Carbon's just an extra layer; it doesn't hugely signify.
Mach+Cocoa+Quartz+Aqua are really just NEXTstep, comprehensively updated, with some bits rewritten to avoid license fees and ported to a new architecture. I don't think it'd be too hard.
But that's not the point. The point is that abandoning PowerPC would really hurt.
There'd be no "blue box" - no Classic MacOS environment, or only one running on a CPU emulator, and thus dog-slow. No existing OS X software would run; it'd all need to be re-compiled.
That costs software houses time and money. Thus they want to sell the result. But who's going to jump ship/again/ and buy all their apps again, even if it's for a bargain price? If you're going to do that, you might as well move to another OS.
These problems face Apple whether it were to move to x86, x86-64 or SPArc or whatever. Mike Spindler did a better job than he generally gets credit for in piloting Apple through the deeply perilous move from 680x0 to PowerPC. The snag is, no CPU now existing has anything like the performance differential over PowerPC that PowerPC had over 680x0, so it can't really repeat the trick...
SPArc isn't a strong architecture, anyway. It's struggling. Sun's moving in an interesting new direction with Niagara, its super-threaded processor, but I think most Sun boxes will be x86-64 in a few years.
But a move to put Solaris on POWER and/or PowerPC isn't impossible, and in that case, yes, it could happen - but I don't think it will. Jobs has his eye firmly on the bottom line now, unlike the idealistic glory days of NeXT back in the '80s.
Just curious - not casting aspersions - where did you get that figure from? I've not heard it.
But even so, that's not amazing growth. 110% means a tenth over twice as much. For all I know, the couple of thousand machines that went into the Virginia Tech cluster might have accounted for all that. I don't think Xserves are exactly flying off the shelves, but I'm glad to hear they're doing better.
Is Darwin up to it? I don't know either. AIX, HP-UX and Solaris are all formidable competitors, tho' HP-UX's fortunes are tied to Itanium's now - so it is probably doomed.
But if it is, it strikes me as a good move.
But you're right: it would be tantamount to clones again. That's why I suggested just OS X Server and only on IBM Servers (not that they're really pushing pSeries workstations any more; x86 has taken over that market, by and large).
But if Darwin ran on non-Apple PPC hardware, it would make it much easier to produce a version of XPostFacto to install Mac OS X on non-Apple hardware. I'd love that - but Steve Jobs wouldn't.
Tell you what would be *much* more doable, though:
Mac OS X Server on IBM pSeries servers - the former RS/6000s. Right CPU, comparable architecture - OpenFirmware, PCI and so on. Great selling point for IBM, validation of enterprise capability for Apple, profitable partnership for both.
I don't *know* that you're wrong, but I have to point out a couple of things:
1. Quartz and Aqua are OPENstep, updated. 2. OPENstep was portable and ran on Solaris and indeed Windows. 3. Sun did have a license to bundle OPENstep: it never used it. NIH syndrome. 4. There was, in the labs, a version of OPENstep/Mach - the whole OS, not just the GUI - that ran on SPARC. And of course Rhapsody's early versions were all on x86.
Much has changed since but I'll bet it's not only technically feasible but not that hard.
However, Sun is SPARC-based with a sideline in x86. Apple is firmly PowerPC-based. That is a problem/much/ harder to overcome.
Re:Brita filters aren't just physical filters
on
Hacking Vodka
·
· Score: 1
Ahhh. All becomes clear. Thankyou for the explanation!
Re:Brita filters aren't just physical filters
on
Hacking Vodka
·
· Score: 1
Oh really? Well, that's good news!
I'm impressed that something so simple manages to scrub so much dissolved calcium carbonate out of the water, then. How does that work? Merely curious - but do have at least 'A' level chemistry, for what it's worth...
In the past, I've used Calgon for this job, but not for drinking water. (My mum used to work for Allbright & Wilson - we got it free.)
Re:Brita filters aren't just physical filters
on
Hacking Vodka
·
· Score: 1
Well, it does say WATER filter in big letters all over all the packaging!
Brita filters aren't just physical filters
on
Hacking Vodka
·
· Score: 1
They will also soften hard water, presumably by chelating some of the metal salts out. This means they contain various reactive chemicals as well as a simple physical filter like fine kaolin china.
If that stuff's reactive, what's it going to do with 35% ethanol/water solution? It might react with the ethanol, or even/dissolve/ in the ethanol. You could be drinking all sorts of nasty chemicals dissolved in the resulting fluid.
I don't care what it tastes like, I wouldn't go near the result, even on a bet...
