FROM : Bill Bumgarner
DATE : Sun Feb 17 21:09:53 2008
On Feb 17, 2008, at 10:52 AM, William Squires wrote:
> Okay, that explains it - weird... I wonder why ObjC did that?
When Objective-C was created, the targeted machines had relatively
little memory -- 4MB or less, typically.
As such, every single byte that could be saved was valuable. So, as
a simple extension of C, Objective-C classes are really just a
glorified C structure that is extended across subclasses and has a
slot at the head that can point to the class of the structure.
Simple. Elegant. Compact. Easy to implement. And prone to the
fragile base class problem.
While memory is still at a premium, eating a couple of pointers worth
of memory to solve the fragile base class problem is now a worthwhile
tradeoff.
It would have been nice if this change could have been made in the
10.0 timeframe when there wasn't a binary compatibility issue
preventing adoption in 32 bits. But, at the time, there were other
priorities that out weighed solving this particular problem.
b.bum
DATE : Sun Feb 17 21:09:53 2008
On Feb 17, 2008, at 10:52 AM, William Squires wrote:
> Okay, that explains it - weird... I wonder why ObjC did that?
When Objective-C was created, the targeted machines had relatively
little memory -- 4MB or less, typically.
As such, every single byte that could be saved was valuable. So, as
a simple extension of C, Objective-C classes are really just a
glorified C structure that is extended across subclasses and has a
slot at the head that can point to the class of the structure.
Simple. Elegant. Compact. Easy to implement. And prone to the
fragile base class problem.
While memory is still at a premium, eating a couple of pointers worth
of memory to solve the fragile base class problem is now a worthwhile
tradeoff.
It would have been nice if this change could have been made in the
10.0 timeframe when there wasn't a binary compatibility issue
preventing adoption in 32 bits. But, at the time, there were other
priorities that out weighed solving this particular problem.
b.bum
| Related mails | Author | Date |
|---|---|---|
| Randall Meadows | Feb 11, 22:14 | |
| Melissa J. Turner | Feb 11, 22:27 | |
| Kyle Sluder | Feb 11, 22:28 | |
| Joshua Emmons | Feb 11, 22:29 | |
| Shawn Erickson | Feb 11, 22:33 | |
| j o a r | Feb 11, 22:35 | |
| Brian Christensen | Feb 11, 22:38 | |
| Randall Meadows | Feb 11, 22:56 | |
| Jens Alfke | Feb 11, 23:11 | |
| Kyle Sluder | Feb 11, 23:32 | |
| j o a r | Feb 11, 23:53 | |
| Adam P Jenkins | Feb 12, 00:33 | |
| Bill Bumgarner | Feb 12, 00:46 | |
| Wade Tregaskis | Feb 12, 00:46 | |
| Kyle Sluder | Feb 12, 01:08 | |
| Nick Zitzmann | Feb 12, 01:58 | |
| Jens Alfke | Feb 12, 04:44 | |
| Andrew Farmer | Feb 12, 06:25 | |
| Sean McBride | Feb 12, 16:58 | |
| j o a r | Feb 12, 19:48 | |
| Jens Alfke | Feb 12, 20:14 | |
| Wade Tregaskis | Feb 12, 20:35 | |
| William Squires | Feb 16, 23:35 | |
| Jean-Daniel Dupas | Feb 17, 01:03 | |
| William Squires | Feb 17, 17:59 | |
| glenn andreas | Feb 17, 18:11 | |
| Bill Bumgarner | Feb 17, 18:11 | |
| Jim Correia | Feb 17, 18:13 | |
| William Squires | Feb 17, 18:47 | |
| mmalc crawford | Feb 17, 18:57 | |
| Bill Bumgarner | Feb 17, 19:03 | |
| Jim Correia | Feb 17, 19:04 | |
| William Squires | Feb 17, 19:52 | |
| Jean-Daniel Dupas | Feb 17, 20:18 | |
| Sherm Pendley | Feb 17, 21:03 | |
| Bill Bumgarner | Feb 17, 21:09 | |
| Bill Bumgarner | Feb 17, 21:17 | |
| Jens Alfke | Feb 18, 00:06 |






Cocoa mail archive

