Read the actual license, not the FAQ (which pretty clearly contradicts the license and tries to make it look better than it is), AC. Your point about downloading being free is a strawman anyway, the problem is distributing it as part of a distribution... After reading the license, if you really can't see what the problems are, read http://lists.debian.org/debian-devel/2006/05/thrd2 .html#00752.
One example:
(c) you do not combine, configure or distribute the Software to
run in conjunction with any additional software that implements
the same or similar functionality or APIs as the Software;
That sounds like distributing gcj or a java library implementation on the same OS is not allowed, although "in conjunction" and "similar functionality" might mean something totally different in lawyer-lingo...
But again, what Google is doing is very common practice in Windows world (and getting quite popular in Linux world). It is by far the best way to keep software updated, and push (actually, really pull) updates and fix vulnerabilities as fast as possible.
It's better than doing nothing, I'll give you that. But do you seriously mean that having hundreds of applications all phone home is somehow better than proper package management? I disagree.
Just for the record, I don't think Linus's interpretation is correct -- he does seem to have a double standard here, just as you imply. I just presented it for discussion.
If you literally cut and paste code from Linux, or code from Linux gets included during the process of linking, you could certainly show derivation and substantial similarity, but where you cannot, i'm afraid you can't use this argument.
You are oversimplifying again. Cut'n'paste is not probably needed, just 'similarity' of execution as you said.
Surely all manufacturers whose peripherals Linux supports would have standing to sue the OSS driver developers because the OSS drivers work in a substantially similar way?
If the OSS developers had access to the original source, and used it as a basis (even without straight copying), this is entirely possible.
He decides he doesn't have time for this so he slicks the drive and re-installs Windows, then goes and makes love to his wife after apologizing about all the cables and how he is spending too much time in front of the computer.
Correction:...re-installs Windows and, 22 reboots later, goes and makes love to his wife...
Screwed up the layout there, this is what it was supposed to look like:
"Derived works" is actually a complicated and not very clearly defined set... You might be right about a binary driver being completely separate work, but it certainly isn't as clear cut as you imply. This is what Linus had to say about this:
Basically: - anything that was written with Linux in mind (whether it then _also_ works on other operating systems or not) is clearly partially a derived work. - anything that has knowledge of and plays with fundamental internal Linux behaviour is clearly a derived work. If you need to muck around with core code, you're derived, no question about it.
...cut a filesystem example...
Does that mean that any kernel module is automatically not a derived work? HELL NO! It has nothing to do with modules per se, except that non-modules clearly are derived works (if they are so central to the kenrel that you can't load them as a module, they are clearly derived works just by virtue of being very intimate - and because the GPL expressly mentions linking).
So being a module is not a sign of not being a derived work. It's just one sign that _maybe_ it might have other arguments for why it isn't derived.
As you can see he views "derived" as including more than just stuff that includes kernel headers. In some ways that sounds reasonable, as the copyright of the kernel developers extends to the way linux works, not just the actual lines of code -- just like a story line of a book may be under copyright, depending on the case.
"Derived works" is actually a complicated and not very clearly defined set... You might be right about a binary driver being completely separate work, but it certainly isn't as clear cut as you imply.
This is what Linus had to say about this:
Basically:
- anything that was written with Linux in mind (whether it then _also_
works on other operating systems or not) is clearly partially a derived
work.
- anything that has knowledge of and plays with fundamental internal
Linux behaviour is clearly a derived work. If you need to muck around
with core code, you're derived, no question about it.
...cut a filesystem example...
Does that mean that any kernel module is automatically not a derived work?
HELL NO! It has nothing to do with modules per se, except that non-modules
clearly are derived works (if they are so central to the kenrel that you
can't load them as a module, they are clearly derived works just by virtue
of being very intimate - and because the GPL expressly mentions linking).
So being a module is not a sign of not being a derived work. It's just
one sign that _maybe_ it might have other arguments for why it isn't
derived.
As you can see he views "derived" as including more than just stuff that includes kernel headers. In some ways that sounds reasonable, as the copyright of the kernel developers extends to the way linux works, not just the actual lines of code -- just like a story line of a book may be under copyright, depending on the case.
