If you mean MSI then it most certainly can. If you mean the installer for Windows the OS then of course it cannot - you aren't meant to install Windows itself into Wine.
That's incorrect. TransGaming have released a few patches but the bulk of their work is proprietary - WineX is based on a fork of Wine from back when it was X11 licensed.
Not really "fixed" as that's trivial to workaround, but it's a good start. The other solutions proposed are probably better ideas - whitelisting XPI install sites is a good one. There just are not that many uses for XPI in an uncontrolled environment like the public net.
Not anywhere near as irritating as FLOSS though. I prefer "free software developer" but 99.9% of people get the wrong idea from that, so I just say "open source" and suck it the semantic fuzziness:)
Hmm, I'd say both those approaches are wrong. What's so bad about
void toggle(bool *b) { *b = !(*b); }
The extra parens around *b probably aren't needed, though. It's certainly clearer than ^= would be. Your way may be slightly faster but I'd wager not fast enough to make up for the less clear (IMHO) code.
This sort of demonstrates what I was talking about w.r.t reimplementing hashtables. I've seen some standard libraries that use things like the gcc __builtin_expect construct to give CPU branch prediction hints: no matter how smart you might be, how many people will optimize their own hashtables to that sort of extent?
Well, I got a lot of (possibly deserved) flaming for that post:) I never said theory was useless though, just that - like many university degrees - they seem to equip you poorly for the real world. Note I said poorly, not that they don't equip you at all.
Mind you, I have a somewhat vocational bent by nature I guess. I'm the sort of person who can't really see the point of studying, say, History for 3 years (apart from to just spend time maturing at university of course).
On the science vs engineering thing, I'd normally agree but it seems that at least in England virtually all universities conflate Software Engineering and CompSci to such a high level that there's little reason to choose one over the other. Worse, most employers apparently see SE as a "softer" subject despite in many courses it being made up of exactly the same modules +/- a few, so the end result is that a ton of people do CS and virtually none do software engineering. It works in reverse too - one of the first modules we did in my CS course was "Software Engineering" which dealt with processes, evaluation cycles etc.
That's definitely a confused state of affairs, but realistically there's a huge job market for software engineers and a very small one for computer scientists so it's not surprising that the courses got merged over time.
Yeah, I'd agree with that. To be frank, there's a huge gaping hole in the education market for courses that teach you how to write software. Computer Science is great and all, and I'm enjoying it, but I know that a lot of what I'm being taught is pretty useless.
OS Design? Fascinating, but ultimately irrelevant for 99% of coders. Implementing your own hashtables? Useful to gain an insight into how they work, but virtually any development platform people work in except raw C these days will provide highly tuned and optimized hashtable implementations. If anything you shouldn't roll your own, as it'll make your code less readable, more bloated and probably slower.
But it seems looking at the course material that important topics for writing real world software is simply not taught. Where are the lectures on writing internationalized code (think character encodings, flippable UI design etc). Where are the lectures on writing code with low startup overhead? What about teaching people the merits of various toolkits? Accessibility? Version control theory? (to be fair I think this is talked about at some point on my course, but I know some that don't)
It's possible for a student to come out of a CompSci course and be unable to explain why a containment-based widget toolkit is better than a positional one, yet can talk about VM swapping algorithms and implementing the fastest hashtable probing all day. Which would you rather employ?
That's a bizarre definition of cross platform. Apps written using the Microsoft.NET stack are not really any more portable than apps written using Win32 - sure we can run them if we reimplement the APIs and write a loader to parse EXE and DLL files, but that's a long way from the vendor shipping an ELF binary which uses GTK+ or Qt to draw its UI.
People get really hung up on cross platform apps - Java was designed to do that but ended up simply inventing a new platform and hauling it around everywhere,.NET was not designed to do that and so Windows.NET apps are easily distinguishable from Linux.NET apps.
He's probably referring to System.Windows.Forms which is just a (leaky) wrapper for the Win32 GUI toolkit. Win32 is not object oriented so you can't really usefully subclass the widgets, so they're declared final.
3) they cover their asses in the case of special software with the Citrix and VMware thing
Special software? I'm really not convinced the Citrix/VMware thing will fly - almost any company of a size that would find desktop Linux interesting will have multiple custom or specialist apps deployed. This is especially true of things like some popular HR apps which despite being "web apps" tend to have bizarre dependencies on Internet Explorer and particular JVMs, etc - basically I find it hard to imagine that a companies employees would be satisfied with running Windows in a box.
