And Xerox borrowed the mouse, overlapping windows, menus, direct manipulation and other things from other places as well. (Many papers, researchers, Apps, etc., that had existed for 5 or 10 years before PARC). They just brought the ideas together and furthered them.... just like Apple did.
Yes, in response to someone who was saying Objective C sucked because it didn't have garbage collection.
ObjC doesn't "suck" because it doesn't have garbage collection. That being said, automatic (or more automatic) garbage collection is better than more manual. ObjC does feel more anachronistic than Java (in that area) because of that.
Java is not yet "realtime", but then 99% of real apps aren't. I doubt your commercial apps are realtime (in the traditional sense of the word). And Moore's law cures all evils. In a few years more automatic and higher level solutions will replace the lower level ones. It is just during these transitions, people whine "too slow, I want more control", then a couple years later 99% of them switch. I've seen it happen from machine code to asm, to C, to C++, and now going to Java.
And certainly nobody has made the case that Objective C is a pale imitation of the state of OO from the 70s, as was the original claim I'm responding to.
Some things might be overstated, but ObjC has plenty of issues that were solved before and after it. Same with every other language. I do think automatic garbage collectors have advantages (in costs, ease, etc.)... we can debate the merits of loose binding (personally, I think it is better, but there is an argument to be made for stronger casting), the whole MI issues (again, I think that designs/languages shouldn't have it, but the argument can be made it is more OOD), and so on. So there are plenty of legitimate points against ObjC as very OOD. Don't take it personal, it is just fact.
I think the biggest issue against ObjectiveC is that it is irrelevant. Except for Apple, no one is using it. So unless there's a huge compelling advantage for it, you should probably just be standard. Smalltalk and other had enough advantages to justify being different (and they still weren't a huge success other than academic experimentation and some verticals). ObjectiveC just doesn't have that... and I don't feel like it has evolved at all in the last 15 years. (Both Java and C++ have). So ObjC is hurting the Mac more than it is helping it. (IMHO). That doesn't mean that they can afford the time/energy to replace it, but just because Apple is using it, doesn't mean it is the best solution out there.
Actually, I do think NeXT had quite a bit of WIH (was invented here) attitude towards Apple technologies (that they hadn't created). Some of it was openness, some was standards, some was ego, some was just to change things, some was to get people back, and so on.
But you can't tell me that if you were really open minded and looking for the best technologies that you couldn't have borrowed more from all the source code that was available. Puhlease.
Did you know that Apple wasn't even going to create carbon? They were going to use the Lattitude stuff, and make everyone port: it was only because one team was able to get through the WIH attitude that Apple considered it. There are many other ideas and technologies that NeXT threw out, not because they wren't valid, but because they weren't NeXTs or because they were Apples.
So there's give and take on both sides. I don't think NeXT is flawless, just generally a *little* better. I think they could be better still, and could have handled the migration a lot better. And just because I disagree with you, doesn't mean that I buy my own press (either way).
Man, it's always a bad sign when an author has to respond defensively to every post about his article.
First, it isn't EVERY post, it is only a few.
Second I've read/. for years. (off and on, lurked and posted). According to you, Authors aren't allowed to feedback or use Slash -- it is only for rodents who are slandering them or taking things out of context? This is only a place for the uninformed to heckle people without trying to understand or confront their victims? While that fits the format sometimes, I think it would add more maturity to *SOME* of these threads, and maybe some of the people, to get more feedback from authors.
Slash has always been a place with some brilliant insights and informed people, mixed with a lot of pissed off little trolls that vent. I'd rather see more bidirectional feedback, rather than just a bunch of DuDe, what A toTal Lamo, M@N. LeTs DIS the GuY beHInD hiS bAcK! Which do you want to be? Oh wait, I think you already answered that.
No offense, but I am always amazed at some posters inability or unwillingness to actually read what others write. Like I said, "Late 70's", and you said I said, "Early 70's".... I said "on-line" and you assumed internet. (Just an observation).
Actually, "on-line" back then was loose networks, private ones and BBS's. I was on DEC-NET and actively involved in many BBS's. My dad was also a VP at Boeing Computer Services, and so I was the only kid on my block with Cray time (which he was leasing). I was also a bit of a hack, crack and phreak, but nothing big (I hacked my colleges computers in 82, and a few things like that). But I didn't want to be a criminal, or caught, so straightened my ways pretty quick.
I'm not familiar with Copland... what were its features? How do its features compare to OS X? What Copland technology did Apple abandon with OS X?
