Google won't switch to WebM exclusively. This will break many mobile devices for example, like iPhone and even some Android devices. If iPhone users suddenly found out that they cannot play YouTube videos anymore, do you think they will blame the device or the service provider? I'll bet it is the latter.
No. VP8 already has better compression efficiency than h.264 and at the same time being on par in quality with h.264. From the technical point of view, WebM has the potential to be a lot better than H.264.
Uhmm... no. Unless of course you have sources for that claim. As for hardware support, which model from ARM/AMD/Broadcom has support for WebM? For example, even the latest ARM Cortex A-15 that will come out next year doesn't support WebM. When will we find support for WebM from ARM? 2015? 2020?
H.264 official name is MPEG-4 AVC. DivX/XviD/H.263 official name is MPEG-4 ASP. MVC official name is MPEG-4 MVC. In other words, H.264 is part of the MPEG-4 moniker, and is closely related to MPEG-4 ASP and also MPEG2 (used in DVD/Blu-ray etc.).
MJPEG (which the camera uses) is not MPEG-4. That camera doesn't seem to encode to MPEG-4 ASP in any way.
Well, the camera produced MJPEG videos, which makes me to have to transcode to H.264 so that I can play it on my Sony feature-phone (that happened to support H.,264 playback). This would need not happen if this camera actually take videos with MPEG4, right?
Did you know what MJPEG (Google/Bing 'Motion JPEG') is? A video codec too just like MPEG4. The video that camera has is in.mp4 container, MJPEG video and AAC audio. No MPEG4 AVC/ASP/MVC anywhere in it.
I just bought a USD150 digital camera from Sony (you know, that company that has plenty of patents in MPEG-LA) for Christmas that can also record videos in 720p, and it uses MJPEG codec for recording (H.264 isn't available as an option). So, I don't have to pay any royalty fees whatsoever. If Sony wants to maximize profits, why is then this Sony DSC-W350 doesn't record in H.264?
"No, being an OSI certified piece of software means its available. Nothing in the BSD license says you have to put the code on the net or anywhere else."
OSI doesn't certify software, it certify licenses. Using an OSI-certified license like BSD and GPL means that source code availability is guaranteed. Since when OSI becomes like those sham ISO9001-like certifications anyway?
"No, no it's not legally guaranteed, please refer to the BSD license, not the OSI."
No, being OSI-certified, it means that source-code availability is guaranteed. No ifs and buts.
"I didn't say BSD licensed code is closed source, I'm saying that to meet the OSI definition of open source software takes more than the provisions in the BSD license. (Which isn't really important to this debate anyway, as the OSI can't legally enforce anything except the use of their trademark, they don't even have the rights to the term "Open Source")"
To get the OSI certification, the BSD license must comply with all the bullet points in the OSD. You cannot even miss one. Absence of a written guarantee implied compliance, else the list of OSI-approved license will dwindle dramatically. While 'Open source' is not trademarked by OSI, the OSI itself has lots of clouts in the IT world.
"False. I can write a program, BSD license it, give it and the source to someone else, they can make a closed derivative, and I am under no obligation to put the code up anywhere. The OSI may not like it, OSI may say what I did is not open source by their definition, but the BSD license allows that just fine and there's nothing anyone can do about it."
Regents of the University of California can sue you for misusing their license and win, easily. That's why you don't see any BSD-licensed programs out there that doesn't make source-code available. Why don't you try it? Make a good BSD-licensed software, then not make the source code available. The people at OSI will not sue you, but the lawyers at SFLC will be. You better be not living in the US.
"Only if it's available, nobody has a legal obligation to do that. Read the fucking license if you don't believe me. The OSI is not important and have no rights here. Further to that, it's not the same code that is running on your device and is a lot less useful as a result."
It will always be available, the source code. The OSI is important, if they found out that BSD license doesn't follow the OSD to the letter, their certification will be revoked. A revocation will be a huge news to the IT world, it is like Apple going banrupt or something like that.
"You're sputtering and not making much sense again. Everyone gets the same rights as each other with GPL too, they just don't get the right to make closed derivatives."
With BSD, you got the right to get closed source alternatives.
"Didn't say the GPL was right all the time, I said it provided a different set of freedoms to BSD, which it does. A set I prefer for any FOSS work I do. You don't, that's up to you.
The BSD license does not provide the end user with the right to the source code for a program they are given. That is a fact. That is a freedom the BSD license does not provide. You cannot get away from this no matter how hard you squirm and insist that the user can go back to an original version or whatever else. It's not the same thing."
BSD license guarantees source code availability, no matter how you spin it. I dare you to make a program under BSD license, make the binary available to public but not the source code. EIther I will report you to SFLC or I can simply sue you at the courts. You underestimate the importance of OSI (even FSF doesn't dare go against them) and the influence that the OSD has (even Microsoft follows the OSD rules). Only you thinks that OSD is unimportant, while the software giants like Sun (Oracle now) and Microsoft follows its rules.
Only in your world, that BSD license makes source code availability optional.
"Now I don't give a flying fuck which you prefer, and which
1. Being an OSI-certified license, means that source code is going to be available, period. This is proven by the fact that all known BSD-licensed programs out there made their source code freely available. Unless sourceforge is down.
"1. It's not in the license. 2. Still doesn't mean people down the chain can necessarily get at it, and they certainly aren't guaranteed access to derived code. 3. You might want to check out this thread in which it is discussed and people come to the conclusion the the OSI term "source must be available" refers to whether a piece of software is open source, not a given license."