Even if VMware, Inc find some way to hack window management a la Classic-mode in OS X into things, the end result is still going to be cludgy, slow and require actually running a full copy of Windows. VMware/Citrix is great for running Windows in a box, but as a corporate migration solution? I think not.
Because fair use is a gaping hole that many (most?) people would exploit to get free stuff.
I've never ever heard anyone talk about fair use outside of Slashdot, period. For most people it just isn't a big deal.
Making a DRM system that works with fair use but still protects artists is really hard, probably impossible. Apples DRM sort of gets there by being weak and easily exploited, but I'm not sure that's really an answer. It's a solution by being half-arsed.
It makes me wonder if the whole system of copyright is rather broken, to be frank. But I don't know of a better way, so I can't really criticize too much.
It's pretty depressing that this was modded down (overrated + flamebait). It's overly long, but basically the points raised by the AC are salient ones. I too have seen people justify piracy like this. It's pretty bad.
Yeah, but the Linux Counter is pretty useless - I know a ton of people who use Linux and haven't registered, mostly either because they don't know about it or can't be bothered filling out their details simply to be counted.
In other words, you're going to bitch about a program that's several years old, and refuse to try the new version which focussed amongst other things on improving the UI?
That's pretty lame dude. Not saying other people don't do it, but it's still pretty lame.
Given the enormous changes (most would say improvements) in the Gimps UI in version 2.0, this is clearly wrong.
One thing that is painfully clear about this thread is that the majority of posters have not actually used the latest version, as many of their criticisms have been fixed there. To go beyond that and claim the Gimp developers don't listen to the "professions" is, to be frank, total garbage.
I always marvel at the fact that Notes runs twice as fast on Windows as it does on Linux. Of course, Wine = Wine is not an Emulator also translates to WinW = Wine is not Windows. To promise performance parity for Wine with Windows is plain stupid.
[snip]
Yes, there are performance problems with Notes (6.5.1 at any rate) on Wine, but they are not severe. I find it perfectly usable.
Regardless, you cannot draw conclusions based on your experiences from one (1) app run under Wine to all apps. I also run Office XP on Wine and it runs just as fast as it does on Windows. The same is true of Photoshop.
They are different programs, work in different ways, and use different codepaths. It's not surprising they run at different speeds.
No, but most people on Slashdot don't know what they're talking about (including you it seems).
Crazney has broken the pointless encryption on streaming things in the iTunes library to other machines on a LAN.
It has nothing to do with iTMS. Repeat after me: it has nothing to do with iTMS.
The encryption on streaming tunes between clients only serves two purposes: to try and keep people on the Apple upgrade treadmill and to force people to use iTunes on all their machines if they want to stream music between them from the iTunes library. This is your own music we're talking about here, no copyright violations are taking place.
To be frank, Apple is taking the piss with this sort of encryption, and now the piss is being taken out of them. Too bad, but it has nothing to do with FairPlay.
People seem to be conflating two unrelated things. iTunes can stream music from its library to other clients on the network. This is entirely independent from iTMS, the music streamed by DAAP can easily be music you ripped yourself.
So really this has nothing to do with hurting Apple, or not agreeing to a "EULA", and it has everything to do with Apple cynically attempting to manipulate network effects. Your brother sharing his CD collection on the home network using iTunes? You can't use WinAmp, WMP, RhythmBox, Muine or whatever to access that, you have to use iTunes too. Then when you share your music, it cascades onwards.
This is especially true in places like homes, student flats and college networks, like the ones crazney is on. Really, Apple have no excuse for this: restricting DAAP can only have one goal and that is to use peer-power of the type that keeps Windows entrenched to give iTunes an upper hand. As such it frankly deserves to be cracked.
I know crazney. He's a good guy. We talk often - he isn't out to screw Apple or steal music. He wants to play the music on his Mac laptop using the iTunes streaming system: this seems totally fair to me.
They seem to be pretty similar, except whereas XAML is based on.NET reflection, Glade is based on GObject reflection.
The main problem is that I don't think anybody has really experimented with using Glade for anything other than GTK widget heirarchies. The other problem is that GObject isn't really a match for the.NET object model currently. With some work it could be a fantastic middle ground between what COM was and what.NET is - you still gain the advantage of actually using the real platforms instead of "skins" (see CPython vs Python.NET sometime), but you can still interop with other platforms at the OO level.