That is a WHOLE different article. A good one. But probably beyond the scope of the decision. As for what technologies were abandoned: all of them.
Apple had a streams solution (OpenTransport) that programmers generally preferred to sockets. There was a higher level 3D API that had many apps. There was the GX engine, which had problems in the first iteration (mainly due to memory requirements), but things had just gotten worked out (mostly), so it was thrown out for something that took far far more memory. There was a different kernel and driver model. Apple's was faster for at least some things (and quite a few I hear) from people who were on the Kernel team. It also required new drivers, but far less changes to get them working. There was a whole different GUI, much more advanced in UI, but less advanced in look and graphics. There was a different way of dealing with many things. Again, totally different article that I might write someday. I should probably get it booting again first though...
I don't agree wiht your staements about "no big innovations". I think it beat MS and UNIX to many innovations. But was of course behind in others. And I do agree that Apple needed fresh attitude/blood -- but there are costs for everything...
I did Rhapsody training and did some YellowBox stuff (and NeXTSTEP/OpenStep before that). Cocoa is nice. Not as nice as many NeXTies think, and many of them don't realize what we gave up. But in general I think it is good; and do think it was a win.
But if Cocoa was that good, it should have been a lot easier to put it on NuKernel + Carbon, rather than reworking everything else to fit to it. (Just my perseption there).
Yes I do. "stupid" is a harsh word, which is why I used it - just as the author did to describe Apple mgmt.
I am the author. The difference is that I knew Apple management (and their decisions) and had dealt with them directly and indirectly for 15+ years before I said they were being stupid (about that one thing). You've known me how long and base it on what, one thing I said that you didn't like in one article?
but unfortunately it had to go just for Apple to remain afloat...Apple made the right decision.
The problem is that they are not all mutually exclusive or tied in. Apple could have kept one and not died because of it. So what we need to do, if we want to honestly review what can be learned, is not say blindly "it was all or nothing".
Apple could have gone the Copland(via Carbon) route and thrived. Or it did go the OS X route and thrived. Overall, I'm more happy with OSX than unhappy, and use it daily. But I'm not a blind bigot that's going to pretend that means it was the only path to success, or that it is perfect, or there were no mistakes made that we can't learn from, and so on.
I think that Appple did have to cut some (many) of the technologies. All? Not so sure.
I think many of the technologies could have been integrated and saved time : but we replaced Apple NIH with NeXT's. So we couldn't not for technical reasons, but for ego ones. Just because NeXT was right about some things doesn't mean they are right about everything.
So there's some balances there, and it isn't all or nothing... not with Apple's old management (saying 100% compatibility), and not with Apple's new management (saying everything's gotta be new and OUR way)... if you can't challenge your assumptions, then you miss opportunities to learn.
Ah, ye who know so much about me, explain more.
I left PowerSchool (not Apple), as we sold it to Apple for $62M. Pretty good mismanagement taking an company from 7 to 200+ employees, and from $1M to $62M in 2 years.
I chose not to work for a large organization (even if it is Apple), because I don't like the politics. This had already trended that way, and after a couple years of 70+ hour weeks it was time to go...
But I like how you can see this as Apple bashing (saying they could have succeeded either way), or me as Apple basher (because I have some problems with the quality on one project; WebObjects). Such is life: people see what they want to.
ust out of curiosity: is there any reason why every now and then, slashdot posters obviously insult people merely because they have a different opinion?
It is called immaturity. I used to do that to in the late 70's, when I was a teen and first on-line, and raging at all the world (especially authority).
Sad, because they actually damage the minor points they might have in there...
Of course there are actually many reasons. BitGeek has some personal thing going because I answered some of his questions once, and then disagreed later (and demonstrated his mistakes in reasoning). Now he just just tries to add libel and slander to his list of other issues.
A tad inaccurate...
I don't ad the [teako170.com] crap everywhere...
What I do have is keyword substition on the little CMS thing I wrote... and haven't yet tuned. Since Apple is a keyword, it is linking you to the stories about Apple (which happens to be one).
I plan on expanding that, to auto-interlink a lot of definitions, stories and information.
I don't find the underline horribly intrusive myself. But I do think the feature needs tuning. (It has the ability to only do the first instance in the document, and I probably need to throw that switch).
This is just absolutely false. The NeXT kernel (Mach/BSD) and Cocoa were ported very quickly
But they had no value until they adapted to their market, which took all the time.
