Re:I love the smell of GNUpalm in the morning. . .
on
Vietnam Going Open Source
·
· Score: 3, Insightful
The only countries that seem to be looking at this type of thing are extremely socialist or totalitarian in nature (Vietnam, China, European Union, etc.)
Excuse me, but may I say WTF?
Since when is the EU "extremely socialist" or "totalitarian in nature"? Since when is it even a country?
In fact, as you seem have written off most of the planet with that throwaway statement, let's move on and look for more of the same...
This is a very bad thing for the American software industry (and by extension programmers.. you know, some of these companies give you a paycheck?)
Hmmm, except maybe all those programmers who aren't American, who maybe might benefit from the new market in OS customization and development that was just opened up.
Commercial software is NOT a bad thing, only the abuse of a monopoly on commercial software.
You can't just assert that and wave your hands around in a "pfft! it's obvious!" manner, you'd have to back that position up.
Sounds silly, but the person who started this whole free software thing was in fact American, and has pretty convicingly argued from first principles that in fact proprietary software (which I assume is what you really meant) is a bad thing. I'm not saying I agree with him, just that if you want to be taken seriously you should tell us why proprietary software is not a bad thing.
We shouldn't discourage companies from producing quality software by threatening to boycott them.
Presumably they wouldn't be boycotted if people were 100% happy with what they produced and had no complaints.
I couldn't believe my eyes first. I asked it to run iexplore.exe and the damn thing did just that. The only problem was that it always complained about content ratings when starting up (Nothing serious), some toolbar buttons were all black outlines, and the managed window didn't have minimize button, which looked pretty ridiculous. But it worked otherwise just fine!
I guess MS has funny ideas of "tight OS integration" =)
Well, you can install IE just like an update to Windows 95 or 98. IE is undoubtably integrated though - many apps expect it to be there and won't run if it isn't.
not to put down their achievement, but how is this any functionally different than running said applications in vmware/bochs/plex86?
It's very different. I suggest you try their trial version and see. For starters, you get practically 100% performance, there is no slowdown due to emulating a CPU or holding an entire copy of Windows in memory. Secondly, you get much better integration - apps appear in your Linux menu system, they use your native window manager (so they support virtual desktops etc), you can copy and paste between native and emulated apps... the list goes on and on.
Most people go "oooh what's the point in that, stupid Wine developers", but actually we put quite a bit of effort into making IE run correctly because so many other programs rely on it (or rather, its components) being there and functioning.
NB: I don't work for CodeWeavers, but if you can get it to install IE works pretty well on WineHQ. Suitable for light browsing certainly.
It's good as it allows people to use a popular windows app in Linux x86. But it's bad because now there is less pressure on Macromedia to develop a native port or for somebody else to write a free Dreamweaver killer.
I think this argument is rather academic. Being able to use popular apps in Linux is undoubtably good, however the "bad" arguments rely on two flawed assumptions:
1) Macromedia might one day do a native port. Not going to happen anytime soon guys. Dreamweaver is a huge app, and I'd be willing to bet that (as with most apps) the majority of the code is platform specific GUI and graphics calls. It would take a truly astonishing amount of manpower to port it to say GTK+, make it fully integrate and so on, and it just isn't economically viable while Linux has only 1% of the desktop market. Even if we had 5% or 10% we'd still be pushing our luck - a port in this sense often means a rewrite.
2) That we'd have an open source dreamweaver killer anytime soon. Quanta is about the only thing that comes close, and while a great effort, is not a Dreamweaver killer. It might be one day, but that's yet another long term dream.
Basically, the best way out of a bad situation here is via emulation, which is exactly what we're doing.
the classic problem with IP is the following:
company X spends 10b over 20 years to develop a drug that effectively treats a human ailment. Once discovered, manufacture of this drug is essentially zero cost per part. The 10b research cost has to be amortized over a certain volume of pills for a reasonable time period to allow return on investment, otherwise the company goes under and brilliant people dont figure out new ways to cure ailments.
You're assuming the cost has to be amortized over time. In fact, if the company involved gets a healthy return on their investment, whatever the means, we should see continued development of drugs.