1. Being an OSI-certified license, mean 2. Derived software source code is not guaranteed being available is a built-in feature for BSD license. Access to the original sourced code is guaranteed legally, but not necessarily technically (sourceforge down or something like that) 3. BSD-licensed programs is definitely an open-source software. Only from you I have heard that BSD-licensed programs is a closed-source software which source-code is not freely available. Any programs that used an OSI-approved license (BSD or GPL or MPL or CDDL etc.) must make the source code available.
"Yet end users still don't have the freedom to get the source code for derived versions under the BSD license, a freedom granted under GPL, and a freedom you don't consider worth protecting. I don't consider the freedom to make closed derivatives worth protecting. If you still don't accept that GPL and BSD provide different freedoms for different people, well, I've done my best. You're still wrong."
Yet end users still doesn't have the freedom to modify and release GPL codes in different licenses that they wanted, a freedom you get with BSD license but was taken away by the GPL license. You can get around the problem of derivative programs not making their source codes available by bypassing the derivative program and going straight to the original source (which will always be available, no matter what bullsh*t you are trying to perpetuate). Like that, end users and other developers will also get the same right as the developers of the closed-source derivative program. You cannot do the same with GPL, and that's a fact.
With BSD license, you can have your cake and eat it too. Not so with GPL. The freedom to make closed alternatives is a freedom worth protecting, and the huge number of people using the likes of MPL, CDDL, Apache License etc. shows that I am right. If GPL is right all the time, those liberal licenses will not exist.
"you have a different understanding of the word mandatory to everyone else in the world. The BSD license does no make it mandatory to do this."
The OSI certification made it mandatory for source code to be made available. Else, the certification will be revoked. Read the OSD again at http://opensource.org/docs/osd and burn the document contents on your mind. With BSD license, making the source code available IS MANDATORY!
"Then you're living in cloud cuckoo land. It's that simple."
With many companies and individuals listed in my previous post out there, I dare say I am living in col-hard reality!
"If you owned the copyright then you wouldn't have to give credit. You have a license to use the original content and you have copyright over the derived versions (or parts of it). There are differences. This is the last I will say on this topic as it's clear you have trouble understanding."
You don't get the copyright unless you do attribution. Nice try at skirting my CC-BY example. Why don't you refute my easily understandable CC-BY example if you really think you are right?
"I'm pointing out the difference between the two licenses, that in a BSD situation customer C might not be able to get source to the software running on their device, but under GPL they can. Do you dispute this? Because if you do we may as well give up here."
The reason for your hypothetical situation is technical, not legal. In this Internet day and age, the mandatory availability of the source code by Developer A, who will likely use third-party repositories like sourceforge/github, will ensure that your scenario will not happen.
OSI certification ensures that the source codes for programs that used BS license will be available.
"Not to a lot of people. What don't you understand about this? How long is it going to take to get through your incredibly thick skull to the slow, half-dead brain-matter that's inside it? This is not seen by everyone as a plus, or "more" free than the other way. In fact many people see that as a massive negative."
Not to a lot of people? Tivo, Google, Linksys, ASUS and Microsoft (and plenty of other companies) seems to disagree. In fact, there are many individuals too that did not agree. Including me. Ever wonder why we have many other permissive licenses such as MPL, MS-RL, Apache License, CDDL etc.? Maybe because there are plenty of people out there who simply doesn't subscribe to what FSF and RMS says?
"This has been my whole point since the first post, people disagree about what constitutes freedom, If you want the freedom to deny freedom to others you go for it. People like me will continue grant freedom to end users in what we do, and deny people like you the freedom to make closed versions of our code. There are different freedoms. GPL gives some, BSD gives others."
My point is BSD will give all the freedoms that GPL gives the users and developers, plus even more. BSD is a super-set of GPL. Different freedoms? Hell no!
"Really, it didn't, you might want to look at the legal situation a bit more closely if you think that's the case. There is no transfer of copyright. You simply have a license to use the code under the terms of the BSD license. These terms are very, very permissive as we know, but they are still there and must be complied with. If copyright was actually transferred you wouldn't have to abide by the attribution clause because you would own the copyright. What you have is a license, how many more times does this need to be said?"
How many times I have said that transfer of copyright is possible, like in CC-BY? I can take a CC-BY article, change a font and typeface here and there, and make the authorship as me as long as the credit is given.
"Yes. It would be totally pointless, but you could, easily do so.
More to the point I was actually trying to make - Developer A writes BSD code. Developer B grabs a copy, alters it, embeds it in some device or program. This takes them 2 years. User C buys the device or program. In those two years Developer A got bored and took down their website. Now there is no code at all available to user C, let alone the code that is actually running on their device. With GPL Developer B would be responsible for getting the code to User C. With BSD nobody has to give it to them."
So, your problem is mostly technical instead of legal. Here I thought Developer A actively prevent User C from accessing Developer A's source code. Then there is nothing wrong at all with what Developer A or B has done, under the BSD license. You have to prove malice upon the part of Developer A to even make this fly on self-respecting USA courts.
"Then you should know that even when using the same ARM core, things can be set up and attached to the core in multiple different ways, with multiple different GPIO/MPP/IRQ configurations, onboard memory devices, ethernet controllers, SATA setups, custom button and LED controllers and a multitude of other board specific stuff."
AFAIK, Western Digital doesn't come out with one distinct Linux mini-distro for each hardware config they have out there. Bare-metal coding does have its drawbacks when it comes to maintainability but the company at least have one (actually, a few) base mini-distro they are basing their firmware with. You are over exaggerating the complexities of the firmware, but then again if you want to say that the developers can be retarded sometimes, I don't really have anything to say against that.
