others had "hangups".
You're describing co-operatively multi-tasking operating systems, which linux is not. Ie systems like Windows 3 and MacOS 9 and earlier.
Linux is a preemptive multi-tasking OS, as is MacOS X, WinNT/2k and (partly) Win9x. Under such a system programmes are given only limited periods of time to run, a programme which does not yield control of the system by itself will be suspended eventually. Typically this is done by the OS setting a hardware timer, upon the expiration of which the hardware forcibly returns control to the OS, where upon it can elect to give control of the system to another process (setting that timer again, if needs be). On most hardware this is done with a timer interrupt, eg IRQ 0 on PC class machines, which fires at a preprogrammed interval (100 (older linux) or 1000HZ (2.6) or 1024HZ (Linux or Digital Unix on Alpha)), when the interrupt goes off, the CPU saves the state of the process (as it always does to handle interrupts) and runs the appropriate interrupt vector as installed by the operating system, which can then elect to run another process (usually requiring the OS to save some state of current running process which CPU hadnt saved or which is OS dependent and then restore state of another process).
Anyway, a process on Linux can *NOT* "hang" the system by refusing to yield control. The OS (with help from hardware) will intervene.
still remains a case that all the programs and the OS are fighting for time on 1 single cpu
This isnt really a good mental image to have of a modern OS. The OS does not "fight for time". The OS only ever runs because:
1. A process calls the OS to perform some service on that processes behalf.
Eg, to do work on that processes behalf such as IO (read/write from disk/network/whatever or IPC IO and deliver it to process/destination), or to setup the OS abstractions needed for IO (filehandles typically on Unix) or to interact with OS abstractions, eg to list a directory or running processes or send a signal to a process, etc.
1a. A subset of 1, where a process calls the OS to voluntarily yield control of the CPU it is executing on. The OS potentially can do some housekeeping here before restoring state of another process and allowing it to run.
2. The hardware directly intervenes and executes OS installed functions, typically in response to an interrupt generated by a timer or other hardware or else some exceptional event (typically a memory fault where a memory address is referenced that does not "exist").
Operating systems will typically try to do as little as possible work in the latter case and will try defer as much as p
You might have had a point if English wasn't the accepted norm for international communications and just something the States imposed on everyone.
Engels als internationaale taal is eigenlijk *wel* een geimposeerde taal. Alleen door het Britse koningrijk, niet America. Een grotere deel van het wereld zat, op een punt in tijd of 'n ander, onder het bestuur van Engeland. Engels was het taal van de overheid en van het handel, en daarom ook nu 'swerelds grootse bijtaal. Dat een van haar voormalig kolonies gegroeid is tot werelds meest machtig staat heeft engels wel geholpen maar heeft niks te maken met reden waarom engels nu zo ruim gesproken is.
(excuses als mijn nederlands vraai slecht is, woon al lang in buitenland).
The problem, IMHO, is that ALL high end routers use HARDWARE routing (see: flow/fast switching in 7500/12000s) instead of software routing. Unless you 're building ASICs to handle stuff in the data plane (VIPs or whatever the 12ks use for dCEF and the like), you're not really in any danger of becoming used by the higher end routing equipment manufacturers.
However, they still run their protocols, control "plane", etc. in software on a commodity general purpose CPU, which is what the likes of XORP, GNU Zebra and Quagga cover. Indeed, the Juniper routing engines are literally PC's running some flavour of BSD off of flash. There is nothing stopping one implementing off-board forwarding cards for a PC - you just end up with Juniper's architecture. Intel for example have ASICs targeted toward the building such boards, the Intel Network Processor range, customised Xscale CPUs with PCI interfaces designed for offloading packet-forwarding.
Still, a PC is *more* than capable of replacing any low-end Cisco, eg 26xx, which btw use software forwarding, not hardware, and even mid-range, provided one is careful to match the PC hardware to the requirements.
Cisco at least notified their large carriers before specific details leaked onto the net - I shudder to think of someone posting 0day exploit code for something like this on Full-Disclosure.
There was a Cisco BGP DoS vulnerability announced recently, GNU Zebra and Quagga were not vulnerable to the DoS. Also, why do you think white hats would leak a DoS for an open project but not for IOS? Or why do you think CERT, would not co-ordinate with an open project for vulnerabilities, when they already do so?
It's an ironic pun by intention.
Also, in Afrikaans (that strange descendent of dutch spoken in South Africa) Zebra's are still called Quagga, or kwagga or somesuch.
There's also Quagga, a fork of the GNU Zebra (thanks Kunihiro), which is further along, more mature, in much wider use than XORP (I've not heard of anyone actually using XORP in production, while GNU Zebra and Quagga most definitely are) and, most importantly, not written in C++ ;).
