He doesn't want to use a piece of software that is so incredibly inconsistent that there is NO ONE WAY TO SET UP A NETWORK INTERFACE
No offence, but it seems logic isn't your boyfriends strong point. This argument could be applied just as well to Windows (the way you set up network interfaces changed between win3.1 -> win98 -> win2000 -> winXP), or MacOS 9 -> X.
What you say? They are just different versions, and that's OK, but multiple distros aren't? It all boils down to the same thing at the end of the day, some computers working differently to others.
He's not willing to touch an emulator.
Why not, what's wrong with emulation? If he's atechnical, why does he care about such a technical detail?
Mac OS X has already done more for open-source software in the real world of Joe Sixpack than Linux (and even *BSD) will ever do, in my opinion. I could be wrong, but that's how I feel on the issue.
I think you are wrong. You know how many people use MacOS X? Virtually none. Check out Apples own figures on this. You know how much free software was developed for MacOS X then ported to other platforms? Virtually none.
Sorry to bitch, but I get tired of people saying "Linux will never go anywhere unless it does A, B and C but oh yeah MacOS r0x0rs". A lot of the things you were saying were being said about the server side only a few years ago, including the multi-distro red herring. Yet it seems to be doing OK anyway.
Linux will get there, on the desktop, eventually. It will take a long time, but it's shifting massive inertia. It will involve a lot of work, but everybody knows that. I absolutely believe it will happen - every single time the Linux developer community has hit a major problem, whether it was usability, compatability, ease of install, artwork and so on, solutions have appeared and been implemented despite the naysayers saying that it could "never be done".
I don't think open source software needs to be "sold" to the general public. There's no real hurry. It'll get there eventually.
If I had to say what would make the biggest difference though, I'd say clever advertising campaigns. With a series of adverts, you can quickly communicate even complex ideas to people. Combined with an enthusiastic community that makes itself heard, that would be a good way IMHO to get the word out. And that's what it's all about really, if people understand that it is free, but still good, and even why it's free, selling it becomes so much simpler.
Sounds like a troll to me. You seriously expect us to believe that ALL the Linux users at your uni only used it because they couldn't afford or wouldn't warez Windows?
I don't know any students who use Linux because Windows is too expensive. If they are into it they use it, otherwise they get XP off Kazaa.
It's more likely they are simply keeping tabs on the competition - I really cannot believe they'd modify their own products to work well with Samba. Can you?
Thanks. I know crazney (from transgaming) was complaining about how gcc makes such large object files - it was taking a game they were working on 20 minutes just to link! - so that all adds up.
I must admit I never really thought about object size, nor have I done any comparisons. Most of the projects I've compile with it have been relatively small however (largest is about 1.5-2mloc) so I suppose it becomes more important the larger the project you are compiling.
Where I work there are several networks, the global corporate (classified, it's a list X company) network and then a group of disconnected unclass nets. Quite a few of us here run Linux on the unclass nets, and that's cool. Can't do it on the classified net, they run scans constantly for unknown software.
If there were a competitive FOSS compiler available, I'm sure that some groups would use it. There isn't. We've certainly had teams do comparitive analyses.
That's interesting. In typical comparisons I've seen gcc comes out on top usually, especially these days with regards to C++ support. What is it lacking? Support for precompiled headers?
Hardware accelerated user interface
xp doesn't have it
Windows has used hardware acceleration for years - or was I just dreaming when I saw the "hardware acceleration" slider in the control panel?
Don't confuse "wierd abuses of OpenGL" with hardware acceleration. All modern operating systems use hardware acceleration.
You can play "my OS has this and yours doesn't ner ner ner" all day. XP is chock full of features that MacOS X doesn't have - like being able to run the vast majority of the worlds websites and unprecedented website and hardware compatability. For most people that's way more important than a fast PDF viewer.
They seem pretty active to me. Then there's all the stuff that doesn't get published, like Detours.
Of course just because some researcher invents something doesn't mean it ever gets implemented. I once read a review with an MS researcher - she said they sent reports off to the IE product team, but never expected to hear anything back from them.
Re:What about the postscript desktop?
on
OpenGL 1.5
·
· Score: 3, Interesting
Yes, all that is absolutely correct. I'd just like to point out that you don't really need to use OpenGL to do those things though, at the end of the day the 2D/3D primitives all boil down to a sequence of instructions to the card. For instance X pixmaps (on XFree) are already stored in video RAM.