"It's a freedom to deny freedom to others. It is a drawback for anybody that you give the derivative program too. They have no rights under BSD."
I don't give the derivative program of my program to anybody. The freedom to deny freedom to others is a plus, unlike what you may have said. If the users of the derivative program doesn't like it, then they can always go to the original program. Even the developers of the derivative program cannot stop them.
"Sorry, can you point to the part in the BSD license that says that I have to make any code I release under BSD available forever and to anyone that asks?
I don't remember that section.
The BSD does nothing to guarantee that someone who receives a binary can get the source, any source. That is the fact of the matter. It does nothing to guarantee anything and that's its strength to you, it's also its weakness to me."
The OSI certification for BSD license (3-clause) stipulates that it must follow the Open Source Definition guidelines. One of the guidelines mandates that source code must always be made available. This alone makes the users of BSD license programs to make available the relevant source codes for their program.
Well, I can be wrong if I somehow miss the news that BSD OSI certification has been revoked. Can you point me to such a source? The OSD is available at http://opensource.org/docs/osd and it applies to BSD license too.
"The GPL puts the responsibility for the source onto t
"Because that's not how it works, the author is not assigning the copyright to you. If they were there would be no legal need for attribution as you would be the copyright owner."
The transfer of copyright happened when the term of the license is complied to (credit). It is comparable to how CC-BY works.
"They have no right to get the source code for the binary they have been given. The original author has no obligation to to provide anything either. The BSD does give the user the right to the source of the code they are running."
So I can actually make a BSD-licensed software but yet not make the source code available? Somehow I feel that argument of yours will not fly in any self-respecting courts in the US.
"Which may or may not be available anywhere and doesn't have any of the stuff in it that make it work on that model, so good luck with that."
No, Western Digital will have to tell their users where on earth they swiped the source code from.
"Except that the device specific initialisation is going to be missing, and it's going to cost you months instead of days to get any useful results."
Considering that WD use off-the-shelf parts for their NAS, I will beg to differ. Disclaimer: I work for Western Digital until the end of last year.
"The GPL doesn't tie you to the policies of RMS and the FSF either."
I raise you GPLv3/APGL. Tivo and Google has done nothing wrong with their usage of Linux, yet FSF and RMS sees it fits to create GPLv3 and AGPL. Thank God Mr. Torvalds doesn't bite. GPLv3 and AGPL reflects the militant stance of FSF and RMS.
"Not of the derivative program I'm running and not really at all, nobody is obliged to provide any source under BSD."
The option by derivative programs to not offer source code is a freedom given to you by the BSD license, not a drawback. That's why BSD is superior than GPL. It is called flexibility. And by using BSD license, source code availability is pretty much guaranteed. Can you cite any famous BSD-licensed project that doesn't make their source code available?
Do you realize that BSD is an OSI-approved license? I wonder if OSI will approve the license if source code is not guaranteed.
"The produced binary, yes. The source, no. Their alterations are under their copyright, the original bits are licensed under the terms of the BSD license. Yes they could, easily. WHAT they could win is not a lot, probably only attribution, but they could still win it because they copyright is not given to the new developer, only licensed. You are confused about copyright law. If you had complete copyright control of the new thing then you wouldn't have to attribute it to anyone as you would own it completely. You are using the code under the terms of the BSD license and must attribute as you yourself admit."
"You really don't. Nobody is assigning you their copyrights by using the BSD license, they are just giving you a very permissive license."
Considering that credit has already being given, why not? BSD is little different than CC-BY anyway. You don't even have to change a single line/bit of the source code material to make it yours, attribution (or credits as I call it) is all is needed. Ever wonder why the CC Board recommends the plain 3-clause BSD license as licenses for software (alongside CC GPL and CC LGPL)? Emphasize on the CC *GPL thing!
"Unless someone has given you the binary, then you don't. If someone has given you the binary, then you should get access to the source because that's what the original developer wanted, because the GPL has given you the freedom that the BSD didn't, to have the source of the program you are running."
Whoa, I have no idea that BSD license prevent the recipient from getting the source code. Since when did this happen?
"Which is useless because it won't run on $Hardware that the extended version came on."
GPLv2 (or BSD etc) doesn't mandate that access to source code must be complemented too with the right to run the compiled code on certain hardware. Even the likes of Linus sees the stupidity of GPLv3/AGPL and vowed to not use the license for Linux. BTW, my example of YAMP 2010 GPL will run on the same platform that the original BSD AMP 2010 will too.
Isn't this more proof why GPL has become even more abhorrent to many individual and companies as timer goes on?
"This is rather the point, that each has different freedoms. If you want users to be able to get the source for a binary they are given, then use GPL. If you want developers to be able to distribute binaries without source further down the line, then use BSD."
With BSD, recipients of the binaries can also get the source code. Since when this isn't so? When I make my program available with BSD license, users will get binaries and source code. I don't control (or want to, unlike some GPL zealots out there) the licenses of what the programmers of the derivatives of my BSD program want to use. If the derivative is closed-source, users can always go to the original source and have their way with it.
BSD allows users & developers to get binaries and source codes of my BSD program. Maybe the derivatives doesn't allow access to their source code, but with proper credit, the users will know where the derivative comes from, and goes straight to the source. This is valid 99% of the time.
