Proprietary software may be ok for applications in some situations, but it's really horrible when the core system software isn't available. You have problems that are simply unfixable - and break all applications too!
his is the problem with closed source ones, of course. They can't be supported by anyone except the company that made them, so they'll break whenever the kernel changes in a way that's incompatible or simply exposes a bug in the driver
OK, a question then. How are drivers developed for Windows? The windows source is closed and I doubt that device manufacturers have access to all of the code for the various microsoft kernels. What is MS providing to third parties to enable them to write device drivers without disturbing the OS core that Linux couldn't also provide?
I'm not trolling and I am a "layman" in these matters, these are genuine questions.
Yeah, me and Tone have been discussing that - we'd like to promote your girlfriend - we just need to find a way of promoting figments of peoples imagination...:)
the kernel developers want to be able to change their internal APIs at will. It's usually not a problem, because they can then change all the drivers to match, because they're right there in the source tr
Hmm. Still not convinced, but as I said, I know nothing about low devel device and kernel driver design/development.
I would have thought that the manufacturors of hardware devices would be able to produce and maintain the best software drivers for their kit.
But if they have to stick to a stable internal API, let alone an internal ABI, they might get stuck with badly-designed APIs
Different versions of the kernel breaking stable drivers, rather than the other way around? Irony abounds!
This has been more than paid for by the thousands of NVIDIA cards sold on the back of supposed "Linux support".
Well, this is precisely why, I and many others, buy nVidia cards. I use a dual boot Win2K/Mandrake box and I know I'm getting excellent drivers for each operating system.
I personally don't see any problems with closed source binary drivers. The kernel provides (or should provide) interface abstraction to a sensible level where it's possible for safe "black box" drivers to be written.
Then again, I don't understand anything about writing device drivers or how tightly integrated to an os kernel they need to be. As a layman, I would have thought that the kernel provides certain services to a driver module which are accessable through an interface?
Card manufacturers have trade secrets to protect - that's the nature of the world we live in. No matter how much you or I may wish that drivers are all open source, if an opensource driver reveals details about the hardware, manufacturers are not going to want to opensource the software.
"Is it really the government's job to decide what's best for the nation's youth after dark?"
Ironically, here in the UK, the problem seems to be the opposite - how to keep the kids off of the streets after dark. There's even legislation for so-called "anti-social orders" which , AFAIK, force a curfew onto kids making sure they're not hanging around on the streets at night in large "intimidating" groups. Trouble is, these orders rarely used because they're very difficult to enforce.
I wouold have thought that giving young people something constructive to do in the evenings (like gaming) was a Good Thing.
I just knew that some idiot would start trolling about KDE vs Gnome.
The reason Gnome has commercial support is purely and simply due to there being a company producing a commercial version of Gnome back when Sun and HP etc were looking at Linux Desktops.
Yada Yada Trolltech, QPL/GPL, C++, yada yada.
Don't worry about what anyone says. The reason Gnome was chosen was because of Helixcode, pure and simple. Sun and the rest are businesses and as we all ought to know, business deals with business. If Trolltech were producing a commercial KDE, you would have seen something very different happening.
As for Ximian being the future of SuSE and KDE being "legacy" - be afraid, be very afraid. Novells only interest in Ximian was MONO, which happens to fit their new Linux story very well.
Go over to go-mono.org and read Miguels report on the recent Microsoft Professional Developer Conference. Look for references to XAML and other plans Microsoft have for Longhorn. Check Miguels assessment of what this means for non-Microsoft desktop Operating Systems. Then check his "solution" to this.
Once you've done that, come back here and tell me with a straight face that Ximian gnome as the standard Linux desktop is a Good Thing.
You, my friend, have obviously never had the "pleasure" of working with Forte.
I have actually - two years worth back in 1997 through 1999 for a CRM company in the UK.
I agree that by mordern standards the IDE is dated - but I disagree strongly about everything apart from app partitioning being an "amateurish piece of shit". In fact, in my experience, I'd have to say the exact opposite. The whole thing was so tightly integrated that everything just worked.
Ok, so we ran into some problems with repository corruptions and a few integration issues, but these were easily surmountable in the first instance by making regular backups and usuing a versions of FTEXEC which were known not to have showstopping bugs and in the second by simple management.
Not sure what your performance issues were, but we quickly discovered that in order to make the app scale to the 200+ desktops it was to run on, we had to implement a seemingly convoluted design pattern using DataFactories, Data Access Controllers and a mechanism for managing the database connections and minimising network traffic.
We also had to provide object level transaction management - easily impemented using inheiritance and interfaces.