The HCl in your stomach is very dilute stuff compared to lab acid - even "dilute" lab acid. Plus it has to go through your mouth and oesophagus first, destroying them on the way...
Another detail most people have missed
on
Netscape Reborn?
·
· Score: 2, Insightful
I agree it's good to have commercial input.
But also, the mere name Netscape is still widely recgonised by websites as a supported browser. There are sites I've seen that will work in Netscape 7.1 or 7.2 but won't work in Mozilla 1.3 or 1.7. Betfair.com is one example but I've seen others.
Netscape still has a lot of cachet as The Other Browser Brand.
[1] Source. Why bother? I have no problems installing from binaries if the resolutions are sorted for me. Why use source? Why waste your time compiling? What do you gain? What does "more configurable" mean?
[2] Documentation - and indeed Gentoo in general. Let's put it this way: I've been using & supporting Unix for 15y+, know multiple distros and commercial variants and literally dozens of OSs. I'm not a Linux guru but on computers in general I'm pretty good. My main desktop is Linux.
I spent 2 days+ trying to install Gentoo & I couldn't even get it to boot.
Configurability: a choice of CRON daemons does not a configurable distro make. Let's talk real differences: I'd like to bin the System V init and use the BSD init of Slackware instead. How about that? That's what real customisability means, not a choice of versions of one component.
Documentation: the handholding intro guide doc goes into masses of length about irrelevant parts like boot managers, which don't really matter, and then skims over critical Gentoo-specific stuff like USE flags. I thought it was, generally, not very good at all, and given that I couldn't even install it, my impressions of the distro were - well, unfit to print.
So let it just rest at this: I disgree with your comments regarding Gentoo, or its suitability as the base for a beginner's distro.
Tho' I *do* like to tinker, I don't want to *have* to to make it work acceptably.
Mind you, KDE is notably better at desktop accessories like chat programs and so on. GNOME's handling of these is poor - but then, so it Mac OS X's, relatively. OS X 10.4's Dashboard might fix this.
1. Development effort *is* going back into Debian. Many Ubuntu developers are Debian team members.
2. If UserLinux had come along earlier, or Ubuntu later, maybe they would. Maybe they yet will. But for one thing, Ubuntu has clearly been underway for quite some time, and for another, by keeping its efforts within a privately-funded company, it's come a lot further than UserLinux has so far. (Saying this, I don't know when the two projects started.)
Not that they're talking about publically, no, there aren't.
However, the goals are quite close to those of UserLinux, so some cooperation there might happen some day.
There's not much common ground with general-purpose catch-all Debian-based distros such as Libranet or Knoppix, though, so I see little point in that.
There is some cross-over with commercial efforts like Xandros and Linspire, but being [a] commercial and [b] KDE-based, I see little room for cooperation there.
Why not xBSD: drivers, hardware and software support, expertise and knowledge, polish of existing systems.
Why not a new OS: there are dozens of people doing this. Look at OpenBeOS, Blue Eyed OS, Atheos, SkyOS, Syllable, AROS, etc. etc. The problems are the same as for BSD, but worse: no drivers, no apps, no-one knows how to use them/install them/tweak them/fix them.
ISTM that Ubuntu is a business/corporate-focussed desktop. There is no server functionality, there's good basic productivity/web/email+PIM/chat, there are few games and modest multimedia support. It's a workmanlike OS for basic office/productivity use, or SoHo. It's not a multimedia system or a gaming system, but you can add that stuff if you want.
It's an all-Free answer to Sun Java Desktop System, SuSE Linux Office Desktop or Red Hat Enterprise Workstation. It's much like UserLinux but it's actually here and works now, and that's in the first preview release.
One answer: there are literally many dozens of KDE-based distros out there. There are few GNOME ones.
Another: as someon who's just migrated from KDE 3.3 (on SuSE 9.1) to GNOME 2.8 (on Ubuntu), GNOME is rapidly pulling ahead in simplicity, speed and elegance. KDE is getting steadily bigger, slower and more complex and that is a *bad thing*. GNOME isn't - it's getting more streamlined, elegant and usable. And right now it feels considerably faster.
The latter is all IMHO but it's an opinion based on lots of use of both systems since they were in their respective versions 1.0.
Because in much of Europe people had - and have - much less disposable income than in the USA. Apple made the first personal computer for under $1000 and it was a huge hit - there. Too expensive here. So "Sinkers" made the first personal computer for under £100 - and became a millionaire.
They didn't sell so well in the USA because they were too cheap. Why buy a limited $200 Timex-Sinclair computer when you could have a much better $400 Atari or Commodore? In the UK, the £200-£300 machines were too expensive for most kids...
