RallyPoint has a "very clever design and has actually created something practical by focusing on a particular domain--the military," says Kortuem.
Translation: "Nobody other than the military had money to waste on this."
Really, there are too many demands on gloves and hands already to burden them with this. Sew this into jackets, arm bands, wrist bands, whatever, but not into gloves.
I just don't see the point of running Sugar on top of Windows or Windows on the OLPC. The only reasons for running Windows over Linux are related to drivers or Windows software. Let's look at those. Out of the box, Linux supports far more devices than Windows, and driver installers simply won't work well for the OLPC user community. And what Windows software does Negroponte think people will want to run on the OLPC? Sugar on Windows would require a lot of porting, and it's unlikely that it would work particularly well. If you want an educational software environment on Windows, get Squeak and eToys.
This is not even taking into account the fact that Microsoft would likely take advantage of any alliance with OLPC to destroy OLPC, like Intel tried, and like they have done with so many other business partners; Microsoft simply isn't a trustworthy business partner. Furthermore, it is reasonable and justifiable for volunteers to have the goal of exposing children to an alternative to the Microsoft Windows monopoly, rather than to further Microsoft's business interests; that's not "fundamentalism", it is long-term rational, economic self-interest. Few people would have volunteered if it had meant developing a free educational software platform for Windows.
So, Windows on the OLPC just doesn't make any sense, and Sugar on Windows also makes little sense. And an alliance with Microsoft doesn't make sense either. I certainly am not going to develop free software for some kind of get-them-hooked-early Windows educational platform. There are plenty of other projects that help children that I can volunteer for. Negroponte either needs to make a more convincing argument (good luck), or he can expect a mass exodus of volunteers; nobody is obligated to work for him or his vision.
In principle, an element doesn't have to be unstable in order to be able to participate in chain reactions (uranium is almost stable). If it were completely stable, you could probably use something else to start the chain reaction.
The main issue with the desktop experience is that the geeky programmers and designers assume too much from the average user.
In all fairness, so do Windows and Macintosh: there is plenty of software on those systems (even in the base system) that also asks questions that only geeky programmers can answer. People solve that by having geeky friends.
And what's wrong with that anyway? If it weren't for obscure, user-hostile software, nobody would talk to us geeks, so don't dare fix this:-)
Adobe should really update their web page. For Ubuntu, to install flash, you install the package through the package manager (you may or may not need to enable non-free). Ubuntu could try to work around this by handling Flash player like they do non-free drivers.
I am getting somewhat concerned, however, about the ubiquity of Flash... we really need to make the free versions work.
Actually, Google has something much better than Pages, namely Google Sites. Unfortunately, you only get it with Google Apps, and you still get Pages for your domain's home page.
I think they should scrap Pages, replace it with Sites, and add subversion access, like they do with the Code Wiki.
Speaking of the Code Wiki, that should probably also be replaced with Google Sites...
It's good that Congress recognizes the importance of reforming law on orphaned works, but this is the wrong way.
Congress should simply require copyright registration for all works every 10 years, for $1/registration, with submission of an electronic copy (or approximation thereof when that's not feasible). Once you don't renew, your work falls in the public domain. The list of all currently registered works should be published.
It's simple, reduces bureaucracy, is self-funding, and it creates legal certainty.
You claim that today's programming field is about big systems with many programmers. I claim that this is because management and technical leadership in most places isn't competent enough to divide up a big system in modular fashion and allow smaller, more flexible teams to solve the little problems before multiplying them all up to solve the big ones.
You're stating the obvious: of course management doesn't have the competency and leadership to do that. Who do you think does? You? God? Some mythical guy that's the company may be able to hire in 5 years?
In real-world projects, there's deadlines, there's a given group of managers and programmers that one has to work with, and there are ill-defined and constantly changing requirements. Nobody knows how to divide up such problems correctly among teams or how to solve many of the subproblems.
This explains how a small company with a few dozen employees can produce software that is better in every way than the competing offering from a larger company with hundreds of developers. [...] This elementary mathematics seems to be beyond a lot of senior management in the software business, and that has far more to do with the need to build monolithic systems maintained by zillions of developers than any actual project requirements do.
Senior management understands that just fine. What you don't seem to understand is that, although by sheer luck, sometimes a small company can get just the magic combination of really smart programmers that are 10x more productive than the average (you don't hear about the other 99.9%), but that one can't run a company assuming that that's the case.
If the product is due in 18 months, you need to get started now, with the programmers and managers you have. And they are going to be average. You can't hire really smart programmers because there simply aren't enough of them to go around.
You claim that today's programming field is not about clever tricks and fast algorithms. I claim that if more people understood these old-fashioned concepts, we would have much better software today.