From what I remember, if the application wasn't designed to be efficient, then you would see problems if you tried to scale it beyond 5 - 10 machines.
From a language standpoint, the "plumbing" was indeed hidden from you. It was absurdly easy to talk to objects cross process or cross machines
Exactly. This is the lesson I'm hoping Sun has learned with this Rave product. Unfourtanately, currently, to produce really efficient J2EE apps requires knowledge of how the EJBs etc work.
BTW, it's nice to meet someone on slashdot who knows that Forte isn't Netbeans!
Could Sun finally have seen the light? Back in 1999, Sun purchased an Enterprise Software company called Forte Software. Forte (not to be confused with the Netbeans rebrand) was an application suite which achieved what J2EE achieves now - but completely painlessly.
All of the plumbing was hidden from developers, leaving them free to concentrate on business logic. Forte shipped with a complete Application Framework and its own language the Transactional Object Oriented Language(TOOL).
Basically (to cut a long story short) Java looked as if it had more potential at the time, so Forte was rebranded to Sun ONE Unified Development Server and allowed to wither. It's officially being end-of-lined by Q1 next year.
The point here is that this Project Rage seems very much like Unified server - but it works in Javaland. It (hopefully) hides all the plumbing of a J2EE application from developers, allowing them to concentrate on business logic. If it's more than Suns version of Eclipse, then it'll certainly be a product to watch. I hope Sun get it right this time and that it's not too late.
Where this leaves IBM and Weblogic remains to be seen - unless this Rage integrates with their app servers. It ought to - seamlessly of course...
:)
Consultant:We have finished our audit, and our conclusion is that it would be cheaper for Newham Council to use Open Source Software.
MS rep: What??
Consultant:Yes, the situation is quite clear. It would be cheaper to use OSS.
MS rep:Oh! We weren't expecting that!
Newham rep:This meeting has been quicker than I thought it would be. Shall we go to the pub for a lunch-time pint?
MS rep:Wait...here's a smegload of money we have from Bill's wallet. Will this suffice as a bribe to make it look better for us?
Newham rep:Yup! That should pay for enough beer to distort reality sufficiently!
Consultant:Duck and Drake or Kings Head?
Newham rep:Nah, Broken Spinner. Now, I've just got to dash to the toilet, so I'll see you both down there in 10 minutes. [to MS rep] Get a tab started will yah...
but it just confirms the ACs point in that the blame shouldn't be thrust entirely on X11's shoulders
And therein lies the problem. X11, to most people, means the entire usable desktop environment - in much the same way that "Linux" means the kernel, GNU tools, X11, toolkits and Desktops.
The underlying bare bones X11 system may well be very quick, but to be useful, it needs as a bare minimum, a window manager. To be useful it requires the toolkits and desktop environment.
I'm no software engineer, but it seems to me that given modern GUIs are all about, something could be done to the core X11 system to facilitate faster execution of window managers and toolkits.
Of course, if the toolkit/wm process is completely seperate from X11 and X11 spends alot of idle time waiting for the toolkit/wm to do stuff, then why not move some of that "stuff" into the X11 server?
What most people experience as "X11 bloat" currently probably consists more of bloat on the widget toolkit side than on the XFree86 side.
Why do people always wheel this argument out in defence of X? We all know that X11 doesn't include the "bloat"ed widget toolkits and desktop environments.
However, that said, X11 is about as useful to most people as a car which consists of just the chassis and engine. Unless the toolkits and desktop environments (body, seats, steeringwheel etc) are added it's not particularly useable or useful to the typical Linux desktop user.
Well, clearly, it's ET, isn't it? Those spots are his eyes. Every time he blinks, the spots vanish.
Who knows, he and Drew Barrymore may be making sweet..er...music out there. Let's hope ET has his cell phone with him - with all this attention, things could get out of hand...
Bah... I could go on but mozilla is for geeks right now.
Why does somebody have to say this every time a piece of software from the open source community is evaluated as a replacement for a piece of proprietary software?
Why? For the same reason that 30%-40% of posts in a discussion about XFree86 are about it needing to be reworked or replaced.
I think IE loses the browser domination the day Sony ships the Playstation 3 (with a non-IE browser, which will probably be Mozilla).
Well, that depends on how visible the browser is.
OK, I accept that a greater number of hits by a Moz based browser would force sites (banks especially) to support Moz, but in the wider context of this discussion - which is about branding and Moz having a strong brand identity - this isn't necessarily true.
