BeOS is still miles ahead. Unix/Linux is excellent as a server/networking platform, but it lacks too many features in the desktop area to make it suitable for musicians. This is the kind of people that can't really be bothered with setting up X, or messing with/etc/resolv.conf.
The thing that troubles me about BeOS is the lacking hardware support, though. Otherwise I'd be doing a lot more with it...
OK, this answer is long overdue, sorry for that...
I've been thinking a bit about and I think we agree on most points. There's little I can contradict in your message.
Yet, I think that you can't pick any OS and bend it to your needs. The security model of Unix, for instance, is rather simple. It works, sure, and that's a merit as well. But if you look at the design of Windows NT, it's actually very nice. The implementation, however, leaves much to be desired.
I still think that an OS can be more suited for a certain purpose if it's designed from the ground up for exactly that purpose. BeOS is an example of that. I admit that software and hardware support is far from optimal, though.
And BeOS (and AmigaOS as well) has a very nice API. The one of Windows and X make me run away. Toolboxes only help to a certain extend.
My solution on the multiple OS situation will be to use multiple computers. My Pentium 200 will be the mail/internet machine under NT and my new one a music workstation (OK, and games).
Then again, I am a very a-typical user. The majority can use what they want, but what I'd want to avoid is that I would have to moo along with the herd, because there's no alternative to the OS of choice - whether that would be Windows or Unix.
I can't see anything bad about less OSes. Having multiple OSes means having apps for each one, and uses to which one is more suited than the other...
Some OSes don't seem like viable merge candidates. The toaster doesn't need to run Linux. But, I can't see why some unix core with a nice GUI should run 99.99% of PCs.
For one thing, Unix isn't real-time (and neither is BeOS, BTW). There are plenty of applications running on PC's where an RTOS is a requirement, notably in the industry. Of course, it only applies for a small percentage of users.
In all the years that Unix exists, nobody has been able to come up with a decent GUI that supports data-sharing between applications, drag-and-drop, the like, whereas interrupt-handler MS-DOS somehow mutated into a fairly usable multitasking OS that does do this.
Less OSes certainly is a Good Thing - why are there more versions of Unix than there are craters on the moon? - but there's no such thing as the ultimate OS for all applications. Jack of all trades, but master of none.
As for me, I don't see the very specific stuff - music and audio - moving from the Mac and Windows to Linux, but BeOS definately has grabbed some ground here.
Microsoft's high end is the desktop, and that's the last market Linux will take over. This is a telling point, and one that is rarely made in analyses of Linux. We'll get there in the end, but only after all the other markets have fallen.
Excuse me for laughing, but do we really want to go from a market dominated by Windows to a market dominated by Linux?
I seem to recall a story about this: kill the devil, and become the devil yourself.
I'd rather be found dead than having to use Linux for my interest: music and multimedia. BeOS is my choice here. Let OS's do what they're good at. Linux has its strong points, so have others. There is no can-all-do-all OS for all situations.
People are probably going to kill me for this, but I chose to use Internet Explorer, simply because I think it's a better product. It surpassed Netscape.
Sure, Microsoft's market monopoly sux major big time but that doesn't imply that they make bad products.
I started using MSIE on my Sun Solaris workstation, because I wanted the same software regardless of the platform I was working on.
Personally, I found that IE was more stable on any platform, and better sticking to the standards. Yes, you read that right. Try creating some HTML 4.0 or CSS, and see what browser complies best. I've tried it and I know it.
I try to keep the religious arguments out of it. You won't win a war with an icon of Linus and the Holy Book of 4.4 BSD in your hands. Make a better product. There's nothing to be gained by using Netscape, just because it doesn't come from Redmond.
My dream is to seperate user interface and application. Use a browser as your GUI! I couldn't care less what browser that would be. I just know that IE offers me the best possibilites at this moment...
You can beat them by creating something new. Make a standard that uses Netscape as the GUI and communicaties over TCP/IP with the application. That's how you can win.
