Virtually as important to that is pulling off Carbon. Without that, any prospective successor to classic Mac OS was doomed to failure.
Sorry, but anything Carbon-alike wasn't in the planning at all, at those days.
Quote:
Copeland, Gershwin and Pink never should have been invested in. Cleaning up and making the Mac OS Toolbox thread-safe should have been Apple's number priority prior to inventing some fancy new operating system.
Gershwin development never started. Copland (named after Aaron Copland, just like Gershwin is named after George Gershwin, who composed Rhapsody - ! - in Blue) gave Apple a set of interesting technologies, some of which are still not implemented (e.g. Piles), whereas others have been implemented with subsequent system versions like 8.5, 9 or even 10.x.
Making the Mac OS Toolbox "thread-safe" would have left us with a disastrously old operating system.
OT, but why do so many people call it Copeland? It's Copland.
Maybe because I pronounce it like "cope + land", not "cop + land", unless I'm wrong on the pronunciation. But ok, I'll spell it like Copland from now on.
Sorry, but anything Carbon-alike wasn't in the planning at all, at those days.
Yes, but the point was it needed to be development during those days. That's all Apple needed to do in the 1992 to 1995 time frame, in conjuction with the transition to PowerPC. Once that's done, Apple had its choice of buzzword compliant operating system kernels to choose from. If Carbon was in development, Apple didn't need to buy BeOS or NeXT.
Quote:
Gershwin development never started. Copland (named after Aaron Copland, just like Gershwin is named after George Gershwin, who composed Rhapsody - ! - in Blue) gave Apple a set of interesting technologies, some of which are still not implemented (e.g. Piles), whereas others have been implemented with subsequent system versions like 8.5, 9 or even 10.x.
Piles? Bah, the Apple ecosystem of operating system and GUI design has been entirely too inbred. Interface design and research has been quite crappy in the 1990s. They are barely interesting. And look at what got implemented and what got dropped. HFS+ is probably the most noteworthy, barely. Making the Finder PowerPC native probably convinced Apple that they could make it a Carbon app, to much our disappointment. Maybe V-Twin will end up to be the most useful.
The big technologies, QuickDraw GX, OpenTransport and OpenDoc, even UI Themes, have all been dropped.
Quote:
Making the Mac OS Toolbox "thread-safe" would have left us with a disastrously old operating system.
As per the Fork, Mac OS X is dominated by almost thread-safe Mac OS Toolbox applications formerly known as Carbon apps, such as the Finder, all Microsoft apps, all Adobe apps, all Macromedia apps, Netscape/Mozilla, etc.
Carbon is largely a thread-safe version of the old Toolbox API, isn't it?
Yes, but Carbon on OS X is "a modernized and simplified thread-safe version of the old Toolbox API on top of a modern operating system base with a Microkernel". That's different from Carbon on OS 9, which is "a modernized and simplified thread-safe version of the old Toolbox API on top of the very same operating system which was old-fashioned when intro'd in 1984".
Yes, but Carbon on OS X is "a modernized and simplified thread-safe version of the old Toolbox API on top of a modern operating system base with a Microkernel". That's different from Carbon on OS 9, which is "a modernized and simplified thread-safe version of the old Toolbox API on top of the very same operating system which was old-fashioned when intro'd in 1984".
Actually, OS 9 has a microkernel architecture and better threading support, etc. Still completely half-assed, because Apple had to keep the old Toolbox around, but better.
Apple should have rolled out the Reentrant Toolbox with a preemtive-multitasking System 7, instead of building a shareware hack into the system. I still think the decision to call MultiFinder "good enough" was one of Apple's most disastrous decisions. System 7 could have been such a great foundation to build on...
That's different from Carbon on OS 9, which is "a modernized and simplified thread-safe version of the old Toolbox API on top of the very same operating system which was old-fashioned when intro'd in 1984".
Yeah, but there wouldn't be any point in making the OS thread-safe without a new kernel.
Quote:
OS 9 has a microkernel architecture
I think they called it a nanokernel (starting with 8.6, I think), but I don't have any idea what that's supposed to mean. And I'm not sure how it could be said to have such an architecture. Any ideas?
I think they called it a nanokernel (starting with 8.6, I think), but I don't have any idea what that's supposed to mean. And I'm not sure how it could be said to have such an architecture. Any ideas?
It means they were able to abstract away the barest minimum of functionality into a core and run that as seperately as possible. It allowed OS 9 to be able to do things like relaunch Finder when it crashed (although you had to quit all your applications first) and "soft restart" instead of letting every error bring the system down.
It was one of those things you didn't notice until things went wrong, but when they did it was kind of nice. Probably made it a bit easier to implement Carbon as well, and also Classic.
Words like BeOS and Gil Amelio make me cringe. Stop it already!
Had Steve been Apple's CEO all along Apple's history, we would have seen what we now see in OS X lonnng ago. I'll guesstimate maybe around 1997, 1998 tops.
It isn't BeOS or Mac OS 9 nor Mac OS X that matters. The difference was made by the Architect guiding it forward. Apple has never had and will never have as good an Architect as Steve.
- BeOS had various features OS X still doesn't have, such as live queries for metadata, an enormously speedy file system and generally SPEEEEED.
mmm....... metadata.....
from The Register (3/29/2002) ...BFS, a fast, 64bit journaled file system written by Dominic Giampaolo, which had many database-like properties......Dominic, we are delighted to learn, has subsequently joined Apple as a file system engineer. He started last week).
Comments
Originally posted by THT
Virtually as important to that is pulling off Carbon. Without that, any prospective successor to classic Mac OS was doomed to failure.
Sorry, but anything Carbon-alike wasn't in the planning at all, at those days.