That's KOWLOON, also known as the New Territories. It was leased from China to Britain later on, after HK island itself.
Re:... a win98 edit.com clone for linux?
on
JOE Hits 3.0
·
· Score: 1
I'm sure you can, but [1] is it really worth all the effort, [2] does it come close to the polish of, say, Thunderbird, when doing so, and [3] how much common user interface does it share with any arbitrary GUI apps running alongside?
Strikes me as a bit of a case of a dog walking on its hind legs.
Re:... a win98 edit.com clone for linux?
on
JOE Hits 3.0
·
· Score: 1
I occasionally use textmode tools, but for day to day work, I want:
- to be able to resize fonts
- to be able to resize the window
- move it from my secondary monitor to my primary one and zoom it
- antialiasing
- inline graphics & formatting
- the choice of mouse or keyboard operation, with toolbars and other handy shortcuts
- to be able to open multiple windows and have their contents viewable concurrently, when searching or comparing
- window layout to be quickly customisable, so I can grab the bar separating message list from preview pane or folder list and drag it around... and so on. Those are just the first few to come to mind.
When I'm working remotely, I don't connect to remote machines, I take one of my notebook computers or my PDA.
Re:... a win98 edit.com clone for linux?
on
JOE Hits 3.0
·
· Score: 1
No, I didn't think you were saying that.:-)
I take your point, but on Windows, say, the basic behaviour of an edit field is common across all edit fields. The same keys, the same mouse actions. Cut, copy, paste, search & replace, bold, italics & other formatting, file load & save, etc., it's all standard.
And DOS EDIT uses the same standards. That's why it's a good familiar tool for Windows users.
These days, most of those same commands also work on the Mac - since the days of System 7 or so - and in cross-platform products such as Mozilla and StarOffice, as well as GUIs such as KDE and GNOME.
So an editor which honours them as well would be a good thing.
Sure, common editors across different programs are great - but WordStar is long dead, so Joe is not that editor for most of us, no more than is VI or EMACS.
I've been in IT for 16yr+ now and I only just remember WordStar - it was dead by 1990. CUA (IBM's Common User Access, the basis for the Windows GUI and most of that of the Mac, DR GEM, Amiga Intuition and others) is the standard now.
Slashdot Mirror
Carbon's just an extra layer; it doesn't hugely signify.
/again/ and buy all their apps again, even if it's for a bargain price? If you're going to do that, you might as well move to another OS.
Mach+Cocoa+Quartz+Aqua are really just NEXTstep, comprehensively updated, with some bits rewritten to avoid license fees and ported to a new architecture. I don't think it'd be too hard.
But that's not the point. The point is that abandoning PowerPC would really hurt.
There'd be no "blue box" - no Classic MacOS environment, or only one running on a CPU emulator, and thus dog-slow. No existing OS X software would run; it'd all need to be re-compiled.
That costs software houses time and money. Thus they want to sell the result. But who's going to jump ship
These problems face Apple whether it were to move to x86, x86-64 or SPArc or whatever. Mike Spindler did a better job than he generally gets credit for in piloting Apple through the deeply perilous move from 680x0 to PowerPC. The snag is, no CPU now existing has anything like the performance differential over PowerPC that PowerPC had over 680x0, so it can't really repeat the trick...
SPArc isn't a strong architecture, anyway. It's struggling. Sun's moving in an interesting new direction with Niagara, its super-threaded processor, but I think most Sun boxes will be x86-64 in a few years.
But a move to put Solaris on POWER and/or PowerPC isn't impossible, and in that case, yes, it could happen - but I don't think it will. Jobs has his eye firmly on the bottom line now, unlike the idealistic glory days of NeXT back in the '80s.
Just curious - not casting aspersions - where did you get that figure from? I've not heard it.
But even so, that's not amazing growth. 110% means a tenth over twice as much. For all I know, the couple of thousand machines that went into the Virginia Tech cluster might have accounted for all that. I don't think Xserves are exactly flying off the shelves, but I'm glad to hear they're doing better.
Is Darwin up to it? I don't know either. AIX, HP-UX and Solaris are all formidable competitors, tho' HP-UX's fortunes are tied to Itanium's now - so it is probably doomed.
But if it is, it strikes me as a good move.
But you're right: it would be tantamount to clones again. That's why I suggested just OS X Server and only on IBM Servers (not that they're really pushing pSeries workstations any more; x86 has taken over that market, by and large).
But if Darwin ran on non-Apple PPC hardware, it would make it much easier to produce a version of XPostFacto to install Mac OS X on non-Apple hardware. I'd love that - but Steve Jobs wouldn't.