"Me, as a user of devices that have come with linux on them, and being interested in customising them, I'm very glad that the likes of western digital are required to release the source for the software on the NAS I bought, because I have the freedom to look at what they did and change it without having to completely reverse engineer the platform support. As a user I now have freedom and control over my device which the BSD license would not have allowed me. And given the terrible state of the source tarball I did get from WD I have no doubt that if they had used BSD software they just wouldn't have released it."
Yet, just in case if Western Digital uses a BSD OS, and did not release the source code, I as a developer and a user, can go straight to the source that Western Digital also uses for their NAS, and get the source code there. And with the terrible job Western Dig
They own the copyright to their changes, as in the previous situation, but they are using the rest under the license granted by the copyright holder, the BSD license."
Yet the YAMP 2010's copyright belongs to the derivative developers, not the owner of the original program. The original developer who makes the BSD-licensed original program cannot go to the courts, claiming ownership to the GPL (or closed source) new program (with proper credit) and win. If I take a BSD-licensed application, change the bare minimum so that trademarked portions of the program was changed/removed, and make it available as mine (under GPL, with credit), I for sure can claim copyright for it.
You need to have the ability to switch copyright ownership to be able to make BSD-licensed code be available under different license(s).
"And that's fine, if you don't mind not having access to stuff that used your code as a base, and if you don't mind users not having access to the source code for the programs they are running. The GPL provides those users with the rights to the source for YAMP 2010, with the freedom to change and redistribute the program. If the people who relicensed your code chose a closed license then their end users would not have these rights."
If the derivative program is not mine, why on earth should I get lawful access to the source code of the derivative program? Developers and users can still get access to my BSD-licensed source code. GPL may give users access to the source code of the YAMP 2010 (a freedom attained), but it prevents developers from using them for whatever purpose and license they want to use (another freedom lost). And if the derivative program developers used a closed source license for YAMP 2010, the end users and developers can still have the source code of AMP 2010, with complete freedom to do what they please with it.
Of course, the copyright of the ORIGINAL source code still belong to the ORIGINAL developers, under the BSD license. But the copyright of source code in the DERIVATIVE program, belongs to the DERIVATIVE programmer, whatever license the derivative programmer used. That's why anyone can lift a BSD-licensed source code and make it closed-source, even if there is no modification done at all to the original source code used in the closed-source version (credits should be supplied though). If what you say is true, a BSD-licensed code cannot be relicensed to another license (GPL or closed-source.)
And your argument about the original developers not wanting to give the source code of a new version of their program to the the developers of the derivative program is simply bollocks. That will only happen if the original developers relicense their new program version to close-source license or a restrictive license like GPL. When that happened, the derivative application programmers will not be able to access the new closed-source version (understandable) or unable to because the new GPL license is incompatible their the license used by the derivative application programmers (unless of course if the derivative program used GPL as shown in my example above). If the new version is available under the BSD license like the older version does, then the derivative application developers can lift whatever they want from the original application's new version and adapt it into their program.
If you have a GPL software (let's name it 'Another Compression Software 2010 (TM)'), and someone else used its source code as a base, make some modifications in a few places like the GUI etc., and then released it as 'Yet Another Compression Software 2010 (TM)' under the GPL, did that person own the whole copyright for the all the source code in the derivative software, or did that person only own the modification he/she has made? I believe the latter is correct, and that's why many people and companies out there avoids GPL software completely, and makes BSD-licensed source codes more appealing.
If I have a BSD software (let's name it 'Another Music Player 2010 (TM)'), and someone else used its source code as a base, made some modifications in a few places like the GUI etc., and then released it as 'Yet Another Music Player 2010 (TM)', under the GPL (yes, you read it right), that person who make the derivative software will own the copyright to every single line of the newly-made GPL-ed software, and all he/she has to do is to give credit (as stipulated by the BSD license) and not violate my trademarks. Oh BTW, my own source code used as the base will still exist and available under the BSD for anyone else to use, modify and release on their own terms. My right will be preserved, and I don't have to whine for access to source codes of any other software which copyright I don't own (Yet Another Music Player 2010).
This is why BSD is always better than GPL any day.
You don't own the copyrights of a BSD-derived software. Why on earth you will to claim copyrights for a software you doesn't even own? Now wonder people hate the GPL license, it tries to give the original copyright holder access to any derived software from the original GPLed source.
The thing about you Firefox supporters is that you aren't heavy web users. You think that a few hours of casual browsing is enough to make a comparison on.
Citation needed. Where the hell you get the idea that Firefox users are not heavy web users? If you want to say the same thing about IE maybe I can accept it.
Actually symbolset implies two things 1. W7 growth is attributed primarily by the defection of Vista users, and 2. Windows XP users doesn't adopt W7. From my observation, symbolset is wrong in both counts, and your link at least shows symbolset being wrong on claim no.2.
When you're finding in the charts the information you want to find regardless of the later outcome, you might as well be looking at Tarot cards or bird entrails. It's clear you and I are not going to agree on how to project the uptake curve of W7 against XP. I see W7 at 15 to 20% at the end of July, nearly a year after RTM, and having gotten nearly all of that from the much reviled and structurally similar Windows Vista. The plateau is plain as day. Though the Vista base continues to erode, adoption by XP users is levelling off and it never was much.
With respect to quality, it's better than either Theora and Dirac, and it's also better than H.264 Baseline. If I understand correctly, the latter is largely what is used on the Net today (including YouTube), and enjoys most hardware decoding support.
Only SD YouTube videos are encodec with Baseline profile. All HD videos on YouTube are encoded using Main or High Profile.