Copeland, Gershwin and Pink never should have been invested in. Cleaning up and making the Mac OS Toolbox thread-safe should have been Apple's number priority prior to inventing some fancy new operating system.
Gershwin development never started. Copland (named after Aaron Copland, just like Gershwin is named after George Gershwin, who composed Rhapsody - ! - in Blue) gave Apple a set of interesting technologies, some of which are still not implemented (e.g. Piles), whereas others have been implemented with subsequent system versions like 8.5, 9 or even 10.x.
Making the Mac OS Toolbox "thread-safe" would have left us with a disastrously old operating system.
Making the Mac OS Toolbox "thread-safe" would have left us with a disastrously old operating system.
Carbon is largely a thread-safe version of the old Toolbox API, isn't it?
Originally posted by JLL
OT, but why do so many people call it Copeland? It's Copland.
Maybe because I pronounce it like "cope + land", not "cop + land", unless I'm wrong on the pronunciation. But ok, I'll spell it like Copland from now on.
Originally posted by Chucker
Sorry, but anything Carbon-alike wasn't in the planning at all, at those days.
Yes, but the point was it needed to be development during those days. That's all Apple needed to do in the 1992 to 1995 time frame, in conjuction with the transition to PowerPC. Once that's done, Apple had its choice of buzzword compliant operating system kernels to choose from. If Carbon was in development, Apple didn't need to buy BeOS or NeXT.
Gershwin development never started. Copland (named after Aaron Copland, just like Gershwin is named after George Gershwin, who composed Rhapsody - ! - in Blue) gave Apple a set of interesting technologies, some of which are still not implemented (e.g. Piles), whereas others have been implemented with subsequent system versions like 8.5, 9 or even 10.x.
Piles? Bah, the Apple ecosystem of operating system and GUI design has been entirely too inbred. Interface design and research has been quite crappy in the 1990s. They are barely interesting. And look at what got implemented and what got dropped. HFS+ is probably the most noteworthy, barely. Making the Finder PowerPC native probably convinced Apple that they could make it a Carbon app, to much our disappointment. Maybe V-Twin will end up to be the most useful.
The big technologies, QuickDraw GX, OpenTransport and OpenDoc, even UI Themes, have all been dropped.
Making the Mac OS Toolbox "thread-safe" would have left us with a disastrously old operating system.
As per the Fork, Mac OS X is dominated by almost thread-safe Mac OS Toolbox applications formerly known as Carbon apps, such as the Finder, all Microsoft apps, all Adobe apps, all Macromedia apps, Netscape/Mozilla, etc.
Originally posted by Mac The Fork
Carbon is largely a thread-safe version of the old Toolbox API, isn't it?
Yes, but Carbon on OS X is "a modernized and simplified thread-safe version of the old Toolbox API on top of a modern operating system base with a Microkernel". That's different from Carbon on OS 9, which is "a modernized and simplified thread-safe version of the old Toolbox API on top of the very same operating system which was old-fashioned when intro'd in 1984".
Originally posted by Chucker
Yes, but Carbon on OS X is "a modernized and simplified thread-safe version of the old Toolbox API on top of a modern operating system base with a Microkernel". That's different from Carbon on OS 9, which is "a modernized and simplified thread-safe version of the old Toolbox API on top of the very same operating system which was old-fashioned when intro'd in 1984".
Actually, OS 9 has a microkernel architecture and better threading support, etc. Still completely half-assed, because Apple had to keep the old Toolbox around, but better.
Apple should have rolled out the Reentrant Toolbox with a preemtive-multitasking System 7, instead of building a shareware hack into the system. I still think the decision to call MultiFinder "good enough" was one of Apple's most disastrous decisions. System 7 could have been such a great foundation to build on...
Originally posted by Amorph
Actually, OS 9 has a microkernel architecture and better threading support, etc.
I didn't know that.
Learn something new every day.
That's different from Carbon on OS 9, which is "a modernized and simplified thread-safe version of the old Toolbox API on top of the very same operating system which was old-fashioned when intro'd in 1984".
Yeah, but there wouldn't be any point in making the OS thread-safe without a new kernel.
OS 9 has a microkernel architecture
I think they called it a nanokernel (starting with 8.6, I think), but I don't have any idea what that's supposed to mean. And I'm not sure how it could be said to have such an architecture. Any ideas?
Originally posted by Mac The Fork
I think they called it a nanokernel (starting with 8.6, I think), but I don't have any idea what that's supposed to mean. And I'm not sure how it could be said to have such an architecture. Any ideas?
It means they were able to abstract away the barest minimum of functionality into a core and run that as seperately as possible. It allowed OS 9 to be able to do things like relaunch Finder when it crashed (although you had to quit all your applications first) and "soft restart" instead of letting every error bring the system down.
It was one of those things you didn't notice until things went wrong, but when they did it was kind of nice. Probably made it a bit easier to implement Carbon as well, and also Classic.
Not sure what BeOS's kernel was called, but I thought I'd toss in Next/OS X's Mach Kernel for good measure. Discuss...
Had Steve been Apple's CEO all along Apple's history, we would have seen what we now see in OS X lonnng ago. I'll guesstimate maybe around 1997, 1998 tops.
It isn't BeOS or Mac OS 9 nor Mac OS X that matters. The difference was made by the Architect guiding it forward. Apple has never had and will never have as good an Architect as Steve.
[B]Remember this:
- BeOS had various features OS X still doesn't have, such as live queries for metadata, an enormously speedy file system and generally SPEEEEED.
mmm....... metadata.....
from The Register (3/29/2002) ...BFS, a fast, 64bit journaled file system written by Dominic Giampaolo, which had many database-like properties......Dominic, we are delighted to learn, has subsequently joined Apple as a file system engineer. He started last week).
here's the whole link..
http://www.theregister.com/content/archive/24648.html
MG