Slashdot Mirror
Looking Back At OS X's Origins
DJRumpy writes "Macworld Weekly has an interesting look at the history of OS X from its early origins in 1985 under NeXT and the Mach Kernel to Rhapsody, to its current iteration as OS X. An interesting, quick read if anyone is curious about the timeline from Apple's shaky '90s to their current position in the market. There's also an interesting link at the bottom talking about the difference between the original beta and the release product that we see today."
Check out Ars' run down too: http://arstechnica.com/apple/reviews/2010/09/macos-x-beta.ars
Thank you, editors.
http://en.wikipedia.org/wiki/Nextstep
It was OPENSTEP 4.2 --- which Apple actually sold for a time, along w/ providing free Y2K patches and free upgrades to NeXTstep 3.3 or OPENSTEP 4.2 to license holders of earlier versions.
Amusing rumour is that ``Yellow Box'' was so named because Bill Gates, when asked if he'd develop for NeXT stated, ``Develop for it? I'll piss on it.''
http://www.theregister.co.uk/2009/11/14/gates_says_jobs_saved_apple/
As nice as Mac OS X is though, I'd still rather have NeXTstep:
- Display PostScript
- built-in PANTONE colour library
- vertical, movable menu bar w/ tear off menus and pop-up menus
- top-level Print, Hide, Quit and Services menu
- TeX provided by default and supported by the nifty TeXview.app
- inspector-provided sort options for Miller-column filebrowser view
- re-sizeable Shelf which can store multiple file selections as a single icon
- nifty apps which made use of Services and Display PostScript like beYAP.app, Altsys Virtuoso, poste.app &c.
William
Sphinx of black quartz, judge my vow.
Double click the resize knob at the bottom of the column, it will size itself to fit all file names in.
Sorry for self reply - my first Mac was a IIci; yes color was missing from the Mac between 1984 and '87.
Wish I could delete my previsou. post
Then what would you say about an OS which:
- was \textsc{unix}
- supported the initial versions of http
- was used to develop a graphical web browser and editor named worldwideweb.app[1]
NeXTstep, available in 1989
William
1 - _Weaving the Web_ by Sir Tim Berners-Lee --- http://www.w3.org/People/Berners-Lee/Weaving/Overview.html
Sphinx of black quartz, judge my vow.
Actually, System 8 (Copland) had a ton of problems without including Jobs. The problem was, it was a disjointed effort where nothing was getting done. If anything blame Ellen Hancock for purchasing NeXT because when she was hired she basically said "screw this, it isn't ever going to get shipped" so they bailed out Jobs.
Copland wasn't going anywhere so Apple decided to cut their losses.
Taxation is legalized theft, no more, no less.
You're buying into Bill Gates' bullshit. Apple didn't "steal" anything; they had an agreement with Xerox. Many of the guys who worked on the Mac were hired from Xerox.
Several conventions originated at Apple, such as the "File Edit View Window Help" menu or the phrase "cut and paste." Lisa was already in development when Apple visited Xerox to see what they were working on, so while they were influenced by what they saw, it wasn't an inspiration to go in some whole new direction.
Much of this is detailed at Herztfeld's site, including sketches and screenshots of their GUI work.
Apple did not steal the GUI from Xerox. They got to tour PARC with permission from Xerox's upper management and compensated Xerox with pre-IPO shares. What the Mac did with the ideas from PARC was very different from what Xerox did with the ideas out of PARC. This is also very different from Microsoft sending an employee to copy implementation details from Apple. Do go waving some out of context quote around without knowing the actual history of the situation.
I'm a loner Dottie, a Rebel.
See, the idea that Apple stole the GUI lock/stock from Xerox and then accused Microsoft of the same thing is a massive myth. Have you even looked at the Alto/Star GUI? It used modal buttons along the bottom of windows; windows were tiled and could not overlap. Yes, the general concept of the GUI was developed at PARC, although that wasn't entirely original (see Douglas Englebart's 1960s demo. Apple made a huge contribution to modern GUIs. Check out the photographic record of the Lisa/Mac GUI development. Apple invented the pull-down menu whilst developing Lisa/Mac, they also invented the clipboard, and the idea of dragging and dropping files, to name just three things. All of these were totally copied by Microsoft, although they failed at it by replicating the menu bar at the top of every window, which some people like now, but was a total waste of screen space 25 years ago.
I don't think I would agree with your assertion.
Absolutely agree that in the very early 90s Mac OS was superior to the DOS/win31 combo. Moving past them, don't forget that even Win95 had preemptive multitasking. Multitasking is--and was--a big deal. Remember hearing the disk grind and not being able to switch applications? Remember copying a file to the network or a disk and not being able to do anything but wait for it to finish?
Well, Hank, I was the guy who wrote the report at Apple that recommended we buy NeXT. It was a simple choice, really, between Be and NeXTStep.
He says NeXTStep was their best option. Is he full of shit or are you?
There are no trails. There are no trees out here.
Apple's in-house OS, MacOS 8, made it to first developer release before Jobs killed it. This is not what Apple eventually released as "MacOS 8"; that was a warmed-over System 7. The real MacOS 8 was a completely new kernel, with protected memory and a CPU dispatcher, both of which the original MacOS lacked.