There are probably other ways. For instance, if the cost of the research was independently audited and then multiplied by 1.2, you would have the amount of money that needed to be returned to the company by the market in order to make the venture worthwhile. If the legal system enforced a barrier to entry whereby you could only produce these goods once you had equalized your share of the pot (so the first competitor has to pay you half the research cost, the second pays a third and the other two get some back etc), then eventually the cost would have been returned while allowing a free market to drive down prices.
This does not of course deal with those who ignore the system and produce generics - you could see open source software as being the software equivalent of generics.
On the SCO issue - given that SCO bought the rights to UNIX and did not develop it itself, I have a hard time feeling sympathy for their "plight". Development is about standing on the shoulders of giants, and whether you clean-room reimplement or buy a good technology seems irrelevant to me.
For example, you've got Red Hat all but forking glibc by adding the NPTL stuff to it... people compile against that and then those binaries aren't going to work on systems that use a more standard glibc.
That is not correct. Binaries built using Red Hats glibc do not have any binary portability issues due to NPTL. The only issues are those that are standard with glibc 2.3, such as thread-local locale models.
In fact Linux is more binary compatible than people tend to think. Under the hood, it's all the same code at the end of the day.
i don't understand why all the linux folks want the mainstream public converted over to linux!
It's fun, we like Linux and want other people to be able to enjoy it as much as we do, more users == more apps == more users etc
i like the fact that because i run linux, i have more capability than the average windows joe.
Running Linux should not be a status symbol, period. If you want the respect of your peers, if you want to be seen as "elite" then learn to hack code - THAT is the right way to demonstrate technical capability, not simply knowing how to use the command line.
i like the fact that i am part of a technical movement that helps each other in times of need (newsgroups, forums).
i like the fact that when my rights are threatened or otherwise, that it's this same group of folks that stand together (SCO?).
Those things won't disappear, even if one day everybody in the world uses Linux. The free software developer community is strong because of the shared experiences of building things together. It doesn't matter how many users you have, that shared experience is still there.
Of course the distinction between user and developer will become greater, but this is a natural progression.
one of the things that MS doesn't have is the 'comminuty closeness' that *nix users have.
*nix (god I hate that stupid term) users have no community. Sorry, that's just bollocks. There are all kinds of Solaris, AIX, Tru64 users all over the world who couldn't give a rats ass about what's happening beyond their office door as long as the paycheque comes in on time.
There is a free software community, and it centers around Linux - don't confuse the fact that Linux is roughly based on UNIX with the community aspects. The word "community" comes from Latin, "cum munere." "Munere" is "to give," and "cum" is "among each other," so, community
means "to give among each other."
As long as people share what they make, there will be a community. It has nothing to do with what command shell you run, never forget that. It has nothing to do with how many people use what is built - the community will always be made up of the subset that share with each other.
i'm afraid that if linux were to ever win the desktop war, this closeness and community won't be as friendly or as helpful.
There are friendly and helpful communities of Windows users. They are called friends and family. Of course there are organized groups too, such as Protonic which is similar to our LinuxQuestions.org site.
Basically I think you worry too much. As long as there are people who care about sharing with each other, the community will remain strong, regardless of how much it grows.
Porting an app to a native toolkit, either Qt or GTKmm would work well there, is definately the best way to go. These toolkits have teams of people working on them fulltime, lots and lots of testing and so on, whereas the Wine toolkit implementation does not have either of these.
I'd personally go for GTKmm, as it's more C++ like and you don't have to pay licensing fees - also portable to Windows etc. You get nice stock artwork too. Qt works well also of course.
For many large projects though, it's just not feasable to rewrite the app to have no Win32 dependencies. In this case, it's normally a better idea to improve Wine to the needed level, especially as that way you don't disrupt the main dev team.
I was thinking primarily of "A tale in the desert" - the game was running on Wine for many months before they did the release (beta testers etc) but they still hired a dev to do a full Linux release, because their customers wanted it and would be more willing to subscribe (it's a MMORPG) if they had a native client.
The other company I can't name, but it was basically decided (or so I believe) that as there was clearly a significant base of users running their app under emulation that they would save their customers a lot of grief by officially producing and supporting a native version.