The main problem with XFrees responsiveness is not whether it uses OpenGL or not (which ultimately makes little difference) but how it interacts with applications and how it pokes the video card. For instance very few drivers fully accelerate RENDER (which is 2D hardware acceleration for alpha channel blending and some other things). That means you end up doing very slow framebuffer reads, compositing in software then upload. I guess part of the reason for using OpenGL was to work around the reluctance of driver manufacturers to write specialist fully optimized drivers for their hardware.
Not to mention that most apps are very slow at processing Expose events. There has been talk of doing what MacOS does here and having apps directly render client-side into a compressed backing store.
I was going to be a smart ass and give a link to the license, as I thought it would be on the first page on Google for "Windows XP license". Well, I'm on page 5 and not found it yet. I know there is a site out there that collects these things, but it's kind of surprising that it's so hard to find a copy (unless you have XP installed and running in which case you already accepted it).
Yeah, but you're assuming everybody would have to do this. If companies get the message that they'll be taken to court over it, they'll cave faster in future then eventually just refund those who ask for it anyway.
If you mean the "FreeBSD" userspace, then maybe, but I've yet to see anybody argue convincingly that MacOS X is actually "freebsd with stuff on top". It has a different kernel, different IO layer, the FreeBSD de-facto graphics layer is X11 not Quartz, and so on. Calling MacOS FreeBSD is like calling Windows with Cygwin Linux. I think this guy knows the difference;)
I should note here that the Borland Object Pascal language has dealt with nearly all of the problems presented in the paper, and in many respects is a wonderful language to work in. I know I certainly prefer most parts of it to their C equivalents. The syntax is clear and readable, typecast scoping is clearer, it has powerful built in types (nobody uses arrays of chars for strings and hasn't for years, there is a builtin string type).
Pascal may be dead, but in Windows-land at any rate its child, Object Pascal lives on. And if you compare how Object Pascal evolved from Pascal to how C++ evolved from C, Pascal wins every time.
The other neat thing (from one perspective, bad from another) is that Pascal was essentially abandoned to Borland. Lacking the need to standardise it, they added many many useful abilities into the language that make it an ideal Win32 workhorse. Unlike MSVC++, when it would have led to better integration with Windows Borland extended it, for instance IDispatch is integrated into the language seamlessly, so you can do:
var
WordDocument:OleVariant;
begin
WordDocument:= CreateOleObject("Word.Document");
WordDocument.SomeMethod();
end;
ie in order to support COM instead of messy cludges that make you do things manually support for late bound object methods were added to the language.
Of course that means you were essentially tied to one companies compiler, but nonetheless that doesn't make the language any less nice.
That's fine as far as it goes, but once you start using hungarian notation it's a slipperly slope and you end up with stuff like LPVOID in an attempt to remain consistant.
Alternatively, how about parameter prototypes like "LPUNKOUTER ppv" - which is what, exactly? Well, I can tell you that's it's a pointer to an outer unknown - oh but wait, an Unknown is also just a pointer, hence the "ppv" bit - pointer to pointer to void, geddit? But what is the parameter actually for? Not a clue;)
int foo, *bar;
creates an integer named foo, and a pointer to an int named bar. Right? Or am I wrong?
What do you mean by the order of type modifiers? Can you give an example?
What I like is the way you can declare a variable const (ie, it's not constant and should not change) yet still assign to it and get away with merely a compiler warning.
Gecko/Moz has a lot more features. We embed IE in an app here at work because it has features we need. It's possible we might move to Gecko/Moz in the future, but KHTML is out of the question - we need things like in place editing, simple licensing, and behaviours/XBL.
And, to tie into this article, the ports collection provides a wealth of great software. There's no issue as to which flavor of Linux you have
FreeBSD is cool and all, but you could equally say "Debian apt is really great! there's no issue as to which CPU arch you're on, if you're running Debian, there will be packages for you".
ie, don't overhype ports. Useful yes, something amazing that only FreeBSD has, no.
As for IE, even 1.0 was written in-house although some code used in it was licensed from NSCA Mosaic. Of course, that code was also included in virtually every other browser on the planet so, again, you don't have a point with this one.
By "licensed" you mean they promised to give SpyGlass a share of the profits, but obviously didn't tell them they were going to give it away for free -> hence, no money for SpyGlass.
There is a known bug in Mozilla (known for AGES now) where it cannot copy'n'paste large amounts of text, due to a screwup with their implementation of the X clipboard protocols. Unfortunately the guy who "owns" that module, has never done anything with it. The bug still remains unfixed.
Yes, there are other client side apps, but none that I know of on Linux. How many people use LimeWire on Linux? Not many. I can't think of any other mainstream apps.
There are a few Mono apps floating about (like Dashboard, but a few others too). There aren't many yet, but then Mono is a lot newer.