Interesting: I do this for a living, writing Perl CGI scripts (on a Linux box, incidently).
There's much to be said about Perl. Its syntax is totally confusing, and the OO implementation is a laugh. C++ is truly elegant, compared to it. As a programming language, I prefer C/C++ over Perl.
However, for CGI scripts the story is somewhat different. Perl is excellent to write relatively small programs that work with text. And with HTTP and HTML, you only work with text and small programs. The CGI interface gives you parameters as name/value pairs, another strong point of Perl.
PHP is not an option either. The whole idea of HTML (and later, CSS) is to seperate content and layout. It's kinda silly to merge content and code. PHP is nice for the incidental webmaster who makes an extended MRTG site with it, but as soon as you work with professional designers who tend to move things around a bit, PHP no longer works.
As for C/C++, it's much more cumbersome to work with texts than it is with Perl. And I don't fancy writing CGI-scripts in (c)sh, tcl, expect or awk. (Emacs anyone?)
With the right libs from CPAN, you can work with template HTML files and CGI Perl scripts that fill these out. The designer makes with the templates, the programmer writes a scripts that does the filling out.
If speed is such an important issue, why the heck would you put it on the WWW anyway? I don't suppose many of us will be building number-crunching sites.
Yes, the Hercules Thriller 3D with the Rendition Vérité 2200 chipset (I have the one with 8 Mb, video input and output, and 3D glasses VESA connector) is a great card, at least for my P200. Beautiful picture and a good OpenGL implementation.
But overall software support is horrible. I've bought BeOS, but I can't do anything with it due to lacking video drivers.
Personally, I'd be more interested in how fast an NT system could boot. My NT Server (P200, 80 Mb RAM, 4,5 Gb FastSCSI) takes about a minute after the BIOS stuff, which is faster than when I had FreeBSD installed. But then again, I used a P133 and a slower SCSI drive back then.
I'll be putting my money in an RME DIGI9652 "Hammerfall" (www.rme-audio.com). This one offers 24 optical digital ins and outs, S/PDIF in and out, external synchronisation and zero latency monitoring.
Unless Yamaha comes out first with a FireWire/mLAN card for my mixing desk, that is.
There's really no use for me to invest in a new Amiga, when extending my current PC or even buying a dual processor system is eventually the cheaper solution.
The Amiga is at a dead end. It just took some years for it to figure that out.
When I started with an Amiga - say, some 12 or 13 years ago - I liked it primarily for its games and its sound. Hey, I still was a kid. I still am, in a way.
After a while I began to appreciate the simple - though limited - and elegant operating system. And because it could do things that no other machine could: because of its dedicated hardware.
Let's be honest: that was what made the Amiga. Operating system wise, we've evolved quite a bit in fifteen years and I wouldn't want to trade in my NT 4.0 Desktop against the Amiga - seriously.
My interest has shifted towards digital audio now and I have a load of professional music equipment at my fingertips. My ideal is to create my own Digital Audio Workstation, but synchronisation - in other words, dedicated hardware - is a problem with both PC's and Macs. BeOS is not going to change that.
The MCC Amiga would have been a new way of looking at things, at taking the multimedia problem at the root. No longer an office machine doing multimedia things, but a real multimedia machine. I'm going to pay 500 euro for a mere sound card - I wouldn't have minded a more expensive machine.
The announcement at the Amiga site is an insult to my intelligence. Gateway have payed $13m to have a bouncing ball on a box that isn't Amiga, on an OS that isn't Amiga.
My choice will be Windows NT/2000, or BeOS if it manages to support my hardware. The Amiga era has finally ended.
Slashdot Mirror
BeOS is still miles ahead. Unix/Linux is excellent as a server/networking platform, but it lacks too many features in the desktop area to make it suitable for musicians. This is the kind of people that can't really be bothered with setting up X, or messing with /etc/resolv.conf.