Just a note to TBE users: Anyone using Tabbrowser extensions should stay silent when the next Firefox-is-a-memory-hog flame fest ensues -- it has been one of the worst culprits in that area...
a) It's hard to permanently render useless a solution that is entirely software-based (Nokia can always give out an update).
b) The software is just a client that supports Telepathy -- if Google stops their servers there will be others. My guess is that the client has nothing to do with Google (except using the same protocol): I believe this co-operation is mostly a PR effort.
c) it's not a phone. It's an Internet Tablet (with VOIP in the future).
By your logic Nokia should still make all their products out of rubber -- they used to be a rubber manufacturer after all.
More seriously, Nokia doesn't want to end up as a puppet for the telcos, and moving to products that work over IP is one of the ways they can achieve that.
it means Google's GTalk client will be ported to Linux
Not necessarily. The maemo roadmap has said for some time that the next release will have Telepathy support -- which means a GoogleTalk client. That's not necessarily GTalk...
Not getting into specifics (I don't know the cases well enough), but you seem to think that the rules are the same for everyone -- they aren't. Things that are legal for everyone else, may be illegal for monopolists. The system is built like this not to prevent monopolies, but to counter-balance the power a monopoly has on those who wish to compete with it...
To sum up: Yes, Microsoft can be sued and convicted for something all their competitors are doing.
Gnome will continue to be pushed as the default desktop of choice for every "Linux on the desktop" that fails (pretty standard in that area) while those of us who use Linux for choice and aren't afraid of tabs, address bar, etc will use KDE or Fluxbox.
Speaking for others (especially with language like that) is pretty arrogant. I'm pretty much a power user and I really prefer Gnome. I can understand your POV, but I'd appreciate it if you didn't twist the truth -- some people do like to work in Gnome, and that is one of the reasons it is the default choice of several desktop-distributions (failing and flourishing ones alike).
Well, my logic goes like this:
1. Computing and computers are extremely important to our society (both culturally and efficiency-wise)
2. The single most important threat to the "growth" that we're getting from computers is, in my opinion, wrong choices regarding the basic platform (file formats, interfaces, other de-facto standards). We've seen time and time again that once formats or de-facto standards are in wide use, we can't get rid of them however stupid they are... We've also seen that we'll have to live with these decisions for a long time.
Making wrong choices in a fast-moving field, such as computing, is natural of course. But not all mistakes are honest ones -- there are quite a few corporate interests that are actually trying to make us make choices that are just plain bad for everyone, except for the company in question.
We could avoid these bad choices. I'm willing to bet that the effects in productivity would be quite significant, if we did -- enough to cover a whole lot of meals anyway. And if individuals start thinking about this, maybe governments will too... Get the idea?
Sorry. I realise my tone was too harsh in the previous post. I'll explain my reasons for considering the MS model broken in the end.
The user will always have to install something though, so I don't think the user doesn't want to install software.. it make thier computer useless. And teh MS was is closer to this ideal then a package manager.
I guess we'll just have to agree to disagree... I think the following use case is very close to "not installing software":
realise you need a new tool because you can't accomplish a task with current ones
open "Add/remove software" (or whatever it is called in your system)
(search with keyword if you don't know the name of the software)
click on the software you want to use
It's not so much installing, more like selecting what program you wish to see on your Applications menu this week... No need to think about whether this sofware works on your OS version or if you want to trust the vendor -- that's already taken care of.
Usually with windows software, you install the software. Everything else you need is already on the system. With linux, this is rarely true
Very strange. This mismatch of experiences must stem from a) different use cases and b) different distros -- we really shouldn't be talking about just "Linux" when we talk about specific experiences.
Anyway, my two cents on this:
I've used Debian and Ubuntu for years on four different machines with varying use cases, like "scientific work", "software development", "home web & email", "music jukebox & file server". I have never experienced dependencies that apt didn't handle -- even through OS version changes. I have encountered maybe five cases where I've needed to get software ouside of the repositories (at least Matlab, Scratchbox dev. environment, newer amaroK, newer GRASS GIS), none of which were really a problem. I've found Debian/Ubuntu to be a very extensive collection of software and mostly my needs have been met by a search in the apt repository (and maybe a google search to get user opinions). I cannot imagine how much time I would have to spend to find and install tools that would accomplish approximately the same on a fresh Windows box.