Yes, I personally would consider a wrapper toolkit like SWT to be not quite as good (if you aren't concerned about super portability) as using GTK directly. Like the original AWT, wrapper toolkits tend to be rather unpredictable, and have other strange problems that only show up on one platform etc. All the major cross platform toolkits (GTK, Qt, Swing) are "emulating".
I have heard of Kazaa, but last time I checked that was not written in Java. Java/GTK is not developed by IBM, and SWT is not GTK - it's a wrapper/abstraction library with all the wierdness, loose semantics and pain that implies. I'm not saying SWT is bad, but I'm saying that it's not a great choice for free software designed primarily to run on Linux, where you might as well just use GTK itself.
Slashdot Mirror
No offence, but it seems logic isn't your boyfriends strong point. This argument could be applied just as well to Windows (the way you set up network interfaces changed between win3.1 -> win98 -> win2000 -> winXP), or MacOS 9 -> X.
What you say? They are just different versions, and that's OK, but multiple distros aren't? It all boils down to the same thing at the end of the day, some computers working differently to others.
He's not willing to touch an emulator.
Why not, what's wrong with emulation? If he's atechnical, why does he care about such a technical detail?
Mac OS X has already done more for open-source software in the real world of Joe Sixpack than Linux (and even *BSD) will ever do, in my opinion. I could be wrong, but that's how I feel on the issue.
I think you are wrong. You know how many people use MacOS X? Virtually none. Check out Apples own figures on this. You know how much free software was developed for MacOS X then ported to other platforms? Virtually none.
Sorry to bitch, but I get tired of people saying "Linux will never go anywhere unless it does A, B and C but oh yeah MacOS r0x0rs". A lot of the things you were saying were being said about the server side only a few years ago, including the multi-distro red herring. Yet it seems to be doing OK anyway.
Linux will get there, on the desktop, eventually. It will take a long time, but it's shifting massive inertia. It will involve a lot of work, but everybody knows that. I absolutely believe it will happen - every single time the Linux developer community has hit a major problem, whether it was usability, compatability, ease of install, artwork and so on, solutions have appeared and been implemented despite the naysayers saying that it could "never be done".
I don't think open source software needs to be "sold" to the general public. There's no real hurry. It'll get there eventually.
If I had to say what would make the biggest difference though, I'd say clever advertising campaigns. With a series of adverts, you can quickly communicate even complex ideas to people. Combined with an enthusiastic community that makes itself heard, that would be a good way IMHO to get the word out. And that's what it's all about really, if people understand that it is free, but still good, and even why it's free, selling it becomes so much simpler.
I don't know any students who use Linux because Windows is too expensive. If they are into it they use it, otherwise they get XP off Kazaa.
Those looks like servers to me - the previous poster was talking about the students own machines.
It's more likely they are simply keeping tabs on the competition - I really cannot believe they'd modify their own products to work well with Samba. Can you?
I must admit I never really thought about object size, nor have I done any comparisons. Most of the projects I've compile with it have been relatively small however (largest is about 1.5-2mloc) so I suppose it becomes more important the larger the project you are compiling.
That's like saying capitalism is just a way of allocating resources - an oversimplification, in other words.
Not for interop purposes though, as far as I'm aware. Obviously, you cannot use code gained from disassembly in your own products.
Where I work there are several networks, the global corporate (classified, it's a list X company) network and then a group of disconnected unclass nets. Quite a few of us here run Linux on the unclass nets, and that's cool. Can't do it on the classified net, they run scans constantly for unknown software.
That's interesting. In typical comparisons I've seen gcc comes out on top usually, especially these days with regards to C++ support. What is it lacking? Support for precompiled headers?
xp doesn't have it
Windows has used hardware acceleration for years - or was I just dreaming when I saw the "hardware acceleration" slider in the control panel?
Don't confuse "wierd abuses of OpenGL" with hardware acceleration. All modern operating systems use hardware acceleration.
You can play "my OS has this and yours doesn't ner ner ner" all day. XP is chock full of features that MacOS X doesn't have - like being able to run the vast majority of the worlds websites and unprecedented website and hardware compatability. For most people that's way more important than a fast PDF viewer.
They seem pretty active to me. Then there's all the stuff that doesn't get published, like Detours.
Of course just because some researcher invents something doesn't mean it ever gets implemented. I once read a review with an MS researcher - she said they sent reports off to the IE product team, but never expected to hear anything back from them.