And actually they really weren't. The software compiled and ran, that doesn't mean they were ready to ship. What about drivers? What about Apps? Apple had to rewrite all the utilities and environments that ran on them to make them useful. WIth Carbon/Copland this would have been much easier. NeXT-Apple threw away things like QD3D, QDGX and so on and started over. The other apps would have run. I'm not saying that was the wrong choice: but it is what cost the time. So it wasn't really running (well) until Jaguar.
NeXT delivered on its promises, it's just that Apple's requirements changed.
Ummm, Apple was NeXT and was the one that changed the requirements.
But I tend to agree that NeXT could have delivered much sooner, if they wanted to kill themselves (since they were then Apple). But the plan was brain-dead from the start; and they either knew it, or should have known it, from day one.
As for why the full move wouldn't have been complete until Gershwin, that was because Apple management wouldn't let Copland team to Carbon and required 100% compatibility. (The point of the article).
Sorta. It is one big what if. There is some ranting at the ranters that claim there's only one true path to success. I've never bought that. Many paths can succeed. OS X was one. But I think Copland could have as well.
I wish he had more information to back up his claims. The article just seemed to be overly bitter calling NeXT "liars" and claiming that the engineers were hit with unrealistic expectations.
What would you like to prove my claims (what claims are in question)?
As for NeXT and liars. I called them marketers, and said I can't tell the difference. I have no doubt Apple people, Be people, or others would have done the same. Marketing is coloring the truth, and selective truths... in engineering that's lying.
I think it takes everyone on a project to make it fail.
I don't. I think one bad decision maker can drive a project into the ground.
I really doubt that if Apple had went with Copland so many alpha geeks would be flocking to Mac
Nice to you think so highly of yourself, and so low of everyone else. Hard to say, but you're probably right (in implication) that you get the attention of UNIX types by conformity. But sometimes change and challenge is good too. Would the world be a better place if we didn't have Be's and Apple's to try to challenge assumptions? I think we get to find out...
Zealotry accusations aside, I always felt that we should fix underlying problems and not superficial ones. The Mac dealt with many (as far as UI), and missed on others. Some saw me promoting the successes as ignoring the failures, even when I called the failures out too. Such is life.
I've developed for 20+ years, on many platforms (VAX, AOS/VS, UNIX, Mac, Windows, DOS, CP/M, and so on). I see strengths and weaknesses in all. But for users, nothing else has yet come close to a Mac.
As for my assumptions : Apple had the money to complete Copland, what they didn't have was time. The reason they ran out of time, was because they'd hyped and missed deliveries, and because they'd shot at the wrong target (100% compatibility). Carbon was much more reasonable than 100% compatibility. Many were saying it, and had said it for years. But they could change the target (or hit the impossible one they were told to shoot at).
And Jobs killed many bad projects, But he also killed many good ones, and didn't really descriminate. He provided a vision, which is better than no vision. But his vision was as biased (or more so) than the ones before it.
No I don't want Apple to go back. But I think we can learn from both the positives and negatives of Steve and the NeXT take-over. Some see that as "grousing", others see it as trying to learn or remember accurately. You can choose which...
Apparently the author is too stupid to realize Apple was sinking. Fast.
Apple was sinking in perception and marketplace. We're talking technology here (there). The point is that if Apple had just dropped the 100% compatibility thing, they would have had Carbon on NuKernal, long before NeXT could deliver OSX. I get the marketing issues, do you get the engineering ones? Carbon was easy (relatively) it was 100% compatibility that was killing the project...
Re:Using rosecolored glasses to look into the past
on
Copland/Gershwin vs. NeXT
·
· Score: 2, Informative
The reason that it had semi-cooperative tasking (actually engine was preemptive, UI thread was cooperative) was because it had to have 100% compatibility. If you took out that requirement, then you had Carbon/Gershwin. That was the point.
Slashdot Mirror
And Xerox borrowed the mouse, overlapping windows, menus, direct manipulation and other things from other places as well. (Many papers, researchers, Apps, etc., that had existed for 5 or 10 years before PARC). They just brought the ideas together and furthered them.... just like Apple did.
Java is not yet "realtime", but then 99% of real apps aren't. I doubt your commercial apps are realtime (in the traditional sense of the word). And Moore's law cures all evils. In a few years more automatic and higher level solutions will replace the lower level ones. It is just during these transitions, people whine "too slow, I want more control", then a couple years later 99% of them switch. I've seen it happen from machine code to asm, to C, to C++, and now going to Java.