Apropos, on the windows part of your comment, "Usually with windows software, you install the software. Everything else you need is already on the system": You left out acquiring the software. You'll either order it (and wait for delivery), buy it by visiting a shop, or download it from a random corner of the net. You need to do this for every piece of software you want to install... Whichever choice you make it's real work. This becomes a real problem when you consider how useless the basic install of Windows is -- you'll need to get a decent web browser, email program, IM program, office software, text editor, image viewer, etc. Most people choose the third option, downloading from the net, for at least some programs. What are the chances that at all of those downloads are from reputable sources?
My experience says that for a typical intelligent non-geek the number is closer to zero than one. Personal anecdote: I've become the "computer guy" for my parents and my parents-in-law. Both families have people who are computer-savvy in the sense that they know when they're "using a screw-driver to hammer in nails" and know how to get new tools -- they both buy and download software. However, they don't seem to have the internet-scam-radar that you and I probably do. They also seem to become illiterate when shown a dialog box (this is a fascinating phenomena, but also frustrating). I'm really out of options here -- They're clearly finding new ways to accomplish things (exactly what computers are for), but at the same time they are getting stuff they don't actually want on their machines. Not vast amounts, but enough to make me uncertain about the security of those machines. I'd really like to try a package management system with them but I cant', because I don't want to disturb their habits (==Windows) too much...
I think I am going to go install Linspire. Let's face it, I don't have the time to hassle with making mp3s and dvd players and voip work on the big distros either, and I am a Linux developer...
That's fine, and I agree with you on Linspire being a valuable addition to the Linux family. Still, I think we (as in, the people who at least partially understand the socio-techno-legal-economical consequencies of using proprietary and patented formats) should not stop educating people on this, in my opinion crucially important, issue... I think the Debian/Ubuntu way of handling this is fine: not installed by default, but not difficult to get installed (in theory at least, it could naturally be easier).
People should be made aware of the choice they're making -- maybe when the next proprietary format comes along, people will actually question it's viability before it becomes an essential part of the computing platform. A lot of the problems this industry is seeing are based on people getting locked-in to a format/software/platform, only because they didn't understand that this was even a possibility (or that there may have been better choices and that they can demand freeness and interoperability from their suppliers).
Getting along doesn't mean keeping your mouth shut.
Um, this is total bull. Very rarely do I need to find anything to install software I bought for Windows;
I wonder if you read my post _and_ the post I replied to.
Nuffsaid argued that downloading and installing stuff is the right way to install software (or at least that that is the problem with linux at the moment) and that intelligent non-geeks wouldn't understand the point of package management...
I disagreed and pointed out that the package management approach is more secure ans actually easier than what he suggested.
You say "total bull" and start talking about (apparently) shrink-wrap software which we had not been discussing.
I'm willing to bet most [intelligent non-geeks] don't care about package management. They just want to install the software and go on.
Actually, they don't want to install software, period. A package manager -approach is, in my opinion, closest to that ideal...
What I meant by "more secure" is that you only need to decide to trust one site and one organization, instead of evaluating a dozens of download sites and hundreds of software producers and deciding who to trust. You also do not need to get familiar with dozens of different software update methods to keep your software updated, and to manually check the websites of the programs that do not phone home...
A single package repository is the equivalent of a centrally governed economy. Nothing stops you from installing that one program that's not included in the repositories by hand -- you'll have to evaluate the security of the software provider of course, and think about how to keep it updated, but this is still better than having to do it dozens or hundreds of times...
I'm not claiming package management is a 100% secure software installation method -- I do believe it is the safest alternative for the general population...
Wow, that Jeff Waugh is a really, really patient guy. You kept ranting about your own preferences, do not accept his well thought out answers and pretty much keep insisting that he's wrong. Through all this you insult him every few lines or at least use diminishing terms to describe their work... His attitude never slips from the professional -- very impressive.