Slashdot Mirror
Google won't switch to WebM exclusively. This will break many mobile devices for example, like iPhone and even some Android devices. If iPhone users suddenly found out that they cannot play YouTube videos anymore, do you think they will blame the device or the service provider? I'll bet it is the latter.
No. VP8 already has better compression efficiency than h.264 and at the same time being on par in quality with h.264. From the technical point of view, WebM has the potential to be a lot better than H.264.
Uhmm... no. Unless of course you have sources for that claim. As for hardware support, which model from ARM/AMD/Broadcom has support for WebM? For example, even the latest ARM Cortex A-15 that will come out next year doesn't support WebM. When will we find support for WebM from ARM? 2015? 2020?
Citation needed. Which OECD country has progressive tax system, and has a middle class and a vibrant economy? Scandinavian countries? Hahahaha!
Working with text documents and spreadsheets are now niche activities? Wow, what world did you live in?
H.264 official name is MPEG-4 AVC. DivX/XviD/H.263 official name is MPEG-4 ASP. MVC official name is MPEG-4 MVC. In other words, H.264 is part of the MPEG-4 moniker, and is closely related to MPEG-4 ASP and also MPEG2 (used in DVD/Blu-ray etc.).
MJPEG (which the camera uses) is not MPEG-4. That camera doesn't seem to encode to MPEG-4 ASP in any way.
Well, the camera produced MJPEG videos, which makes me to have to transcode to H.264 so that I can play it on my Sony feature-phone (that happened to support H.,264 playback). This would need not happen if this camera actually take videos with MPEG4, right?
Did you know what MJPEG (Google/Bing 'Motion JPEG') is? A video codec too just like MPEG4. The video that camera has is in .mp4 container, MJPEG video and AAC audio. No MPEG4 AVC/ASP/MVC anywhere in it.
MP4 is the container, but the video in it uses MJPEG.
I just bought a USD150 digital camera from Sony (you know, that company that has plenty of patents in MPEG-LA) for Christmas that can also record videos in 720p, and it uses MJPEG codec for recording (H.264 isn't available as an option). So, I don't have to pay any royalty fees whatsoever. If Sony wants to maximize profits, why is then this Sony DSC-W350 doesn't record in H.264?
"No, being an OSI certified piece of software means its available. Nothing in the BSD license says you have to put the code on the net or anywhere else."
OSI doesn't certify software, it certify licenses. Using an OSI-certified license like BSD and GPL means that source code availability is guaranteed. Since when OSI becomes like those sham ISO9001-like certifications anyway?
"No, no it's not legally guaranteed, please refer to the BSD license, not the OSI."
No, being OSI-certified, it means that source-code availability is guaranteed. No ifs and buts.
"I didn't say BSD licensed code is closed source, I'm saying that to meet the OSI definition of open source software takes more than the provisions in the BSD license. (Which isn't really important to this debate anyway, as the OSI can't legally enforce anything except the use of their trademark, they don't even have the rights to the term "Open Source")"
To get the OSI certification, the BSD license must comply with all the bullet points in the OSD. You cannot even miss one. Absence of a written guarantee implied compliance, else the list of OSI-approved license will dwindle dramatically. While 'Open source' is not trademarked by OSI, the OSI itself has lots of clouts in the IT world.
"False. I can write a program, BSD license it, give it and the source to someone else, they can make a closed derivative, and I am under no obligation to put the code up anywhere. The OSI may not like it, OSI may say what I did is not open source by their definition, but the BSD license allows that just fine and there's nothing anyone can do about it."
Regents of the University of California can sue you for misusing their license and win, easily. That's why you don't see any BSD-licensed programs out there that doesn't make source-code available. Why don't you try it? Make a good BSD-licensed software, then not make the source code available. The people at OSI will not sue you, but the lawyers at SFLC will be. You better be not living in the US.
"Only if it's available, nobody has a legal obligation to do that. Read the fucking license if you don't believe me. The OSI is not important and have no rights here. Further to that, it's not the same code that is running on your device and is a lot less useful as a result."
It will always be available, the source code. The OSI is important, if they found out that BSD license doesn't follow the OSD to the letter, their certification will be revoked. A revocation will be a huge news to the IT world, it is like Apple going banrupt or something like that.
"You're sputtering and not making much sense again. Everyone gets the same rights as each other with GPL too, they just don't get the right to make closed derivatives."
With BSD, you got the right to get closed source alternatives.
"Didn't say the GPL was right all the time, I said it provided a different set of freedoms to BSD, which it does. A set I prefer for any FOSS work I do. You don't, that's up to you.
The BSD license does not provide the end user with the right to the source code for a program they are given. That is a fact. That is a freedom the BSD license does not provide. You cannot get away from this no matter how hard you squirm and insist that the user can go back to an original version or whatever else. It's not the same thing."
BSD license guarantees source code availability, no matter how you spin it. I dare you to make a program under BSD license, make the binary available to public but not the source code. EIther I will report you to SFLC or I can simply sue you at the courts. You underestimate the importance of OSI (even FSF doesn't dare go against them) and the influence that the OSD has (even Microsoft follows the OSD rules). Only you thinks that OSD is unimportant, while the software giants like Sun (Oracle now) and Microsoft follows its rules.
Only in your world, that BSD license makes source code availability optional.
"Now I don't give a flying fuck which you prefer, and which
"1. Being an OSI-certified license, mean"
completing the sentence above, it reads
1. Being an OSI-certified license, means that source code is going to be available, period. This is proven by the fact that all known BSD-licensed programs out there made their source code freely available. Unless sourceforge is down.