Firstly, writing a new driver for Linux is several orders of magnitude easier than rewriting an entire application for portability. A program like MS Office will never, ever be a "native" app, say that uses GTK, Qt, whatever - it's far too huge to ever rewrite in that way.
A driver, on the other hand, is not so complex, and can be more easily developed independantly, especially by the community.
Really, Wine is useful primarily for apps that will never, ever be ported. I've seen little evidence that it discourages companies from doing native ports. I've seen quite a few companies do native ports even when their programs worked just fine in Wine. Only time will tell if this has bad effects or not.
You're not understanding that there is a fundamental distinction between support and software. They are entirely unrelated.
They are not limiting your rights to install their products on another server. They are limiting your rights to do that AND get support for it as well. If you don't care about support, no problem!
They are allowed to do this, in exactly the same way that I am allowed to give you Knoppix and say "I will give you help learning Linux as long as you buy me a beer". I am not restricting your rights nor am I violating the GPL. I am placing conditions on my help - just the same as they are.
There are no restrictions on the software. You can do what you like with the software. There are restrictions on the support given - they are entirely separate.
On the other end of the spectrum are the many commercial and internally developed applications that have been written for Windows but not ported to Linux. With those, Handy says, "there isn't interoperability" at all.
As somebody who is available for hire to make apps (any apps) work on Linux via Wine, I must point out that this is just blatently not true. I (and many others) have been hired before by companies wishing to move their infrastructure to Linux. For custom software, the job is often reasonably straightforward as the source is available, but even for 3rd party apps the company uses it is still possible.
So, to say there is no interoperability is not true. Typically, if you do the math, you may find it is cheaper to hire a Wine developer for a time to make your apps work on Linux than continue to license Windows for all the machines needed.
It has nothing to do with the GPL, so stop spreading FUD.
You can download Advanced Server, install it on as many machines as you like, all for free - both beer and speech. If you want support with that, you have to agree to their license agreements, and this is reasonable - otherwise a shop with 100 machines would take out 20 support licenses, then when one broke it would always be conveniently one of the 20.
An API is either documented and exported (in which case its behaviour its clearly defined) or internal and not documented, in which case its behaviour can change.
Really, there should be very few APIs that are documented in public documentation but internal. Also, the percentages of apps that seem to have been broken by this upgrade is quite high - I find it hard to believe they are all dodgy GUI hacks and suchlike. Perhaps they have been changing the API internals without realising that some of the details of the internals had leaked out?
The latest versions of both Gaim and aMSN work, because they support version 9 of the MSN protocol. Any clients that do not support this version of the protocol can no longer log in - clients that do, can.
Development houses using the MS.NET stuff _do not want_ alternatives. If they wanted alternatives, they would be using Java, Python, etc.
Java does not offer a compelling desktop development platform (swing is dire, sorry...) and Python has little momentum outside of Linux/GNU - how many desktop apps do you know that are written for Win32? How many for GTK+?
In fact, most Windows developers can choose either:
1) C++
2) Delphi
3).NET
The other alternatives aren't mature enough, aren't large enough, or simply aren't backed by big enough companies. It's no wonder they are jumping for.NET
A plan that includes selling product to someone that doesn't want the thing you are selling, doesn't sound so good to me.
In fact I've met quite a few people who are interested in Mono for cross platform compatability purposes, but they can't justify writing 100% portable code right now. They can't justify using Python (a slow, dynamically typed language) and want to move away from C++ - Java is not an option because it's too clunky on the desktop. So they use C#, while still wanting Mono.
Andi, you seem to be assuming that if Mono didn't exist and offer cross platform compatability, nobody would use.NET - clearly this isn't the case. The arguments over Mono are worth having, but are mostly about whether we should use it for our own software.
We are doomed to have to recreate their APIs regardless of what our community decides: a few clued in people are seeing that maybe.NET gives them greater compatability, and if they've done their homework they'll realise that they still have to choose their APIs and S.W.F is not all that portable. That has to be better than people continuing to churn out enormous and baroque pieces of C++ for us to reverse engineer.