The main problem with sucking Glade files through SOAP/XML-RPC is that you'd have to relay all the signals to an instance on the server. GTK+ wasn't really designed for high latency feedback like that. It *could* work, but it probably wouldn't work well.
So there seem to be several alternatives:
1) Use Mono with our own XAML/Avalon implementation..NET already has code access security and other things you want for sandboxing, though given the mess that Java applets/security was I'm not sure it's necessary a good thing. Pros: compatibility, Cons: we're chasing a standard we don't have any influence over so will always be behind.
2) Improve Mozilla/XUL - some previous poster suggested fully documenting it, integrating it with Apache/Tomcat, writing corporate webapps in it etc - YES. These are excellent suggestions for how to improve XUL as an app delivery platform. I used to be quite into XUL and knew the creator of JabberZilla : the things you could do with it even in the pre 1.0 days were mind boggling. XUL may never be a good way to write client side apps, but for internally deployed webapps it's fantastic.
3) Leverage current Linux technologies to produce something like ActiveX but with security. I was thinking about that this morning. If you have an SELinux/DBUS enabled system you can get quite good sandboxing even for native code. You'd want some kernel mods to prevent certain "attacks", but I think you can get pretty close to what.NET CAS provides except you can also reuse the entire free desktop platform as well (all the libraries that don't have.NET bindings for instance).
You could then layer some simple technologies on top of the web browser to write web-deployed client-side apps using GTK/Glade or Qt or whatever.
In other words, it'd have all the good bits of ActiveX, but without the bad bits. I'm not sure this is a good way to go though - ActiveX/Java Applets seem to be a mostly dead idea, except in a few old-skool corporate webapps.
Common misconception. Actually there is no "true geek" distro - at CodeWeavers which is staffed almost entirely by geeks, you'll see everything from Debian Testing (the CTO), to Slack to bleeding edge Fedora/SuSE releases. They are all Linux, after all.
The iTunes installer requires some COM functionality not presently available on Wine (service-based activation, iirc). Because iTunes only works in 2k+ you can't use native DCOM to hack around this, the only solution is to implement support for it in Wines own OLE code.
Apparently, iTunes does actually work if you install it on Windows then copy it (and the registry entries it creates) across, though it's not really perfect.
Anyway, it's pretty high up on the CodeWeavers wishlist, so expect to see us work on it at some point. Just don't expect it soon - iTunes isn't really a killer app and there are some nice native equivalents like Muine or RhythmBox available.
Last time I checked rhel was entirely under the GPL. You have to pay for support which seems to have confused a lot of people who conflated gratis with libre, but it's definitely free software.
The Win32 system makes calls into the kernel to get system things done (like disk and screen I/O), but tracks all the Win32 stuff itself (like windows and message queues).
No, windows and message queues (and other similar GDI/USER objects) are managed by the kernel in Windows.
Win32 is, in essence, just a 'personality' running on the NT core.
I guess you're thinking of the kernel32 -> ntdll mapping.
I believe this could work. It would not, however, remove the need to own a copy of Windows, so its use would be more for the pragmatists ('we want it to work') than the idealists ('we want it to be free')
It could work, though it'd be extremely difficult to do. It's also almost certainly not what they've done (or haven't done) as their site says you don't need a copy of Windows when using David.
Slashdot Mirror
If you mean MSI then it most certainly can. If you mean the installer for Windows the OS then of course it cannot - you aren't meant to install Windows itself into Wine.
That's incorrect. TransGaming have released a few patches but the bulk of their work is proprietary - WineX is based on a fork of Wine from back when it was X11 licensed.
Not really "fixed" as that's trivial to workaround, but it's a good start. The other solutions proposed are probably better ideas - whitelisting XPI install sites is a good one. There just are not that many uses for XPI in an uncontrolled environment like the public net.
Not anywhere near as irritating as FLOSS though. I prefer "free software developer" but 99.9% of people get the wrong idea from that, so I just say "open source" and suck it the semantic fuzziness :)
This sort of demonstrates what I was talking about w.r.t reimplementing hashtables. I've seen some standard libraries that use things like the gcc __builtin_expect construct to give CPU branch prediction hints: no matter how smart you might be, how many people will optimize their own hashtables to that sort of extent?
Well, I got a lot of (possibly deserved) flaming for that post :) I never said theory was useless though, just that - like many university degrees - they seem to equip you poorly for the real world. Note I said poorly, not that they don't equip you at all.
Mind you, I have a somewhat vocational bent by nature I guess. I'm the sort of person who can't really see the point of studying, say, History for 3 years (apart from to just spend time maturing at university of course).