"1. It's not in the license.
2. Still doesn't mean people down the chain can necessarily get at it, and they certainly aren't guaranteed access to derived code.
3. You might want to check out this thread in which it is discussed and people come to the conclusion the the OSI term "source must be available" refers to whether a piece of software is open source, not a given license."
1. Being an OSI-certified license, mean
2. Derived software source code is not guaranteed being available is a built-in feature for BSD license. Access to the original sourced code is guaranteed legally, but not necessarily technically (sourceforge down or something like that)
3. BSD-licensed programs is definitely an open-source software. Only from you I have heard that BSD-licensed programs is a closed-source software which source-code is not freely available. Any programs that used an OSI-approved license (BSD or GPL or MPL or CDDL etc.) must make the source code available.
"Yet end users still don't have the freedom to get the source code for derived versions under the BSD license, a freedom granted under GPL, and a freedom you don't consider worth protecting. I don't consider the freedom to make closed derivatives worth protecting.
If you still don't accept that GPL and BSD provide different freedoms for different people, well, I've done my best. You're still wrong."
Yet end users still doesn't have the freedom to modify and release GPL codes in different licenses that they wanted, a freedom you get with BSD license but was taken away by the GPL license. You can get around the problem of derivative programs not making their source codes available by bypassing the derivative program and going straight to the original source (which will always be available, no matter what bullsh*t you are trying to perpetuate). Like that, end users and other developers will also get the same right as the developers of the closed-source derivative program. You cannot do the same with GPL, and that's a fact.
With BSD license, you can have your cake and eat it too. Not so with GPL. The freedom to make closed alternatives is a freedom worth protecting, and the huge number of people using the likes of MPL, CDDL, Apache License etc. shows that I am right. If GPL is right all the time, those liberal licenses will not exist.
"you have a different understanding of the word mandatory to everyone else in the world. The BSD license does no make it mandatory to do this."
The OSI certification made it mandatory for source code to be made available. Else, the certification will be revoked. Read the OSD again at http://opensource.org/docs/osd and burn the document contents on your mind. With BSD license, making the source code available IS MANDATORY!
"Then you're living in cloud cuckoo land. It's that simple."
With many companies and individuals listed in my previous post out there, I dare say I am living in col-hard reality!
"If you owned the copyright then you wouldn't have to give credit.
You have a license to use the original content and you have copyright over the derived versions (or parts of it). There are differences.
This is the last I will say on this topic as it's clear you have trouble understanding."
You don't get the copyright unless you do attribution. Nice try at skirting my CC-BY example. Why don't you refute my easily understandable CC-BY example if you really think you are right?
"I'm pointing out the difference between the two licenses, that in a BSD situation customer C might not be able to get source to the software running on their device, but under GPL they can. Do you dispute this? Because if you do we may as well give up here."
The reason for your hypothetical situation is technical, not legal. In this Internet day and age, the mandatory availability of the source code by Developer A, who will likely use third-party repositories like sourceforge/github, will ensure that your scenario will not happen.
OSI certification ensures that the source codes for programs that used BS license will be available.
"Not to a lot of people. What don't you understand about this? How long is it going to take to get through your incredibly thick skull to the slow, half-dead brain-matter that's inside it? This is not seen by everyone as a plus, or "more" free than the other way. In fact many people see that as a massive negative."
Not to a lot of people? Tivo, Google, Linksys, ASUS and Microsoft (and plenty of other companies) seems to disagree. In fact, there are many individuals too that did not agree. Including me. Ever wonder why we have many other permissive licenses such as MPL, MS-RL, Apache License, CDDL etc.? Maybe because there are plenty of people out there who simply doesn't subscribe to what FSF and RMS says?
"This has been my whole point since the first post, people disagree about what constitutes freedom, If you want the freedom to deny freedom to others you go for it. People like me will continue grant freedom to end users in what we do, and deny people like you the freedom to make closed versions of our code. There are different freedoms. GPL gives some, BSD gives others."
My point is BSD will give all the freedoms that GPL gives the users and developers, plus even more. BSD is a super-set of GPL. Different freedoms? Hell no!
"Really, it didn't, you might want to look at the legal situation a bit more closely if you think that's the case. There is no transfer of copyright. You simply have a license to use the code under the terms of the BSD license. These terms are very, very permissive as we know, but they are still there and must be complied with. If copyright was actually transferred you wouldn't have to abide by the attribution clause because you would own the copyright. What you have is a license, how many more times does this need to be said?"
How many times I have said that transfer of copyright is possible, like in CC-BY? I can take a CC-BY article, change a font and typeface here and there, and make the authorship as me as long as the credit is given.
"Yes. It would be totally pointless, but you could, easily do so.
More to the point I was actually trying to make - Developer A writes BSD code. Developer B grabs a copy, alters it, embeds it in some device or program. This takes them 2 years. User C buys the device or program. In those two years Developer A got bored and took down their website. Now there is no code at all available to user C, let alone the code that is actually running on their device. With GPL Developer B would be responsible for getting the code to User C. With BSD nobody has to give it to them."
So, your problem is mostly technical instead of legal. Here I thought Developer A actively prevent User C from accessing Developer A's source code. Then there is nothing wrong at all with what Developer A or B has done, under the BSD license. You have to prove malice upon the part of Developer A to even make this fly on self-respecting USA courts.