The thing that troubles me about BeOS is the lacking hardware support, though. Otherwise I'd be doing a lot more with it...
OK, this answer is long overdue, sorry for that...
I've been thinking a bit about and I think we agree on most points. There's little I can contradict in your message.
Yet, I think that you can't pick any OS and bend it to your needs. The security model of Unix, for instance, is rather simple. It works, sure, and that's a merit as well. But if you look at the design of Windows NT, it's actually very nice. The implementation, however, leaves much to be desired.
I still think that an OS can be more suited for a certain purpose if it's designed from the ground up for exactly that purpose. BeOS is an example of that. I admit that software and hardware support is far from optimal, though.
And BeOS (and AmigaOS as well) has a very nice API. The one of Windows and X make me run away. Toolboxes only help to a certain extend.
My solution on the multiple OS situation will be to use multiple computers. My Pentium 200 will be the mail/internet machine under NT and my new one a music workstation (OK, and games).
Then again, I am a very a-typical user. The majority can use what they want, but what I'd want to avoid is that I would have to moo along with the herd, because there's no alternative to the OS of choice - whether that would be Windows or Unix.
I can't see anything bad about less OSes. Having multiple OSes means having apps for each one, and uses to which one is more suited than the other...
Some OSes don't seem like viable merge candidates. The toaster doesn't need to run Linux. But, I can't see why some unix core with a nice GUI should run 99.99% of PCs.
For one thing, Unix isn't real-time (and neither is BeOS, BTW). There are plenty of applications running on PC's where an RTOS is a requirement, notably in the industry. Of course, it only applies for a small percentage of users.
In all the years that Unix exists, nobody has been able to come up with a decent GUI that supports data-sharing between applications, drag-and-drop, the like, whereas interrupt-handler MS-DOS somehow mutated into a fairly usable multitasking OS that does do this.
Less OSes certainly is a Good Thing - why are there more versions of Unix than there are craters on the moon? - but there's no such thing as the ultimate OS for all applications. Jack of all trades, but master of none.
As for me, I don't see the very specific stuff - music and audio - moving from the Mac and Windows to Linux, but BeOS definately has grabbed some ground here.
And BeOS builds heavily on Unix anyway... :-)
Microsoft's high end is the desktop, and that's the last market Linux will take over.
This is a telling point, and one that is rarely made in analyses of Linux. We'll get there in the end, but only after all the other markets have fallen.
Excuse me for laughing, but do we really want to go from a market dominated by Windows to a market dominated by Linux?
I seem to recall a story about this: kill the devil, and become the devil yourself.
I'd rather be found dead than having to use Linux for my interest: music and multimedia. BeOS is my choice here. Let OS's do what they're good at. Linux has its strong points, so have others. There is no can-all-do-all OS for all situations.
I'd be really set if BeOS would support my hardware, or the hardware I intend to buy. :-)
BeOS is an operating system that deserves to flourish. I have it at home, but I can't use it with my video board (bummer).
Anyway, I'm waiting with my new PC until BeOS supports a video board that contains a DVI connector...
People are probably going to kill me for this, but I chose to use Internet Explorer, simply because I think it's a better product. It surpassed Netscape.
Sure, Microsoft's market monopoly sux major big time but that doesn't imply that they make bad products.
I started using MSIE on my Sun Solaris workstation, because I wanted the same software regardless of the platform I was working on.
Personally, I found that IE was more stable on any platform, and better sticking to the standards. Yes, you read that right. Try creating some HTML 4.0 or CSS, and see what browser complies best. I've tried it and I know it.
I try to keep the religious arguments out of it. You won't win a war with an icon of Linus and the Holy Book of 4.4 BSD in your hands. Make a better product. There's nothing to be gained by using Netscape, just because it doesn't come from Redmond.
My dream is to seperate user interface and application. Use a browser as your GUI! I couldn't care less what browser that would be. I just know that IE offers me the best possibilites at this moment...