If Sony integrate the browser correctly, the only brand which will be visible is the Sony brand - PS3 users won't care what they're surfing with, all they'll care about is that they are surfing. Essentially, the browser vanishes.
Microsoft have picked up on this too - hence the line that IE6 is the last browser all future browsing capability being integrated with Windows. IE will vanish, but browsing will be accessible from explorer/outlook/word or however they implement it.
From the "Joe Sixpack" of view, having a "browser" as a seperate application begins to look like an optional and unecessary extra and a strong brand just serves to emphasise its "seperateness".
Read the Article on Saturday - was surprisingly good.
The article mentions that Acornsoft, the original publishers, didn't want to pay royalty rates as high as Braben and Bell wanted, so they retained publishing rights enabling them to sell those rights at auction to BTSoft, leading to Elite on other platforms.
IMO, by far the best version produced was for the Acorn Archimedes in 1992 - circa 10 years after the original.
Many happy hours as an undergraduate whiled away back then...
This topic reminded me of an article in the print version of UK mag PC Pro by Jon Honeyball. He tells of one of his (consultancy) "clients" looking for a SAN/RAID solution from Dell and EMC, which would cost 120,000.
Aparently, Apples offering was 2TB storage for 9000 - vs 80,000 for 2TB from Dell/EMC.
In the article he says;The obvious question is whether you can put fibre channel cards into your Windows servers and connect them up too. Apple tells me this is possible The idea being to use the storage for SQL*Server databases and the like.
I'm not surprised this is such a good solution for use with Linux.
Dr Miguel Nicolelis knew he had nailed it when the monkey stopped using her arm to play the computer game.
Should this story be in the Ximian section? And isn't it "Dr Miguel d'incaza"? Also, I didn't know the Ximian monkey was female...
Slashdot Mirror
Proprietary software may be ok for applications in some situations, but it's really horrible when the core system software isn't available. You have problems that are simply unfixable - and break all applications too!
his is the problem with closed source ones, of course. They can't be supported by anyone except the company that made them, so they'll break whenever the kernel changes in a way that's incompatible or simply exposes a bug in the driver
OK, a question then. How are drivers developed for Windows? The windows source is closed and I doubt that device manufacturers have access to all of the code for the various microsoft kernels. What is MS providing to third parties to enable them to write device drivers without disturbing the OS core that Linux couldn't also provide?
I'm not trolling and I am a "layman" in these matters, these are genuine questions.
Yeah, me and Tone have been discussing that - we'd like to promote your girlfriend - we just need to find a way of promoting figments of peoples imagination... :)
the kernel developers want to be able to change their internal APIs at will. It's usually not a problem, because they can then change all the drivers to match, because they're right there in the source tr
Hmm. Still not convinced, but as I said, I know nothing about low devel device and kernel driver design/development. I would have thought that the manufacturors of hardware devices would be able to produce and maintain the best software drivers for their kit.
But if they have to stick to a stable internal API, let alone an internal ABI, they might get stuck with badly-designed APIs
Different versions of the kernel breaking stable drivers, rather than the other way around? Irony abounds!
I'm Secretary of State for Health, be careful what you say...
This has been more than paid for by the thousands of NVIDIA cards sold on the back of supposed "Linux support".
Well, this is precisely why, I and many others, buy nVidia cards. I use a dual boot Win2K/Mandrake box and I know I'm getting excellent drivers for each operating system.
I personally don't see any problems with closed source binary drivers. The kernel provides (or should provide) interface abstraction to a sensible level where it's possible for safe "black box" drivers to be written.
Then again, I don't understand anything about writing device drivers or how tightly integrated to an os kernel they need to be. As a layman, I would have thought that the kernel provides certain services to a driver module which are accessable through an interface?
Card manufacturers have trade secrets to protect - that's the nature of the world we live in. No matter how much you or I may wish that drivers are all open source, if an opensource driver reveals details about the hardware, manufacturers are not going to want to opensource the software.
Can't believe Ive just typed "mod parent up" but i have to. The most sensible observations that I've read on this topic.
as if your original post wasn't disgraceful enough, you continue to post this nonsense.
Perhaps your girlfriend sufferes from paranoid delusions?
Managers are the single cause of most of the inefficiency, coruption and waste in the NHS.
"Is it really the government's job to decide what's best for the nation's youth after dark?"
Ironically, here in the UK, the problem seems to be the opposite - how to keep the kids off of the streets after dark. There's even legislation for so-called "anti-social orders" which , AFAIK, force a curfew onto kids making sure they're not hanging around on the streets at night in large "intimidating" groups. Trouble is, these orders rarely used because they're very difficult to enforce.