"Then you should know that even when using the same ARM core, things can be set up and attached to the core in multiple different ways, with multiple different GPIO/MPP/IRQ configurations, onboard memory devices, ethernet controllers, SATA setups, custom button and LED controllers and a multitude of other board specific stuff."
AFAIK, Western Digital doesn't come out with one distinct Linux mini-distro for each hardware config they have out there. Bare-metal coding does have its drawbacks when it comes to maintainability but the company at least have one (actually, a few) base mini-distro they are basing their firmware with. You are over exaggerating the complexities of the firmware, but then again if you want to say that the developers can be retarded sometimes, I don't really have anything to say against that.
"It's a freedom to deny freedom to others. It is a drawback for anybody that you give the derivative program too. They have no rights under BSD."
I don't give the derivative program of my program to anybody. The freedom to deny freedom to others is a plus, unlike what you may have said. If the users of the derivative program doesn't like it, then they can always go to the original program. Even the developers of the derivative program cannot stop them.
"Sorry, can you point to the part in the BSD license that says that I have to make any code I release under BSD available forever and to anyone that asks?
I don't remember that section.
The BSD does nothing to guarantee that someone who receives a binary can get the source, any source.
That is the fact of the matter.
It does nothing to guarantee anything and that's its strength to you, it's also its weakness to me."
The OSI certification for BSD license (3-clause) stipulates that it must follow the Open Source Definition guidelines. One of the guidelines mandates that source code must always be made available. This alone makes the users of BSD license programs to make available the relevant source codes for their program.
Well, I can be wrong if I somehow miss the news that BSD OSI certification has been revoked. Can you point me to such a source? The OSD is available at http://opensource.org/docs/osd and it applies to BSD license too.
"The GPL puts the responsibility for the source onto t
"Because that's not how it works, the author is not assigning the copyright to you. If they were there would be no legal need for attribution as you would be the copyright owner."
The transfer of copyright happened when the term of the license is complied to (credit). It is comparable to how CC-BY works.
"They have no right to get the source code for the binary they have been given. The original author has no obligation to to provide anything either. The BSD does give the user the right to the source of the code they are running."
So I can actually make a BSD-licensed software but yet not make the source code available? Somehow I feel that argument of yours will not fly in any self-respecting courts in the US.
"Which may or may not be available anywhere and doesn't have any of the stuff in it that make it work on that model, so good luck with that."
No, Western Digital will have to tell their users where on earth they swiped the source code from.
"Except that the device specific initialisation is going to be missing, and it's going to cost you months instead of days to get any useful results."
Considering that WD use off-the-shelf parts for their NAS, I will beg to differ. Disclaimer: I work for Western Digital until the end of last year.
"The GPL doesn't tie you to the policies of RMS and the FSF either."
I raise you GPLv3/APGL. Tivo and Google has done nothing wrong with their usage of Linux, yet FSF and RMS sees it fits to create GPLv3 and AGPL. Thank God Mr. Torvalds doesn't bite. GPLv3 and AGPL reflects the militant stance of FSF and RMS.
"Not of the derivative program I'm running and not really at all, nobody is obliged to provide any source under BSD."
The option by derivative programs to not offer source code is a freedom given to you by the BSD license, not a drawback. That's why BSD is superior than GPL. It is called flexibility.
And by using BSD license, source code availability is pretty much guaranteed. Can you cite any famous BSD-licensed project that doesn't make their source code available?
Do you realize that BSD is an OSI-approved license? I wonder if OSI will approve the license if source code is not guaranteed.
"The produced binary, yes. The source, no. Their alterations are under their copyright, the original bits are licensed under the terms of the BSD license.
Yes they could, easily. WHAT they could win is not a lot, probably only attribution, but they could still win it because they copyright is not given to the new developer, only licensed. You are confused about copyright law. If you had complete copyright control of the new thing then you wouldn't have to attribute it to anyone as you would own it completely. You are using the code under the terms of the BSD license and must attribute as you yourself admit."
"You really don't. Nobody is assigning you their copyrights by using the BSD license, they are just giving you a very permissive license."
Considering that credit has already being given, why not? BSD is little different than CC-BY anyway. You don't even have to change a single line/bit of the source code material to make it yours, attribution (or credits as I call it) is all is needed. Ever wonder why the CC Board recommends the plain 3-clause BSD license as licenses for software (alongside CC GPL and CC LGPL)? Emphasize on the CC *GPL thing!
"Unless someone has given you the binary, then you don't. If someone has given you the binary, then you should get access to the source because that's what the original developer wanted, because the GPL has given you the freedom that the BSD didn't, to have the source of the program you are running."
Whoa, I have no idea that BSD license prevent the recipient from getting the source code. Since when did this happen?
"Which is useless because it won't run on $Hardware that the extended version came on."
GPLv2 (or BSD etc) doesn't mandate that access to source code must be complemented too with the right to run the compiled code on certain hardware. Even the likes of Linus sees the stupidity of GPLv3/AGPL and vowed to not use the license for Linux. BTW, my example of YAMP 2010 GPL will run on the same platform that the original BSD AMP 2010 will too.
Isn't this more proof why GPL has become even more abhorrent to many individual and companies as timer goes on?
"This is rather the point, that each has different freedoms. If you want users to be able to get the source for a binary they are given, then use GPL. If you want developers to be able to distribute binaries without source further down the line, then use BSD."
With BSD, recipients of the binaries can also get the source code. Since when this isn't so? When I make my program available with BSD license, users will get binaries and source code. I don't control (or want to, unlike some GPL zealots out there) the licenses of what the programmers of the derivatives of my BSD program want to use. If the derivative is closed-source, users can always go to the original source and have their way with it.