Some things might be overstated, but ObjC has plenty of issues that were solved before and after it. Same with every other language. I do think automatic garbage collectors have advantages (in costs, ease, etc.)... we can debate the merits of loose binding (personally, I think it is better, but there is an argument to be made for stronger casting), the whole MI issues (again, I think that designs/languages shouldn't have it, but the argument can be made it is more OOD), and so on. So there are plenty of legitimate points against ObjC as very OOD. Don't take it personal, it is just fact.
I think the biggest issue against ObjectiveC is that it is irrelevant. Except for Apple, no one is using it. So unless there's a huge compelling advantage for it, you should probably just be standard. Smalltalk and other had enough advantages to justify being different (and they still weren't a huge success other than academic experimentation and some verticals). ObjectiveC just doesn't have that... and I don't feel like it has evolved at all in the last 15 years. (Both Java and C++ have). So ObjC is hurting the Mac more than it is helping it. (IMHO). That doesn't mean that they can afford the time/energy to replace it, but just because Apple is using it, doesn't mean it is the best solution out there.
Actually, I do think NeXT had quite a bit of WIH (was invented here) attitude towards Apple technologies (that they hadn't created). Some of it was openness, some was standards, some was ego, some was just to change things, some was to get people back, and so on.
But you can't tell me that if you were really open minded and looking for the best technologies that you couldn't have borrowed more from all the source code that was available. Puhlease.
Did you know that Apple wasn't even going to create carbon? They were going to use the Lattitude stuff, and make everyone port: it was only because one team was able to get through the WIH attitude that Apple considered it. There are many other ideas and technologies that NeXT threw out, not because they wren't valid, but because they weren't NeXTs or because they were Apples.
So there's give and take on both sides. I don't think NeXT is flawless, just generally a *little* better. I think they could be better still, and could have handled the migration a lot better. And just because I disagree with you, doesn't mean that I buy my own press (either way).
Second I've read
Slash has always been a place with some brilliant insights and informed people, mixed with a lot of pissed off little trolls that vent. I'd rather see more bidirectional feedback, rather than just a bunch of DuDe, what A toTal Lamo, M@N. LeTs DIS the GuY beHInD hiS bAcK! Which do you want to be? Oh wait, I think you already answered that.
No offense, but I am always amazed at some posters inability or unwillingness to actually read what others write. Like I said, "Late 70's", and you said I said, "Early 70's".... I said "on-line" and you assumed internet. (Just an observation). Actually, "on-line" back then was loose networks, private ones and BBS's. I was on DEC-NET and actively involved in many BBS's. My dad was also a VP at Boeing Computer Services, and so I was the only kid on my block with Cray time (which he was leasing). I was also a bit of a hack, crack and phreak, but nothing big (I hacked my colleges computers in 82, and a few things like that). But I didn't want to be a criminal, or caught, so straightened my ways pretty quick.
A tad inaccurate... I don't ad the [teako170.com] crap everywhere... What I do have is keyword substition on the little CMS thing I wrote... and haven't yet tuned. Since Apple is a keyword, it is linking you to the stories about Apple (which happens to be one). I plan on expanding that, to auto-interlink a lot of definitions, stories and information. I don't find the underline horribly intrusive myself. But I do think the feature needs tuning. (It has the ability to only do the first instance in the document, and I probably need to throw that switch).
Zealotry accusations aside, I always felt that we should fix underlying problems and not superficial ones. The Mac dealt with many (as far as UI), and missed on others. Some saw me promoting the successes as ignoring the failures, even when I called the failures out too. Such is life. I've developed for 20+ years, on many platforms (VAX, AOS/VS, UNIX, Mac, Windows, DOS, CP/M, and so on). I see strengths and weaknesses in all. But for users, nothing else has yet come close to a Mac. As for my assumptions : Apple had the money to complete Copland, what they didn't have was time. The reason they ran out of time, was because they'd hyped and missed deliveries, and because they'd shot at the wrong target (100% compatibility). Carbon was much more reasonable than 100% compatibility. Many were saying it, and had said it for years. But they could change the target (or hit the impossible one they were told to shoot at). And Jobs killed many bad projects, But he also killed many good ones, and didn't really descriminate. He provided a vision, which is better than no vision. But his vision was as biased (or more so) than the ones before it. No I don't want Apple to go back. But I think we can learn from both the positives and negatives of Steve and the NeXT take-over. Some see that as "grousing", others see it as trying to learn or remember accurately. You can choose which...
The reason that it had semi-cooperative tasking (actually engine was preemptive, UI thread was cooperative) was because it had to have 100% compatibility. If you took out that requirement, then you had Carbon/Gershwin. That was the point.