You can beat them by creating something new. Make a standard that uses Netscape as the GUI and communicaties over TCP/IP with the application. That's how you can win.
- Peter
Interesting: I do this for a living, writing Perl CGI scripts (on a Linux box, incidently).
There's much to be said about Perl. Its syntax is totally confusing, and the OO implementation is a laugh. C++ is truly elegant, compared to it. As a programming language, I prefer C/C++ over Perl.
However, for CGI scripts the story is somewhat different. Perl is excellent to write relatively small programs that work with text. And with HTTP and HTML, you only work with text and small programs. The CGI interface gives you parameters as name/value pairs, another strong point of Perl.
PHP is not an option either. The whole idea of HTML (and later, CSS) is to seperate content and layout. It's kinda silly to merge content and code. PHP is nice for the incidental webmaster who makes an extended MRTG site with it, but as soon as you work with professional designers who tend to move things around a bit, PHP no longer works.
As for C/C++, it's much more cumbersome to work with texts than it is with Perl. And I don't fancy writing CGI-scripts in (c)sh, tcl, expect or awk. (Emacs anyone?)
With the right libs from CPAN, you can work with template HTML files and CGI Perl scripts that fill these out. The designer makes with the templates, the programmer writes a scripts that does the filling out.
If speed is such an important issue, why the heck would you put it on the WWW anyway? I don't suppose many of us will be building number-crunching sites.
Well, just my 0.02 euro worth.
- Peter
Yes, the Hercules Thriller 3D with the Rendition Vérité 2200 chipset (I have the one with 8 Mb, video input and output, and 3D glasses VESA connector) is a great card, at least for my P200. Beautiful picture and a good OpenGL implementation.
But overall software support is horrible. I've bought BeOS, but I can't do anything with it due to lacking video drivers.
Personally, I'd be more interested in how fast an NT system could boot. My NT Server (P200, 80 Mb RAM, 4,5 Gb FastSCSI) takes about a minute after the BIOS stuff, which is faster than when I had FreeBSD installed. But then again, I used a P133 and a slower SCSI drive back then.
- Peter
I'll be putting my money in an RME DIGI9652 "Hammerfall" (www.rme-audio.com). This one offers 24 optical digital ins and outs, S/PDIF in and out, external synchronisation and zero latency monitoring.
Unless Yamaha comes out first with a FireWire/mLAN card for my mixing desk, that is.
There's really no use for me to invest in a new Amiga, when extending my current PC or even buying a dual processor system is eventually the cheaper solution.
The Amiga is at a dead end. It just took some years for it to figure that out.
I think the Amiga was the best 2D games machine ever. But also I think 2D is something of the past as well.
Too bad: my PC needs to double speed every so many months, or otherwise I can't keep playing games...
When I started with an Amiga - say, some 12 or 13 years ago - I liked it primarily for its games and its sound. Hey, I still was a kid. I still am, in a way.
After a while I began to appreciate the simple - though limited - and elegant operating system. And because it could do things that no other machine could: because of its dedicated hardware.
Let's be honest: that was what made the Amiga. Operating system wise, we've evolved quite a bit in fifteen years and I wouldn't want to trade in my NT 4.0 Desktop against the Amiga - seriously.
My interest has shifted towards digital audio now and I have a load of professional music equipment at my fingertips. My ideal is to create my own Digital Audio Workstation, but synchronisation - in other words, dedicated hardware - is a problem with both PC's and Macs. BeOS is not going to change that.
The MCC Amiga would have been a new way of looking at things, at taking the multimedia problem at the root. No longer an office machine doing multimedia things, but a real multimedia machine. I'm going to pay 500 euro for a mere sound card - I wouldn't have minded a more expensive machine.
The announcement at the Amiga site is an insult to my intelligence. Gateway have payed $13m to have a bouncing ball on a box that isn't Amiga, on an OS that isn't Amiga.
My choice will be Windows NT/2000, or BeOS if it manages to support my hardware. The Amiga era has finally ended.