Most software today isn't slow because of lack of clever tricks or lack of fast algorithms, it's slow because people get basics like memory management, communications, and system design wrong. And the most serious problem with software is not speed anyway, they are usability, crashes, data loss, and security holes. Your attitudes of "let's get out the C compiler and the algorithms textbook to speed this up" is quite common, and it's completely the wrong way to go, and (no offence) to me is the mark of an average and fairly inexperienced programmer.
No offence meant, but I think your preconceptions may be clouding your judgement here.
My preconceptions are based on working in industry for 25 years and building products (in addition to teaching).
Remember that some things in engineering and geophysics are still fully CPU bound and that even a 1% improvement saves you more than an hour per process per week
And there are a small number of specialists that deal with that sort of thing, just like there are specialists for Thai localization or GPU design. You can be a programmer and be such a specialist as well, but programmers in general aren't.
Mathematicians should know better than to have arguments about terms that haven't been defined. "Invented" and "discovered" aren't well-defined terms, and debating about whether mathematics is one or the other is pointless until one has defined them.
As for Roger Penrose, despite his mathematical insights, he seems to be prone to ill-founded leaps of logic and unfounded generalizations. Being able to reason logically in one domain apparently doesn't generalize to the consistent application of logic in other areas.
Knuth's view of programming seems to be that of clever tricks and fast algorithms. That may have been true when he got into the field, but it isn't anymore. Today, it's about creating big systems that need to be maintained by many programmers, not all of whom are as smart as Knuth. Algorithms come in libraries and are implemented by specialists. So, maybe Knuth doesn't need unit tests for his projects, but real-world projects do.
The bottom line is, you appear not to know what you're talking about
You know, you can posture, attempt to change the subject, and insult all you want, the basic fact remains:
Compared to other big companies, Apple's contributions to the research literature, sponsorship of academic research, and hiring of research-oriented graduates is negligible.
Nobody has to take my word for it; it's something anybody can easily veriy for themselves by going to the apple.com website and looking at the job postings and academic sponsorship programs, and checking for peer reviewed publications by Apple employees in computer science literature databases.
The data does not suggest that the bursts of enrollment we saw in the early 80's and late 90's are or should be the norm. When there is another technology boom, there will likely be another enrollment boom in CS.
So, in effect, you're saying that it doesn't really matter that Apple doesn't support computer scientists because nobody really needs them anyway.
I guess that's not surprising for someone coming from a company that still basically ships the same UI as it did in the 1980's, plus a few visual effects. I happen to believe that we need to make progress, both in areas of usability and software dependability; OS X sucks in both those areas just as badly as all the other platforms out there.
As far as what Apple has contributed, how about consumer-priced machines shipping with: Unix, stereographic display support, Apple/Genentech BLAST libraries, optimized FFT and imaging libraries, vector libraries, XGrid, zeroconf, launchd, and, yeah, sure, a Smalltalk-80 variant that can communicate with C++, C libs, several scripting languages, AppleScript, etc.
Funny, I had those before OS X came out. And most of what you list are simply Apple derivatives or Apple reimplementations of software that previously was already widely available for Linux and UNIX.
Further, if you'd actually -used- the Unix of 30 years ago, or worked on a system of any scale, I don't think you'd be so down on the progress that has been made since.
Yeah, sure, UNIX has improved. Too bad that Apple/NeXT wasn't part of that improvement: anything Apple considers remotely valuable or distinctive, they guard jealously. It took years just to get NeXT to comply with the GPL on the Objective C compiler. Apple could have contributed by releasing a core Objective C runtime and libraries, but, no, nothing.
In any case, what does UNIX software engineering have to do with computer science?
I've given my credentials. What are yours?
I'm not going to get into a pissing contest with you over credentials. Stick to the facts, please.
The problem is that you can't tell the system when to collect garbage. So you may find that during a critical section your VM stops everything and starts collecting up the garbage and throwing it out.
That is completely false. Even with a regular garbage collector, you can disable the garbage collector when you want/need to. Concurrent garbage collectors don't stop the system at all. And real-time garbage collectors give you precise upper limits for every memory allocation.
If you manage memory yourself, you can prove that your system will not do this.
That, too, is false, in two ways.
First, having a garbage collector doesn't prevent you from doing manual storage management.
Second, malloc/free and new/delete can have arbitrarily high latencies, and they do in practice.
The vast majority of computer science research takes place at universities
Of course. And Apple uses the fruits of that research extensively; without it, Apple wouldn't exist.
But computer scientists need jobs and funding, neither of which Apple provides. Maybe you haven't been paying attention, but academic computer science is in trouble, with student enrollment down about 50%. And why should people go into CS, if the only perspective companies like Apple give them is to become coding monkeys?