No offence, but I hope I never have to meet you in a situation like that. Being critical or demanding does not require bad manners.
Let's make one thing clear: Installing software from a single repository using a single UI is both simpler and more secure than doing it with a hundred different UIs from a hundred different web sites. The windows way of installing software is the opposite of "It Just Works" -- people are just used to going through all the hoops that particular installer requires (which is fine, they can keep using what works for them).
Explaining the concept of package management to intelligent non-geeks is not difficult. In fact in my experience people "get it" quite naturally -- they just need to be told about the pros and cons, not just "click here, here and here".
I disagree. It's a hack that doesn't fix a problem, but does create another one.
I'm willing to bet it won't prevent a significant amount of malware installations -- it's still just another OK button to press. People become illiterate when shown a security notice, showing it longer won't help.
Although it might help a few people, it is an annoyance for everyone -- another example of a program "knowing" what I want better than I do...
The field of "human factors" recognized that controls and displays need to be designed so that it's possible for a well trained human to get things right even in a hurry.
Agreed 100%. The problem is that a lot of people aren't ready to admit that secure computing is going to require learning... An example: E-mail applications with working crypto can (and hopefully will) be a lot easier to use than they are now, but they will always require some understanding of the underlying system -- otherwise social engineering will always work...
It's not about OSS or not OSS, it's about license incompatibilities. The Sun Java license _is_ unclear, but my interpretation is that it would still be against the license terms for Suse to distribute any other Java VMs (regardless of how they split their distribution on CDs)...
Slashdot Mirror
One example:
That sounds like distributing gcj or a java library implementation on the same OS is not allowed, although "in conjunction" and "similar functionality" might mean something totally different in lawyer-lingo...I Like Pudding: Noooooooooo!
Yeah, I counted. Get's old pretty fast.
"Derived works" is actually a complicated and not very clearly defined set... You might be right about a binary driver being completely separate work, but it certainly isn't as clear cut as you imply. This is what Linus had to say about this:
As you can see he views "derived" as including more than just stuff that includes kernel headers. In some ways that sounds reasonable, as the copyright of the kernel developers extends to the way linux works, not just the actual lines of code -- just like a story line of a book may be under copyright, depending on the case.
Just a note to TBE users: Anyone using Tabbrowser extensions should stay silent when the next Firefox-is-a-memory-hog flame fest ensues -- it has been one of the worst culprits in that area...
b) The software is just a client that supports Telepathy -- if Google stops their servers there will be others. My guess is that the client has nothing to do with Google (except using the same protocol): I believe this co-operation is mostly a PR effort.
c) it's not a phone. It's an Internet Tablet (with VOIP in the future).
More seriously, Nokia doesn't want to end up as a puppet for the telcos, and moving to products that work over IP is one of the ways they can achieve that.
Lottery winners make even more. What's your point?
To sum up: Yes, Microsoft can be sued and convicted for something all their competitors are doing.
1. Computing and computers are extremely important to our society (both culturally and efficiency-wise)
2. The single most important threat to the "growth" that we're getting from computers is, in my opinion, wrong choices regarding the basic platform (file formats, interfaces, other de-facto standards). We've seen time and time again that once formats or de-facto standards are in wide use, we can't get rid of them however stupid they are... We've also seen that we'll have to live with these decisions for a long time.
Making wrong choices in a fast-moving field, such as computing, is natural of course. But not all mistakes are honest ones -- there are quite a few corporate interests that are actually trying to make us make choices that are just plain bad for everyone, except for the company in question.
We could avoid these bad choices. I'm willing to bet that the effects in productivity would be quite significant, if we did -- enough to cover a whole lot of meals anyway. And if individuals start thinking about this, maybe governments will too... Get the idea?