You think Wine does not infringe any patents Microsoft has? That's almost certainly incorrect - fortunately both Mono and Wine have companies (ie funding) behind them these days, such that if they were to go to court, they stand a chance of getting the patents written off.
Look - if we sit around moaning about software patents all day, we'd never get any software written at all, it's that simple. I think peoples fears about Mono are mostly unjustified, and I think that because I've never seen a write up of how Mono will damage the movement that doesn't sound like it was written by a conspiracy theorist.
When I see cold, hard, objections based on fact and not "what if, what if, what if", then maybe I will be wary. This paper certainly is not it.
It was such a startlingly good idea that I am not at all surprised no-one appears to have done anything similar.
It's not used much because it sucks. Time and time again it was found that users cannot reliably rate their own skill, or they are scared of missing out on features. Typically almost all users chose "Advanced", despite their actual level of proficiency.
It also tended to be used as a way to cover up the fact that a program had too much stuff crammed into it in a bad way.
Correct me if I'm wrong, but isn't Pango supposed to be able to render and edit Indian text correctly these days? I'm certain I've seen screenshots if GTK2 apps doing that
Mac OS X lets you run all the Adobe stuff natively, a real UNIX underneath for porting, and with the G5, a processor fast enough to get in the ballpark of x86 for price/performance
Well, they don't need anything other than Photoshop from Adobe, which runs just fine on Wine nowadays (and other than the colour of the controls it is basically "native"), and Linux/x86 is far more attractive pricewise than a Mac...
Slashdot Mirror
Excuse me, but may I say WTF?
Since when is the EU "extremely socialist" or "totalitarian in nature"? Since when is it even a country?
In fact, as you seem have written off most of the planet with that throwaway statement, let's move on and look for more of the same...
This is a very bad thing for the American software industry (and by extension programmers.. you know, some of these companies give you a paycheck?)
Hmmm, except maybe all those programmers who aren't American, who maybe might benefit from the new market in OS customization and development that was just opened up.
Commercial software is NOT a bad thing, only the abuse of a monopoly on commercial software.
You can't just assert that and wave your hands around in a "pfft! it's obvious!" manner, you'd have to back that position up.
Sounds silly, but the person who started this whole free software thing was in fact American, and has pretty convicingly argued from first principles that in fact proprietary software (which I assume is what you really meant) is a bad thing. I'm not saying I agree with him, just that if you want to be taken seriously you should tell us why proprietary software is not a bad thing.
We shouldn't discourage companies from producing quality software by threatening to boycott them.
Presumably they wouldn't be boycotted if people were 100% happy with what they produced and had no complaints.
Indeed it does, and it can render even complex web pages. Here it shows the Adobe SVG plugin, as you can see it renders correctly and the issue with the ratings doesn't occur. Here it is viewing microsoft.com
I guess MS has funny ideas of "tight OS integration" =)
Well, you can install IE just like an update to Windows 95 or 98. IE is undoubtably integrated though - many apps expect it to be there and won't run if it isn't.
It's very different. I suggest you try their trial version and see. For starters, you get practically 100% performance, there is no slowdown due to emulating a CPU or holding an entire copy of Windows in memory. Secondly, you get much better integration - apps appear in your Linux menu system, they use your native window manager (so they support virtual desktops etc), you can copy and paste between native and emulated apps ... the list goes on and on.
NB: I don't work for CodeWeavers, but if you can get it to install IE works pretty well on WineHQ. Suitable for light browsing certainly.
I think this argument is rather academic. Being able to use popular apps in Linux is undoubtably good, however the "bad" arguments rely on two flawed assumptions:
1) Macromedia might one day do a native port. Not going to happen anytime soon guys. Dreamweaver is a huge app, and I'd be willing to bet that (as with most apps) the majority of the code is platform specific GUI and graphics calls. It would take a truly astonishing amount of manpower to port it to say GTK+, make it fully integrate and so on, and it just isn't economically viable while Linux has only 1% of the desktop market. Even if we had 5% or 10% we'd still be pushing our luck - a port in this sense often means a rewrite.