On the science vs engineering thing, I'd normally agree but it seems that at least in England virtually all universities conflate Software Engineering and CompSci to such a high level that there's little reason to choose one over the other. Worse, most employers apparently see SE as a "softer" subject despite in many courses it being made up of exactly the same modules +/- a few, so the end result is that a ton of people do CS and virtually none do software engineering. It works in reverse too - one of the first modules we did in my CS course was "Software Engineering" which dealt with processes, evaluation cycles etc.
That's definitely a confused state of affairs, but realistically there's a huge job market for software engineers and a very small one for computer scientists so it's not surprising that the courses got merged over time.
OS Design? Fascinating, but ultimately irrelevant for 99% of coders. Implementing your own hashtables? Useful to gain an insight into how they work, but virtually any development platform people work in except raw C these days will provide highly tuned and optimized hashtable implementations. If anything you shouldn't roll your own, as it'll make your code less readable, more bloated and probably slower.
But it seems looking at the course material that important topics for writing real world software is simply not taught. Where are the lectures on writing internationalized code (think character encodings, flippable UI design etc). Where are the lectures on writing code with low startup overhead? What about teaching people the merits of various toolkits? Accessibility? Version control theory? (to be fair I think this is talked about at some point on my course, but I know some that don't)
It's possible for a student to come out of a CompSci course and be unable to explain why a containment-based widget toolkit is better than a positional one, yet can talk about VM swapping algorithms and implementing the fastest hashtable probing all day. Which would you rather employ?
People get really hung up on cross platform apps - Java was designed to do that but ended up simply inventing a new platform and hauling it around everywhere, .NET was not designed to do that and so Windows .NET apps are easily distinguishable from Linux .NET apps.
He's probably referring to System.Windows.Forms which is just a (leaky) wrapper for the Win32 GUI toolkit. Win32 is not object oriented so you can't really usefully subclass the widgets, so they're declared final.
Special software? I'm really not convinced the Citrix/VMware thing will fly - almost any company of a size that would find desktop Linux interesting will have multiple custom or specialist apps deployed. This is especially true of things like some popular HR apps which despite being "web apps" tend to have bizarre dependencies on Internet Explorer and particular JVMs, etc - basically I find it hard to imagine that a companies employees would be satisfied with running Windows in a box.
Even if VMware, Inc find some way to hack window management a la Classic-mode in OS X into things, the end result is still going to be cludgy, slow and require actually running a full copy of Windows. VMware/Citrix is great for running Windows in a box, but as a corporate migration solution? I think not.
I've never ever heard anyone talk about fair use outside of Slashdot, period. For most people it just isn't a big deal.
Making a DRM system that works with fair use but still protects artists is really hard, probably impossible. Apples DRM sort of gets there by being weak and easily exploited, but I'm not sure that's really an answer. It's a solution by being half-arsed.
It makes me wonder if the whole system of copyright is rather broken, to be frank. But I don't know of a better way, so I can't really criticize too much.
It's pretty depressing that this was modded down (overrated + flamebait). It's overly long, but basically the points raised by the AC are salient ones. I too have seen people justify piracy like this. It's pretty bad.
On modern desktops, Arabic works just fine.
Yeah, but the Linux Counter is pretty useless - I know a ton of people who use Linux and haven't registered, mostly either because they don't know about it or can't be bothered filling out their details simply to be counted.
http://bylands.dur.ac.uk/~mh/egypt-installfest/ind ex.html
Last time I checked, FreeBSD had got a few test cases out of it, a few minor bugfixes and that's it
That's pretty lame dude. Not saying other people don't do it, but it's still pretty lame.
One thing that is painfully clear about this thread is that the majority of posters have not actually used the latest version, as many of their criticisms have been fixed there. To go beyond that and claim the Gimp developers don't listen to the "professions" is, to be frank, total garbage.
Yes, there are performance problems with Notes (6.5.1 at any rate) on Wine, but they are not severe. I find it perfectly usable.
Regardless, you cannot draw conclusions based on your experiences from one (1) app run under Wine to all apps. I also run Office XP on Wine and it runs just as fast as it does on Windows. The same is true of Photoshop.
They are different programs, work in different ways, and use different codepaths. It's not surprising they run at different speeds.
Crazney has broken the pointless encryption on streaming things in the iTunes library to other machines on a LAN.
It has nothing to do with iTMS. Repeat after me: it has nothing to do with iTMS.