-
realise you need a new tool because you can't accomplish a task with current ones
- open "Add/remove software" (or whatever it is called in your system)
- (search with keyword if you don't know the name of the software)
- click on the software you want to use
It's not so much installing, more like selecting what program you wish to see on your Applications menu this week... No need to think about whether this sofware works on your OS version or if you want to trust the vendor -- that's already taken care of. Very strange. This mismatch of experiences must stem from a) different use cases and b) different distros -- we really shouldn't be talking about just "Linux" when we talk about specific experiences.Anyway, my two cents on this:
I've used Debian and Ubuntu for years on four different machines with varying use cases, like "scientific work", "software development", "home web & email", "music jukebox & file server". I have never experienced dependencies that apt didn't handle -- even through OS version changes. I have encountered maybe five cases where I've needed to get software ouside of the repositories (at least Matlab, Scratchbox dev. environment, newer amaroK, newer GRASS GIS), none of which were really a problem. I've found Debian/Ubuntu to be a very extensive collection of software and mostly my needs have been met by a search in the apt repository (and maybe a google search to get user opinions). I cannot imagine how much time I would have to spend to find and install tools that would accomplish approximately the same on a fresh Windows box.
Apropos, on the windows part of your comment, "Usually with windows software, you install the software. Everything else you need is already on the system": You left out acquiring the software. You'll either order it (and wait for delivery), buy it by visiting a shop, or download it from a random corner of the net. You need to do this for every piece of software you want to install... Whichever choice you make it's real work. This becomes a real problem when you consider how useless the basic install of Windows is -- you'll need to get a decent web browser, email program, IM program, office software, text editor, image viewer, etc. Most people choose the third option, downloading from the net, for at least some programs. What are the chances that at all of those downloads are from reputable sources?
My experience says that for a typical intelligent non-geek the number is closer to zero than one. Personal anecdote: I've become the "computer guy" for my parents and my parents-in-law. Both families have people who are computer-savvy in the sense that they know when they're "using a screw-driver to hammer in nails" and know how to get new tools -- they both buy and download software. However, they don't seem to have the internet-scam-radar that you and I probably do. They also seem to become illiterate when shown a dialog box (this is a fascinating phenomena, but also frustrating). I'm really out of options here -- They're clearly finding new ways to accomplish things (exactly what computers are for), but at the same time they are getting stuff they don't actually want on their machines. Not vast amounts, but enough to make me uncertain about the security of those machines. I'd really like to try a package management system with them but I cant', because I don't want to disturb their habits (==Windows) too much...
People should be made aware of the choice they're making -- maybe when the next proprietary format comes along, people will actually question it's viability before it becomes an essential part of the computing platform. A lot of the problems this industry is seeing are based on people getting locked-in to a format/software/platform, only because they didn't understand that this was even a possibility (or that there may have been better choices and that they can demand freeness and interoperability from their suppliers).
Getting along doesn't mean keeping your mouth shut.
- Nuffsaid argued that downloading and installing stuff is the right way to install software (or at least that that is the problem with linux at the moment) and that intelligent non-geeks wouldn't understand the point of package management...
- I disagreed and pointed out that the package management approach is more secure ans actually easier than what he suggested.
- You say "total bull" and start talking about (apparently) shrink-wrap software which we had not been discussing.
Actually, they don't want to install software, period. A package manager -approach is, in my opinion, closest to that ideal...What I meant by "more secure" is that you only need to decide to trust one site and one organization, instead of evaluating a dozens of download sites and hundreds of software producers and deciding who to trust. You also do not need to get familiar with dozens of different software update methods to keep your software updated, and to manually check the websites of the programs that do not phone home...
A single package repository is the equivalent of a centrally governed economy. Nothing stops you from installing that one program that's not included in the repositories by hand -- you'll have to evaluate the security of the software provider of course, and think about how to keep it updated, but this is still better than having to do it dozens or hundreds of times...
I'm not claiming package management is a 100% secure software installation method -- I do believe it is the safest alternative for the general population...
No offence, but I hope I never have to meet you in a situation like that. Being critical or demanding does not require bad manners.
Explaining the concept of package management to intelligent non-geeks is not difficult. In fact in my experience people "get it" quite naturally -- they just need to be told about the pros and cons, not just "click here, here and here".
It's not about OSS or not OSS, it's about license incompatibilities. The Sun Java license _is_ unclear, but my interpretation is that it would still be against the license terms for Suse to distribute any other Java VMs (regardless of how they split their distribution on CDs)...