The main problem with XFrees responsiveness is not whether it uses OpenGL or not (which ultimately makes little difference) but how it interacts with applications and how it pokes the video card. For instance very few drivers fully accelerate RENDER (which is 2D hardware acceleration for alpha channel blending and some other things). That means you end up doing very slow framebuffer reads, compositing in software then upload. I guess part of the reason for using OpenGL was to work around the reluctance of driver manufacturers to write specialist fully optimized drivers for their hardware.
Not to mention that most apps are very slow at processing Expose events. There has been talk of doing what MacOS does here and having apps directly render client-side into a compressed backing store.
I was going to be a smart ass and give a link to the license, as I thought it would be on the first page on Google for "Windows XP license". Well, I'm on page 5 and not found it yet. I know there is a site out there that collects these things, but it's kind of surprising that it's so hard to find a copy (unless you have XP installed and running in which case you already accepted it).
Yeah, but you're assuming everybody would have to do this. If companies get the message that they'll be taken to court over it, they'll cave faster in future then eventually just refund those who ask for it anyway.
If you mean the "FreeBSD" userspace, then maybe, but I've yet to see anybody argue convincingly that MacOS X is actually "freebsd with stuff on top". It has a different kernel, different IO layer, the FreeBSD de-facto graphics layer is X11 not Quartz, and so on. Calling MacOS FreeBSD is like calling Windows with Cygwin Linux. I think this guy knows the difference ;)
Pascal may be dead, but in Windows-land at any rate its child, Object Pascal lives on. And if you compare how Object Pascal evolved from Pascal to how C++ evolved from C, Pascal wins every time.
The other neat thing (from one perspective, bad from another) is that Pascal was essentially abandoned to Borland. Lacking the need to standardise it, they added many many useful abilities into the language that make it an ideal Win32 workhorse. Unlike MSVC++, when it would have led to better integration with Windows Borland extended it, for instance IDispatch is integrated into the language seamlessly, so you can do:
var WordDocument :OleVariant;
begin
WordDocument := CreateOleObject("Word.Document");
WordDocument.SomeMethod();
end;
ie in order to support COM instead of messy cludges that make you do things manually support for late bound object methods were added to the language.
Of course that means you were essentially tied to one companies compiler, but nonetheless that doesn't make the language any less nice.
Alternatively, how about parameter prototypes like "LPUNKOUTER ppv" - which is what, exactly? Well, I can tell you that's it's a pointer to an outer unknown - oh but wait, an Unknown is also just a pointer, hence the "ppv" bit - pointer to pointer to void, geddit? But what is the parameter actually for? Not a clue ;)
Well, define "expressive" first, then maybe people will be able to give better answers....
int foo, *bar; creates an integer named foo, and a pointer to an int named bar. Right? Or am I wrong?
What do you mean by the order of type modifiers? Can you give an example?
What I like is the way you can declare a variable const (ie, it's not constant and should not change) yet still assign to it and get away with merely a compiler warning.
Gecko/Moz has a lot more features. We embed IE in an app here at work because it has features we need. It's possible we might move to Gecko/Moz in the future, but KHTML is out of the question - we need things like in place editing, simple licensing, and behaviours/XBL.
FreeBSD is cool and all, but you could equally say "Debian apt is really great! there's no issue as to which CPU arch you're on, if you're running Debian, there will be packages for you".
ie, don't overhype ports. Useful yes, something amazing that only FreeBSD has, no.
By "licensed" you mean they promised to give SpyGlass a share of the profits, but obviously didn't tell them they were going to give it away for free -> hence, no money for SpyGlass.
There is a known bug in Mozilla (known for AGES now) where it cannot copy'n'paste large amounts of text, due to a screwup with their implementation of the X clipboard protocols. Unfortunately the guy who "owns" that module, has never done anything with it. The bug still remains unfixed.
Yes, there are other client side apps, but none that I know of on Linux. How many people use LimeWire on Linux? Not many. I can't think of any other mainstream apps.
There are a few Mono apps floating about (like Dashboard, but a few others too). There aren't many yet, but then Mono is a lot newer.
Yes, I personally would consider a wrapper toolkit like SWT to be not quite as good (if you aren't concerned about super portability) as using GTK directly. Like the original AWT, wrapper toolkits tend to be rather unpredictable, and have other strange problems that only show up on one platform etc. All the major cross platform toolkits (GTK, Qt, Swing) are "emulating".
I have heard of Kazaa, but last time I checked that was not written in Java. Java/GTK is not developed by IBM, and SWT is not GTK - it's a wrapper/abstraction library with all the wierdness, loose semantics and pain that implies. I'm not saying SWT is bad, but I'm saying that it's not a great choice for free software designed primarily to run on Linux, where you might as well just use GTK itself.