I wouold have thought that giving young people something constructive to do in the evenings (like gaming) was a Good Thing.
I just knew that some idiot would start trolling about KDE vs Gnome.
The reason Gnome has commercial support is purely and simply due to there being a company producing a commercial version of Gnome back when Sun and HP etc were looking at Linux Desktops.
Yada Yada Trolltech, QPL/GPL, C++, yada yada.
Don't worry about what anyone says. The reason Gnome was chosen was because of Helixcode, pure and simple. Sun and the rest are businesses and as we all ought to know, business deals with business. If Trolltech were producing a commercial KDE, you would have seen something very different happening.
As for Ximian being the future of SuSE and KDE being "legacy" - be afraid, be very afraid. Novells only interest in Ximian was MONO, which happens to fit their new Linux story very well.
Go over to go-mono.org and read Miguels report on the recent Microsoft Professional Developer Conference. Look for references to XAML and other plans Microsoft have for Longhorn. Check Miguels assessment of what this means for non-Microsoft desktop Operating Systems. Then check his "solution" to this.
Once you've done that, come back here and tell me with a straight face that Ximian gnome as the standard Linux desktop is a Good Thing.
You, my friend, have obviously never had the "pleasure" of working with Forte.
I have actually - two years worth back in 1997 through 1999 for a CRM company in the UK.
I agree that by mordern standards the IDE is dated - but I disagree strongly about everything apart from app partitioning being an "amateurish piece of shit". In fact, in my experience, I'd have to say the exact opposite. The whole thing was so tightly integrated that everything just worked.
Ok, so we ran into some problems with repository corruptions and a few integration issues, but these were easily surmountable in the first instance by making regular backups and usuing a versions of FTEXEC which were known not to have showstopping bugs and in the second by simple management.
Not sure what your performance issues were, but we quickly discovered that in order to make the app scale to the 200+ desktops it was to run on, we had to implement a seemingly convoluted design pattern using DataFactories, Data Access Controllers and a mechanism for managing the database connections and minimising network traffic.
We also had to provide object level transaction management - easily impemented using inheiritance and interfaces.
From what I remember, if the application wasn't designed to be efficient, then you would see problems if you tried to scale it beyond 5 - 10 machines.
From a language standpoint, the "plumbing" was indeed hidden from you. It was absurdly easy to talk to objects cross process or cross machines
Exactly. This is the lesson I'm hoping Sun has learned with this Rave product. Unfourtanately, currently, to produce really efficient J2EE apps requires knowledge of how the EJBs etc work.
BTW, it's nice to meet someone on slashdot who knows that Forte isn't Netbeans!
Could Sun finally have seen the light? Back in 1999, Sun purchased an Enterprise Software company called Forte Software. Forte (not to be confused with the Netbeans rebrand) was an application suite which achieved what J2EE achieves now - but completely painlessly.
All of the plumbing was hidden from developers, leaving them free to concentrate on business logic. Forte shipped with a complete Application Framework and its own language the Transactional Object Oriented Language(TOOL).
Basically (to cut a long story short) Java looked as if it had more potential at the time, so Forte was rebranded to Sun ONE Unified Development Server and allowed to wither. It's officially being end-of-lined by Q1 next year.
The point here is that this Project Rage seems very much like Unified server - but it works in Javaland. It (hopefully) hides all the plumbing of a J2EE application from developers, allowing them to concentrate on business logic. If it's more than Suns version of Eclipse, then it'll certainly be a product to watch. I hope Sun get it right this time and that it's not too late.
Where this leaves IBM and Weblogic remains to be seen - unless this Rage integrates with their app servers. It ought to - seamlessly of course...
:)
Rave is based on NetBeans technology but they don't mention that in the article.
Actually, there's a sidebar titled "Enabling technologies" which lists netbeans as one of the enablers.
However, the importance of this tool isn't in the netbeans base, it's in the integration of the server side J2EE tech etc.
When I saw "Idiot" and "Howard" in the same sentence, I immediately thought of Michael Howard
Sorry...
Consultant:We have finished our audit, and our conclusion is that it would be cheaper for Newham Council to use Open Source Software.
MS rep: What??
Consultant:Yes, the situation is quite clear. It would be cheaper to use OSS.
MS rep:Oh! We weren't expecting that!
Newham rep:This meeting has been quicker than I thought it would be. Shall we go to the pub for a lunch-time pint?
MS rep:Wait...here's a smegload of money we have from Bill's wallet. Will this suffice as a bribe to make it look better for us?