Nice idea. Doubt it'll happen.
Tell you what would be *much* more doable, though:
Mac OS X Server on IBM pSeries servers - the former RS/6000s. Right CPU, comparable architecture - OpenFirmware, PCI and so on. Great selling point for IBM, validation of enterprise capability for Apple, profitable partnership for both.
I don't *know* that you're wrong, but I have to point out a couple of things:
/much/ harder to overcome.
1. Quartz and Aqua are OPENstep, updated.
2. OPENstep was portable and ran on Solaris and indeed Windows.
3. Sun did have a license to bundle OPENstep: it never used it. NIH syndrome.
4. There was, in the labs, a version of OPENstep/Mach - the whole OS, not just the GUI - that ran on SPARC. And of course Rhapsody's early versions were all on x86.
Much has changed since but I'll bet it's not only technically feasible but not that hard.
However, Sun is SPARC-based with a sideline in x86. Apple is firmly PowerPC-based. That is a problem
Ahhh. All becomes clear. Thankyou for the explanation!
Oh really? Well, that's good news!
I'm impressed that something so simple manages to scrub so much dissolved calcium carbonate out of the water, then. How does that work? Merely curious - but do have at least 'A' level chemistry, for what it's worth...
In the past, I've used Calgon for this job, but not for drinking water. (My mum used to work for Allbright & Wilson - we got it free.)
Well, it does say WATER filter in big letters all over all the packaging!
They will also soften hard water, presumably by chelating some of the metal salts out. This means they contain various reactive chemicals as well as a simple physical filter like fine kaolin china.
/dissolve/ in the ethanol. You could be drinking all sorts of nasty chemicals dissolved in the resulting fluid.
If that stuff's reactive, what's it going to do with 35% ethanol/water solution? It might react with the ethanol, or even
I don't care what it tastes like, I wouldn't go near the result, even on a bet...
The HCl in your stomach is very dilute stuff compared to lab acid - even "dilute" lab acid. Plus it has to go through your mouth and oesophagus first, destroying them on the way...
I agree it's good to have commercial input.
But also, the mere name Netscape is still widely recgonised by websites as a supported browser. There are sites I've seen that will work in Netscape 7.1 or 7.2 but won't work in Mozilla 1.3 or 1.7. Betfair.com is one example but I've seen others.
Netscape still has a lot of cachet as The Other Browser Brand.
Well...
[1] Source. Why bother? I have no problems installing from binaries if the resolutions are sorted for me. Why use source? Why waste your time compiling? What do you gain? What does "more configurable" mean?
[2] Documentation - and indeed Gentoo in general. Let's put it this way: I've been using & supporting Unix for 15y+, know multiple distros and commercial variants and literally dozens of OSs. I'm not a Linux guru but on computers in general I'm pretty good. My main desktop is Linux.
I spent 2 days+ trying to install Gentoo & I couldn't even get it to boot.
Configurability: a choice of CRON daemons does not a configurable distro make. Let's talk real differences: I'd like to bin the System V init and use the BSD init of Slackware instead. How about that? That's what real customisability means, not a choice of versions of one component.
Documentation: the handholding intro guide doc goes into masses of length about irrelevant parts like boot managers, which don't really matter, and then skims over critical Gentoo-specific stuff like USE flags. I thought it was, generally, not very good at all, and given that I couldn't even install it, my impressions of the distro were - well, unfit to print.
So let it just rest at this: I disgree with your comments regarding Gentoo, or its suitability as the base for a beginner's distro.
Agreed.
Tho' I *do* like to tinker, I don't want to *have* to to make it work acceptably.
Mind you, KDE is notably better at desktop accessories like chat programs and so on. GNOME's handling of these is poor - but then, so it Mac OS X's, relatively. OS X 10.4's Dashboard might fix this.
1. Development effort *is* going back into Debian. Many Ubuntu developers are Debian team members.
2. If UserLinux had come along earlier, or Ubuntu later, maybe they would. Maybe they yet will. But for one thing, Ubuntu has clearly been underway for quite some time, and for another, by keeping its efforts within a privately-funded company, it's come a lot further than UserLinux has so far. (Saying this, I don't know when the two projects started.)
Not that they're talking about publically, no, there aren't.
However, the goals are quite close to those of UserLinux, so some cooperation there might happen some day.
There's not much common ground with general-purpose catch-all Debian-based distros such as Libranet or Knoppix, though, so I see little point in that.
There is some cross-over with commercial efforts like Xandros and Linspire, but being [a] commercial and [b] KDE-based, I see little room for cooperation there.