NB: I'm biased.
it's about the root name servers
No, it's about "one" particular root nameserver, F-root, which is the root ISC operate. It's one IPv4 address, but actually a whole bunch of machines located across the world.
I'm replying to this, rather than the actual (AC posted) post I should be replying to, so that you get to read it ;)
:). Just to emphasise the point, please visit the website for the website of the dutch embassy in madrid, http://www.embajadapaisesbajos.es/.
The Netherlands/low-countries did once upon a very long time ago consist of more than it does now. However, the French spun off several provinces into the seperate states of Belgium and Luxembourg. See the wikipedia entry for the seventeen provinces. As that entry notes, one must be careful if one wishes to distinguish between the Netherlands as it was then and the Netherlands as it is now. However, this problem applies to nearly all European countries, as, other than countries whose borders are immutably defined by geography (eg island nations such as Ireland and Britain, and even these two have different borders and state of nationship at differing points in history.), borders, indeed entire states, have waxed and waned through history, including Spain.
However, the fact remains that there is no such country as Holland or Holanda and that the name of the country is Nederland (in dutch), the Netherlands in english and, most definitely, los Paises Bajos in Spanish
The funny thing is that, if Islamic extremism is the biggest threat to the US's security these days, Iraq under Saddam was the far more benign than what Iraq is today and, even more so, than what Iraq is likely to become.
;). And whatever stable government that will eventually emerge from the power vacuum will very likely be quite Islamist in nature and anti-american. The majority of the population in Iraq, repressed by Saddam, are Shi'a, as in Iran, and are generally "stricter" muslims. The chaos and security problems are not conducive to having moderates gain power.
:(
Iraq under Saddam was very secular, adherence to even basic Islamic principles was not much of a concern, one of the top men in Saddam's regime, Tariq Aziz, was a christian. However, ow that Iraq has been utterly destabilised and the US military is likely to have a presence there for quite some time, Iraq has become a magnet for Islamic fundamentalists. Quite ironic that the US invasion of Iraq is what has allowed "al Queda" to really get hold in Iraq - Saddam would not have allowed it
Israel has to be the primary reason why one would invade Iraq, given the idealogy of the Bush administration. There was just nothing in the way of any credible security threat presented by Iraq to the USA to justify the invasion. However, Saddam was a threat to Israel, not least because he funded suicide bombers, also because he had bombed Israel in the first Gulf War, and might possibly have tried to do so again if he had ever managed to reacquire the capability to do so.
Also, this Bush administration has well-documented links with the military-industrial complex in the US. War, occupation and rebuilding requires significant spending. And what better favour to do for your old boardroom buddies than by opening the petcocks of the Federal purse and pouring money into Halliburton, et al by way of contracts awarded without tender, never mind the military supply spending? Kind of like how Saddam would divert a lot of the UN Oil-for-Food money into building nice palaces, etc.. for himself and his buddies, except instead of Saddam thieving money that should have been spent on food and medicine, the Bush administration is thieving its own nation's taxpayers money, who get body bags in return (and a skyrocketing deficit).
Anyway...
France had major interests in Iraq. ELF, the state-owned French petro-chemical giant, and Total[1], had major interests in Iraq, see eg:/ 19 97/msg01212.html or just google for "elf iraq". There are pictures of Jacques Chirac with Saddam from the 1970s. Of course, there are more recent pictures of Donald Rumsfeld with Saddam.
http://www.payk.net/mailingLists/iran-news/html
The Mujahideen (NB: there are a variety of english spellings for the arabic word, as with most arabic words.) btw does not equal the taliban. See the wikipedia entry for Mujahideen. It's a general word. In the afghani case, the taliban were but one faction of the collective resistance movement known as the mujahideen. After the war with the Russians, there was civil war between the Taliban and the other factions, the taliban gaining control of most, but not all, of Afghanistan.
As for motives. Let's be honest, every major power which takes an interest in the middle-east does so because of oil. Additionally, the US has a strong political affiliation with Israel, and has long been very involved in assuring Israeli security. The current administration in particular is quite interested in Israel. See Project for a New American Century (PNAC), there are papers there dating to before the present administration gained power making the case for taking out Iraq, reasoned by way of taking out a potential threat of WMD proliferation and stabilising the middle-east and gaining security for Israel. So taking out Iraq is something the the people behind the Bush administration have had as a goal since long before 20010911.
1. Elf, Total and the belgian PetroFina have all since merged together into TotalFinaElf. Total bought Petrofina at some point and then TotalFina merged with Elf in 2000.