2) That we'd have an open source dreamweaver killer anytime soon. Quanta is about the only thing that comes close, and while a great effort, is not a Dreamweaver killer. It might be one day, but that's yet another long term dream.
Basically, the best way out of a bad situation here is via emulation, which is exactly what we're doing.
You're assuming the cost has to be amortized over time. In fact, if the company involved gets a healthy return on their investment, whatever the means, we should see continued development of drugs.
There are probably other ways. For instance, if the cost of the research was independently audited and then multiplied by 1.2, you would have the amount of money that needed to be returned to the company by the market in order to make the venture worthwhile. If the legal system enforced a barrier to entry whereby you could only produce these goods once you had equalized your share of the pot (so the first competitor has to pay you half the research cost, the second pays a third and the other two get some back etc), then eventually the cost would have been returned while allowing a free market to drive down prices.
This does not of course deal with those who ignore the system and produce generics - you could see open source software as being the software equivalent of generics.
On the SCO issue - given that SCO bought the rights to UNIX and did not develop it itself, I have a hard time feeling sympathy for their "plight". Development is about standing on the shoulders of giants, and whether you clean-room reimplement or buy a good technology seems irrelevant to me.
That is not correct. Binaries built using Red Hats glibc do not have any binary portability issues due to NPTL. The only issues are those that are standard with glibc 2.3, such as thread-local locale models.
In fact Linux is more binary compatible than people tend to think. Under the hood, it's all the same code at the end of the day.
It's fun, we like Linux and want other people to be able to enjoy it as much as we do, more users == more apps == more users etc
i like the fact that because i run linux, i have more capability than the average windows joe.
Running Linux should not be a status symbol, period. If you want the respect of your peers, if you want to be seen as "elite" then learn to hack code - THAT is the right way to demonstrate technical capability, not simply knowing how to use the command line.
i like the fact that i am part of a technical movement that helps each other in times of need (newsgroups, forums). i like the fact that when my rights are threatened or otherwise, that it's this same group of folks that stand together (SCO?).
Those things won't disappear, even if one day everybody in the world uses Linux. The free software developer community is strong because of the shared experiences of building things together. It doesn't matter how many users you have, that shared experience is still there.
Of course the distinction between user and developer will become greater, but this is a natural progression.
one of the things that MS doesn't have is the 'comminuty closeness' that *nix users have.
*nix (god I hate that stupid term) users have no community. Sorry, that's just bollocks. There are all kinds of Solaris, AIX, Tru64 users all over the world who couldn't give a rats ass about what's happening beyond their office door as long as the paycheque comes in on time.
There is a free software community, and it centers around Linux - don't confuse the fact that Linux is roughly based on UNIX with the community aspects. The word "community" comes from Latin, "cum munere." "Munere" is "to give," and "cum" is "among each other," so, community means "to give among each other."
As long as people share what they make, there will be a community. It has nothing to do with what command shell you run, never forget that. It has nothing to do with how many people use what is built - the community will always be made up of the subset that share with each other.
i'm afraid that if linux were to ever win the desktop war, this closeness and community won't be as friendly or as helpful.
There are friendly and helpful communities of Windows users. They are called friends and family. Of course there are organized groups too, such as Protonic which is similar to our LinuxQuestions.org site.
Basically I think you worry too much. As long as there are people who care about sharing with each other, the community will remain strong, regardless of how much it grows.
I'd personally go for GTKmm, as it's more C++ like and you don't have to pay licensing fees - also portable to Windows etc. You get nice stock artwork too. Qt works well also of course.
For many large projects though, it's just not feasable to rewrite the app to have no Win32 dependencies. In this case, it's normally a better idea to improve Wine to the needed level, especially as that way you don't disrupt the main dev team.
The other company I can't name, but it was basically decided (or so I believe) that as there was clearly a significant base of users running their app under emulation that they would save their customers a lot of grief by officially producing and supporting a native version.
Firstly, writing a new driver for Linux is several orders of magnitude easier than rewriting an entire application for portability. A program like MS Office will never, ever be a "native" app, say that uses GTK, Qt, whatever - it's far too huge to ever rewrite in that way.
A driver, on the other hand, is not so complex, and can be more easily developed independantly, especially by the community.