Why not xBSD: drivers, hardware and software support, expertise and knowledge, polish of existing systems.
Why not a new OS: there are dozens of people doing this. Look at OpenBeOS, Blue Eyed OS, Atheos, SkyOS, Syllable, AROS, etc. etc. The problems are the same as for BSD, but worse: no drivers, no apps, no-one knows how to use them/install them/tweak them/fix them.
Because Gentoo is aimed at expert users who want to compile their own code for optimal performance and don't need hand-holding.
Ubuntu is meant to be an easy simple beginner-friendly distro for non-technical users who are unfamiliar with Linux.
There is no common ground between these two.
ISTM that Ubuntu is a business/corporate-focussed desktop. There is no server functionality, there's good basic productivity/web/email+PIM/chat, there are few games and modest multimedia support. It's a workmanlike OS for basic office/productivity use, or SoHo. It's not a multimedia system or a gaming system, but you can add that stuff if you want.
It's an all-Free answer to Sun Java Desktop System, SuSE Linux Office Desktop or Red Hat Enterprise Workstation. It's much like UserLinux but it's actually here and works now, and that's in the first preview release.
One answer: there are literally many dozens of KDE-based distros out there. There are few GNOME ones.
Another: as someon who's just migrated from KDE 3.3 (on SuSE 9.1) to GNOME 2.8 (on Ubuntu), GNOME is rapidly pulling ahead in simplicity, speed and elegance. KDE is getting steadily bigger, slower and more complex and that is a *bad thing*. GNOME isn't - it's getting more streamlined, elegant and usable. And right now it feels considerably faster.
The latter is all IMHO but it's an opinion based on lots of use of both systems since they were in their respective versions 1.0.
It's simple, really.
Capitalism works better than socialism or communism. Field trials show this.
Capitalism depends on growth; growth requires new supplies of resources and room.
There is nowhere else to grow in the world. It's getting pretty full.
The only way is up.
Short term, it creates money and resources for those on the ground. Longer term, it is the only thing that'll give their kids somewhere to go.
There is.
The Sprinter, from Peters Plus. OK, it's only 21MHz, but it's a modern Spectrum with ISA slots, a hard disk and so on.
There's also the C-One, a modern Commodore 64.
I'm sure there are more besides...
Because in much of Europe people had - and have - much less disposable income than in the USA. Apple made the first personal computer for under $1000 and it was a huge hit - there. Too expensive here. So "Sinkers" made the first personal computer for under £100 - and became a millionaire.
They didn't sell so well in the USA because they were too cheap. Why buy a limited $200 Timex-Sinclair computer when you could have a much better $400 Atari or Commodore? In the UK, the £200-£300 machines were too expensive for most kids...
It's all about money.
That's KOWLOON, also known as the New Territories. It was leased from China to Britain later on, after HK island itself.
I'm sure you can, but [1] is it really worth all the effort, [2] does it come close to the polish of, say, Thunderbird, when doing so, and [3] how much common user interface does it share with any arbitrary GUI apps running alongside?
Strikes me as a bit of a case of a dog walking on its hind legs.
I occasionally use textmode tools, but for day to day work, I want: ... and so on. Those are just the first few to come to mind.
- to be able to resize fonts
- to be able to resize the window
- move it from my secondary monitor to my primary one and zoom it
- antialiasing
- inline graphics & formatting
- the choice of mouse or keyboard operation, with toolbars and other handy shortcuts
- to be able to open multiple windows and have their contents viewable concurrently, when searching or comparing
- window layout to be quickly customisable, so I can grab the bar separating message list from preview pane or folder list and drag it around
When I'm working remotely, I don't connect to remote machines, I take one of my notebook computers or my PDA.
No, I didn't think you were saying that. :-)
I take your point, but on Windows, say, the basic behaviour of an edit field is common across all edit fields. The same keys, the same mouse actions. Cut, copy, paste, search & replace, bold, italics & other formatting, file load & save, etc., it's all standard.
And DOS EDIT uses the same standards. That's why it's a good familiar tool for Windows users.
These days, most of those same commands also work on the Mac - since the days of System 7 or so - and in cross-platform products such as Mozilla and StarOffice, as well as GUIs such as KDE and GNOME.
So an editor which honours them as well would be a good thing.
Sure, common editors across different programs are great - but WordStar is long dead, so Joe is not that editor for most of us, no more than is VI or EMACS.
I've been in IT for 16yr+ now and I only just remember WordStar - it was dead by 1990. CUA (IBM's Common User Access, the basis for the Windows GUI and most of that of the Mac, DR GEM, Amiga Intuition and others) is the standard now.