The encryption on streaming tunes between clients only serves two purposes: to try and keep people on the Apple upgrade treadmill and to force people to use iTunes on all their machines if they want to stream music between them from the iTunes library. This is your own music we're talking about here, no copyright violations are taking place.
To be frank, Apple is taking the piss with this sort of encryption, and now the piss is being taken out of them. Too bad, but it has nothing to do with FairPlay.
So really this has nothing to do with hurting Apple, or not agreeing to a "EULA", and it has everything to do with Apple cynically attempting to manipulate network effects. Your brother sharing his CD collection on the home network using iTunes? You can't use WinAmp, WMP, RhythmBox, Muine or whatever to access that, you have to use iTunes too. Then when you share your music, it cascades onwards.
This is especially true in places like homes, student flats and college networks, like the ones crazney is on. Really, Apple have no excuse for this: restricting DAAP can only have one goal and that is to use peer-power of the type that keeps Windows entrenched to give iTunes an upper hand. As such it frankly deserves to be cracked.
I know crazney. He's a good guy. We talk often - he isn't out to screw Apple or steal music. He wants to play the music on his Mac laptop using the iTunes streaming system: this seems totally fair to me.
The main problem is that I don't think anybody has really experimented with using Glade for anything other than GTK widget heirarchies. The other problem is that GObject isn't really a match for the .NET object model currently. With some work it could be a fantastic middle ground between what COM was and what .NET is - you still gain the advantage of actually using the real platforms instead of "skins" (see CPython vs Python.NET sometime), but you can still interop with other platforms at the OO level.
The main problem with sucking Glade files through SOAP/XML-RPC is that you'd have to relay all the signals to an instance on the server. GTK+ wasn't really designed for high latency feedback like that. It *could* work, but it probably wouldn't work well.
So there seem to be several alternatives:
1) Use Mono with our own XAML/Avalon implementation. .NET already has code access security and other things you want for sandboxing, though given the mess that Java applets/security was I'm not sure it's necessary a good thing. Pros: compatibility, Cons: we're chasing a standard we don't have any influence over so will always be behind.
2) Improve Mozilla/XUL - some previous poster suggested fully documenting it, integrating it with Apache/Tomcat, writing corporate webapps in it etc - YES. These are excellent suggestions for how to improve XUL as an app delivery platform. I used to be quite into XUL and knew the creator of JabberZilla : the things you could do with it even in the pre 1.0 days were mind boggling. XUL may never be a good way to write client side apps, but for internally deployed webapps it's fantastic.
3) Leverage current Linux technologies to produce something like ActiveX but with security. I was thinking about that this morning. If you have an SELinux/DBUS enabled system you can get quite good sandboxing even for native code. You'd want some kernel mods to prevent certain "attacks", but I think you can get pretty close to what .NET CAS provides except you can also reuse the entire free desktop platform as well (all the libraries that don't have .NET bindings for instance).
You could then layer some simple technologies on top of the web browser to write web-deployed client-side apps using GTK/Glade or Qt or whatever.
In other words, it'd have all the good bits of ActiveX, but without the bad bits. I'm not sure this is a good way to go though - ActiveX/Java Applets seem to be a mostly dead idea, except in a few old-skool corporate webapps.
Common misconception. Actually there is no "true geek" distro - at CodeWeavers which is staffed almost entirely by geeks, you'll see everything from Debian Testing (the CTO), to Slack to bleeding edge Fedora/SuSE releases. They are all Linux, after all.
Apparently, iTunes does actually work if you install it on Windows then copy it (and the registry entries it creates) across, though it's not really perfect.
Anyway, it's pretty high up on the CodeWeavers wishlist, so expect to see us work on it at some point. Just don't expect it soon - iTunes isn't really a killer app and there are some nice native equivalents like Muine or RhythmBox available.
Last time I checked rhel was entirely under the GPL. You have to pay for support which seems to have confused a lot of people who conflated gratis with libre, but it's definitely free software.
No, windows and message queues (and other similar GDI/USER objects) are managed by the kernel in Windows.
Win32 is, in essence, just a 'personality' running on the NT core.
I guess you're thinking of the kernel32 -> ntdll mapping.
I believe this could work. It would not, however, remove the need to own a copy of Windows, so its use would be more for the pragmatists ('we want it to work') than the idealists ('we want it to be free')
It could work, though it'd be extremely difficult to do. It's also almost certainly not what they've done (or haven't done) as their site says you don't need a copy of Windows when using David.