Newham rep:Yup! That should pay for enough beer to distort reality sufficiently!
Consultant:Duck and Drake or Kings Head?
Newham rep:Nah, Broken Spinner. Now, I've just got to dash to the toilet, so I'll see you both down there in 10 minutes. [to MS rep] Get a tab started will yah...
but it just confirms the ACs point in that the blame shouldn't be thrust entirely on X11's shoulders
And therein lies the problem. X11, to most people, means the entire usable desktop environment - in much the same way that "Linux" means the kernel, GNU tools, X11, toolkits and Desktops.
The underlying bare bones X11 system may well be very quick, but to be useful, it needs as a bare minimum, a window manager. To be useful it requires the toolkits and desktop environment.
I'm no software engineer, but it seems to me that given modern GUIs are all about, something could be done to the core X11 system to facilitate faster execution of window managers and toolkits.
Of course, if the toolkit/wm process is completely seperate from X11 and X11 spends alot of idle time waiting for the toolkit/wm to do stuff, then why not move some of that "stuff" into the X11 server?
What most people experience as "X11 bloat" currently probably consists more of bloat on the widget toolkit side than on the XFree86 side.
Why do people always wheel this argument out in defence of X? We all know that X11 doesn't include the "bloat"ed widget toolkits and desktop environments.
However, that said, X11 is about as useful to most people as a car which consists of just the chassis and engine. Unless the toolkits and desktop environments (body, seats, steeringwheel etc) are added it's not particularly useable or useful to the typical Linux desktop user.
Why is xfree86 called xfree86
Does the 86 mean 1986? That's the connotation that always springs to my mind, in these days of "95","98","millenium","2000" and "2003"
You do know what a joke is, right?
Well, clearly, it's ET, isn't it? Those spots are his eyes. Every time he blinks, the spots vanish.
..er...music out there. Let's hope ET has his cell phone with him - with all this attention, things could get out of hand...
Who knows, he and Drew Barrymore may be making sweet
Mozilla has replaced my browser, and the browser of all my friends (all ten of them! woo!)
/. ...
What kind of friends have you got!? If I had 10 friends, I wouldn't need to be trolling on
Bah... I could go on but mozilla is for geeks right now.
:)
Why does somebody have to say this every time a piece of software from the open source community is evaluated as a replacement for a piece of proprietary software?
Why? For the same reason that 30%-40% of posts in a discussion about XFree86 are about it needing to be reworked or replaced.
Mostly, because it's true
I think IE loses the browser domination the day Sony ships the Playstation 3 (with a non-IE browser, which will probably be Mozilla).
Well, that depends on how visible the browser is.
OK, I accept that a greater number of hits by a Moz based browser would force sites (banks especially) to support Moz, but in the wider context of this discussion - which is about branding and Moz having a strong brand identity - this isn't necessarily true.
If Sony integrate the browser correctly, the only brand which will be visible is the Sony brand - PS3 users won't care what they're surfing with, all they'll care about is that they are surfing. Essentially, the browser vanishes.
Microsoft have picked up on this too - hence the line that IE6 is the last browser all future browsing capability being integrated with Windows. IE will vanish, but browsing will be accessible from explorer/outlook/word or however they implement it.
From the "Joe Sixpack" of view, having a "browser" as a seperate application begins to look like an optional and unecessary extra and a strong brand just serves to emphasise its "seperateness".
Read the Article on Saturday - was surprisingly good.
The article mentions that Acornsoft, the original publishers, didn't want to pay royalty rates as high as Braben and Bell wanted, so they retained publishing rights enabling them to sell those rights at auction to BTSoft, leading to Elite on other platforms.
IMO, by far the best version produced was for the Acorn Archimedes in 1992 - circa 10 years after the original.
Many happy hours as an undergraduate whiled away back then...
This topic reminded me of an article in the print version of UK mag PC Pro by Jon Honeyball. He tells of one of his (consultancy) "clients" looking for a SAN/RAID solution from Dell and EMC, which would cost 120,000.
Aparently, Apples offering was 2TB storage for 9000 - vs 80,000 for 2TB from Dell/EMC.
In the article he says;The obvious question is whether you can put fibre channel cards into your Windows servers and connect them up too. Apple tells me this is possible The idea being to use the storage for SQL*Server databases and the like.
I'm not surprised this is such a good solution for use with Linux.
Dr Miguel Nicolelis knew he had nailed it when the monkey stopped using her arm to play the computer game.
Should this story be in the Ximian section? And isn't it "Dr Miguel d'incaza"? Also, I didn't know the Ximian monkey was female...