Really, Wine is useful primarily for apps that will never, ever be ported. I've seen little evidence that it discourages companies from doing native ports. I've seen quite a few companies do native ports even when their programs worked just fine in Wine. Only time will tell if this has bad effects or not.
They are not limiting your rights to install their products on another server. They are limiting your rights to do that AND get support for it as well. If you don't care about support, no problem!
They are allowed to do this, in exactly the same way that I am allowed to give you Knoppix and say "I will give you help learning Linux as long as you buy me a beer". I am not restricting your rights nor am I violating the GPL. I am placing conditions on my help - just the same as they are.
There are no restrictions on the software. You can do what you like with the software. There are restrictions on the support given - they are entirely separate.
As somebody who is available for hire to make apps (any apps) work on Linux via Wine, I must point out that this is just blatently not true. I (and many others) have been hired before by companies wishing to move their infrastructure to Linux. For custom software, the job is often reasonably straightforward as the source is available, but even for 3rd party apps the company uses it is still possible.
So, to say there is no interoperability is not true. Typically, if you do the math, you may find it is cheaper to hire a Wine developer for a time to make your apps work on Linux than continue to license Windows for all the machines needed.
You can download Advanced Server, install it on as many machines as you like, all for free - both beer and speech. If you want support with that, you have to agree to their license agreements, and this is reasonable - otherwise a shop with 100 machines would take out 20 support licenses, then when one broke it would always be conveniently one of the 20.
An API is either documented and exported (in which case its behaviour its clearly defined) or internal and not documented, in which case its behaviour can change.
Really, there should be very few APIs that are documented in public documentation but internal. Also, the percentages of apps that seem to have been broken by this upgrade is quite high - I find it hard to believe they are all dodgy GUI hacks and suchlike. Perhaps they have been changing the API internals without realising that some of the details of the internals had leaked out?
For now, it seems to be that simple.
Java does not offer a compelling desktop development platform (swing is dire, sorry...) and Python has little momentum outside of Linux/GNU - how many desktop apps do you know that are written for Win32? How many for GTK+?
In fact, most Windows developers can choose either:
1) C++ .NET
2) Delphi
3)
The other alternatives aren't mature enough, aren't large enough, or simply aren't backed by big enough companies. It's no wonder they are jumping for .NET
A plan that includes selling product to someone that doesn't want the thing you are selling, doesn't sound so good to me.
In fact I've met quite a few people who are interested in Mono for cross platform compatability purposes, but they can't justify writing 100% portable code right now. They can't justify using Python (a slow, dynamically typed language) and want to move away from C++ - Java is not an option because it's too clunky on the desktop. So they use C#, while still wanting Mono.
We are doomed to have to recreate their APIs regardless of what our community decides: a few clued in people are seeing that maybe .NET gives them greater compatability, and if they've done their homework they'll realise that they still have to choose their APIs and S.W.F is not all that portable. That has to be better than people continuing to churn out enormous and baroque pieces of C++ for us to reverse engineer.
Look - if we sit around moaning about software patents all day, we'd never get any software written at all, it's that simple. I think peoples fears about Mono are mostly unjustified, and I think that because I've never seen a write up of how Mono will damage the movement that doesn't sound like it was written by a conspiracy theorist.
When I see cold, hard, objections based on fact and not "what if, what if, what if", then maybe I will be wary. This paper certainly is not it.
working on it..... we're workin on it. now time for bed.
No it's not, they get good startup time at least partly by optimising the working set.
It's not used much because it sucks. Time and time again it was found that users cannot reliably rate their own skill, or they are scared of missing out on features. Typically almost all users chose "Advanced", despite their actual level of proficiency.
It also tended to be used as a way to cover up the fact that a program had too much stuff crammed into it in a bad way.
Correct me if I'm wrong, but isn't Pango supposed to be able to render and edit Indian text correctly these days? I'm certain I've seen screenshots if GTK2 apps doing that
Well, they don't need anything other than Photoshop from Adobe, which runs just fine on Wine nowadays (and other than the colour of the controls it is basically "native"), and Linux/x86 is far more attractive pricewise than a Mac...