BSD allows users & developers to get binaries and source codes of my BSD program. Maybe the derivatives doesn't allow access to their source code, but with proper credit, the users will know where the derivative comes from, and goes straight to the source. This is valid 99% of the time.
"Me, as a user of devices that have come with linux on them, and being interested in customising them, I'm very glad that the likes of western digital are required to release the source for the software on the NAS I bought, because I have the freedom to look at what they did and change it without having to completely reverse engineer the platform support. As a user I now have freedom and control over my device which the BSD license would not have allowed me. And given the terrible state of the source tarball I did get from WD I have no doubt that if they had used BSD software they just wouldn't have released it."
Yet, just in case if Western Digital uses a BSD OS, and did not release the source code, I as a developer and a user, can go straight to the source that Western Digital also uses for their NAS, and get the source code there. And with the terrible job Western Dig
"False.
They own the copyright to their changes, as in the previous situation, but they are using the rest under the license granted by the copyright holder, the BSD license."
Yet the YAMP 2010's copyright belongs to the derivative developers, not the owner of the original program. The original developer who makes the BSD-licensed original program cannot go to the courts, claiming ownership to the GPL (or closed source) new program (with proper credit) and win. If I take a BSD-licensed application, change the bare minimum so that trademarked portions of the program was changed/removed, and make it available as mine (under GPL, with credit), I for sure can claim copyright for it.
You need to have the ability to switch copyright ownership to be able to make BSD-licensed code be available under different license(s).
"And that's fine, if you don't mind not having access to stuff that used your code as a base, and if you don't mind users not having access to the source code for the programs they are running. The GPL provides those users with the rights to the source for YAMP 2010, with the freedom to change and redistribute the program. If the people who relicensed your code chose a closed license then their end users would not have these rights."
If the derivative program is not mine, why on earth should I get lawful access to the source code of the derivative program? Developers and users can still get access to my BSD-licensed source code. GPL may give users access to the source code of the YAMP 2010 (a freedom attained), but it prevents developers from using them for whatever purpose and license they want to use (another freedom lost). And if the derivative program developers used a closed source license for YAMP 2010, the end users and developers can still have the source code of AMP 2010, with complete freedom to do what they please with it.
Of course, the copyright of the ORIGINAL source code still belong to the ORIGINAL developers, under the BSD license. But the copyright of source code in the DERIVATIVE program, belongs to the DERIVATIVE programmer, whatever license the derivative programmer used. That's why anyone can lift a BSD-licensed source code and make it closed-source, even if there is no modification done at all to the original source code used in the closed-source version (credits should be supplied though). If what you say is true, a BSD-licensed code cannot be relicensed to another license (GPL or closed-source.)
And your argument about the original developers not wanting to give the source code of a new version of their program to the the developers of the derivative program is simply bollocks. That will only happen if the original developers relicense their new program version to close-source license or a restrictive license like GPL. When that happened, the derivative application programmers will not be able to access the new closed-source version (understandable) or unable to because the new GPL license is incompatible their the license used by the derivative application programmers (unless of course if the derivative program used GPL as shown in my example above). If the new version is available under the BSD license like the older version does, then the derivative application developers can lift whatever they want from the original application's new version and adapt it into their program.
If you have a GPL software (let's name it 'Another Compression Software 2010 (TM)'), and someone else used its source code as a base, make some modifications in a few places like the GUI etc., and then released it as 'Yet Another Compression Software 2010 (TM)' under the GPL, did that person own the whole copyright for the all the source code in the derivative software, or did that person only own the modification he/she has made? I believe the latter is correct, and that's why many people and companies out there avoids GPL software completely, and makes BSD-licensed source codes more appealing.
If I have a BSD software (let's name it 'Another Music Player 2010 (TM)'), and someone else used its source code as a base, made some modifications in a few places like the GUI etc., and then released it as 'Yet Another Music Player 2010 (TM)', under the GPL (yes, you read it right), that person who make the derivative software will own the copyright to every single line of the newly-made GPL-ed software, and all he/she has to do is to give credit (as stipulated by the BSD license) and not violate my trademarks. Oh BTW, my own source code used as the base will still exist and available under the BSD for anyone else to use, modify and release on their own terms. My right will be preserved, and I don't have to whine for access to source codes of any other software which copyright I don't own (Yet Another Music Player 2010).
This is why BSD is always better than GPL any day.
You don't own the copyrights of a BSD-derived software. Why on earth you will to claim copyrights for a software you doesn't even own? Now wonder people hate the GPL license, it tries to give the original copyright holder access to any derived software from the original GPLed source.
The thing about you Firefox supporters is that you aren't heavy web users. You think that a few hours of casual browsing is enough to make a comparison on.
Citation needed. Where the hell you get the idea that Firefox users are not heavy web users? If you want to say the same thing about IE maybe I can accept it.
Actually symbolset implies two things 1. W7 growth is attributed primarily by the defection of Vista users, and 2. Windows XP users doesn't adopt W7. From my observation, symbolset is wrong in both counts, and your link at least shows symbolset being wrong on claim no.2.
Citation needed. What is your source?
With respect to quality, it's better than either Theora and Dirac, and it's also better than H.264 Baseline. If I understand correctly, the latter is largely what is used on the Net today (including YouTube), and enjoys most hardware decoding support.
Only SD YouTube videos are encodec with Baseline profile. All HD videos on YouTube are encoded using Main or High Profile.