As others have pointed out, Jobs didn't kill Copland (the OS you're referring to). Apple's pre-Jobs executive team of Gil Amelio and Ellen Hancock did. As of about the time when that developer release was "released" (only to device driver developers because it was too dysfunctional for anybody working at a higher level, and actually too dysfunctional even to do device driver development on, but they had missed so many deadlines there was a lot of pressure to release something), Amelio and Hancock were convinced that Copland was going nowhere fast, would require a ground-up rewrite to meet specifications, and that the software development management at Apple was too broken to accomplish that rewrite. On top of which, Apple was in sorry shape financially and had a huge 3rd party developer confidence crisis to manage (it wasn't only insiders who knew that Apple's organization was a mess; a lot of Apple's current obsession with secrecy dates from those days when internal Apple political wars were routinely fought out in the press through deliberate leaks). So, they decided to cancel Copland and seek an outside OS for the next generation MacOS through merger or acquisition, because if they didn't have a credible OS story quick developers were going to bolt. The winner of that search was eventually NeXT.
And you know, they weren't wrong. If you'd ever tried to install and run that Copland developer release, you'd know why.
(Deep down, the original MacOS was like DOS - no memory management, no CPU dispatching, no I/O concurrency, and way too many low-level hacks into the OS at the app level. It had to fit in 64K, remember.)
It most certainly did have memory management. There were system calls for allocation and deallocation, and by making applications use handles (double indirect pointers) for allocated memory instead of raw pointers, the OS could even move allocated blocks around behind the application's back in order to defragment free space. Better yet, it could even temporarily unload some types of allocated memory resources not currently in use to make room for other things. A clumsy-yet-ingenious workaround for the lack of a MMU, in other words.
It was actually the relative sophistication of what they did in the 1980s which came back to bite them. A lot of it was a horrible fit to preemptive multitasking and MMU-based memory management.
The PowerPC transition had driven away many developers; most of the engineering apps were never ported, because the PowerPC had a shorter FPU length than the M68000 or Intel x86 lines, there were major data compatibility problems.
Oh, what a load of garbage. The engineering apps were ported early, and enthusiastically. The lack of 80-bit FP was no barrier because few applications truly depended on it (*), and the performance leap from 68K was extreme.
* - Have you noticed that these days x86 is slowly but surely migrating away from 80-bit FP too? It's only supported in x87, and the modern preferred way to do FP on x86 is through SSE (it's not just for vectors). SSE doesn't support 80-bit FP formats, only 64-bit. Also, you seem to be under the delusion that this creates a _data_ compatibility problem. It doesn't. 80-bit IEEE mode for both 68K and x86 was internal-only. When you load and store doubles, they're read and written in the 64-bit format. There is no valid in-memory 80-bit format. The 80-bit extended precision is only maintained so long as values stay inside processor registers. Soon as you write to memory, it gets rounded to 64-bit. So all that really happens is that some algorithms see less precision during calculation chains involving intermediate values which aren't written to memory.
Jobs' real job at the time was to cut a deal with Microsoft to keep Office on the Mac.
More trollish garbage...
Ah this again.
Apple had over a billion in the bank when Microsoft paid them off.
Paid them off because Microsoft and Intel were caught stealing Quicktime code. Shortly afterwards Apple was able to spend billions they didn't have while not touching their balance, somehow. Then Microsoft publicly paid them the $150 million. Apple was not that close to dead, at that point. It's made for some great stories though.
The (annotated) story, if you're actually interested.
Uh, he didn't. The company was close to bankruptcy when he took the reigns, and much of this happened after the announcement that Microsoft would put in $150 million. They also paid a farther, undisclosed sum which had quite a bit to do with legal battles, both patent infringement and stolen code.
Despite losing $850 million the year before, over a billion dollars in 1997--of which around 600 million was related to buying NeXT, and suffering a billion dollar drop in revenues between 1997-1998, Apple mysteriously managed to maintain its investments and actually accumulated cash.
It wasn't until 1998 that Apple began selling off its shares in ARM, and those sales took place over several years. Prior to that, how did Apple manage to spend nearly two billion dollars more than it earned across two years, lose 14% of its income, and still manage to sit on the same $1.2 billion in cash without pawning anything?
That depends on what you mean the competition. It was quite comparable to the Windows 95/98 OSes technically, but was technically far inferior to NT.
Windows 3.1 would be a much more accurate comparison. Co-operative multitasking, no memory protection, static disk cache, etc. Windows 9x was essentially a generation ahead of MacOS, NT another generation again.
The one thing that can't be done with keyboard and that drives me insane is switching to the non-default option in Yes/No boxes. Neither arrow keys, nor Tab works.
System Preferences -> Keyboard -> Keyboard Shortcuts, at the bottom you'll see Full Keyboard Access, select All Controls
You can also hit control-F7 to toggle it without going into System Preferences.
Now tab to the button you want to activate (click) and hit the space bar to activate the button. You can also shift-tab to move backwards in the tab order, which helps because usually the rightmost button is the default active one.
Some other shortcuts:There is a nice summary of various Mac keyboard shortcuts here:
Mac OS X keyboard shortcuts
Sapere aude!