Personally, I think Apple is better off doing what it does well
Of course Apple is better off in the short term not investing in the next generation of computer scientists: it saves a lot of money.
which is making a great platform for others to do their work on
Oh, please, spare me the Apple marketing spiel, it's getting tiresome.
If Microsoft had done that, rather than investing so heavily in research, I dare say they would have done a lot more good for the world
And what good has Apple done for the world? What has Apple actually contributed other than make nice looking boxes for people with above-average incomes?
Take OS X: except for design and some tinkering, it's basically Smalltalk-80 on Mach. Where is the innovation? Where is the perspective for computer scientists if Apple's interfaces, programming, networking etc. are essentially 30 year old technology invented by other people?
While it's true that Apple does a lot less pure research than it did in the 90's
Apple didn't do much "pure research" even in the 90's, and there has been essentially nothing since. That's particularly galling because Apple has been using so much of the results of academic and other people's corporate research.
Microsoft is evil when it comes to their business practices, but they at least employ a lot of computer scientists and have some programs to support academic research. Apple just leeches off other people's results. Any academic computer scientist who supports Apple is a fool: if all computer companies were as stingy as Apple, computer science would cease to exist.
Show me a PC company that has a similar track record of visionary planning and impeccable execution.
What are your criteria? If it's market success, Microsoft is clearly better. If it's technology, there are plenty of companies that had better technology than Apple but that Apple harmed and killed just like Microsoft harmed and killed Apple and other competitors.
Yes, thanks for listing those examples for me. Why don't you add Java, OpenLook, NeWS, and NFS to that while you're at it. There are plenty more Sun engineering disasters.
Considering the numerous versions and variations, there's obviously some things that everyone just can't agree on for a licensing model.
The problem isn't that Sun uses a different license, the problem is that Sun reserves special rights for themselves that no other contributor has. They do that even with GPL'ed projects.
The chance of running into all kinds of memory problems is gradually outweighing the performance penalty you have to pay for garbage collection
It's a myth that there is a performance penalty for garbage collection; garbage collection usually has less overhead than doing the same kind of memory management with malloc/free.
The reason that garbage collection has a reputation for being slower is that once people have garbage collection, they get sloppy and wasteful. And languages using garbage collection often just aren't designed for efficiency (e.g., Java).
It would be nice to have a safe, garbage collected systems language, and such languages are possible. Unfortunately, all we get is Java and C#, two bloated languages that make writing efficient programs harder than eating tomato soup with chopsticks.
Apple invests in technology years in advance while the Dells and HPs are running their businesses on a quarterly basis.
Apple doesn't invest much in technology: they shuttered their research labs in the 1990's and outsourced a lot of the engineering. Since then, they are largely a design and marketing firm, with the occasional bit of software development.
Sun has the reigns to make certain that everything works
Yes, and that's the problem: Sun holds the reigns. Even if they were using that control well, it would be a problem. But Sun engineering and management doesn't know what they are doing, and that makes it worse. Of course, it doesn't matter for Solaris (who cares?), but Java has become so much a part of the computing infrastructure that Sun's continued screw-ups are a real problem.
Can you even download a copy of RedHat Enterprise Server and use it, with the ability to get patches for free? Last time I checked, you weren't allowed to do that - unless you count the CentOS project.
Who cares? The problem is not with companies making proprietary add-ons to open source software; if that was all Sun is doing (and they are doing plenty of that), I wouldn't complain.
The problem is with a company setting up projects such that they obtain special rights to contributions, and using tricks to retain control of projects that they should have lost long ago because of their incompetence.
Solaris >=2 is based on SunOS 5 which was derived from closed source system v. SunOS 4 was indeed based on some kind of BSD, but got killed by Solaris 2 long time ago
Solaris >=2 still contains plenty of BSD code. Furthermore, System V contains stuff derived from Berkeley as well.
Slashdot Mirror
RallyPoint has a "very clever design and has actually created something practical by focusing on a particular domain--the military," says Kortuem.
Translation: "Nobody other than the military had money to waste on this."
Really, there are too many demands on gloves and hands already to burden them with this. Sew this into jackets, arm bands, wrist bands, whatever, but not into gloves.
The unit runs the SSD Linux operating system, which straps NetBSD userland functions onto the Linux kernel
I'm sorry, but that just seems completely wrong... or, rather, backwards.
I just don't see the point of running Sugar on top of Windows or Windows on the OLPC. The only reasons for running Windows over Linux are related to drivers or Windows software. Let's look at those. Out of the box, Linux supports far more devices than Windows, and driver installers simply won't work well for the OLPC user community. And what Windows software does Negroponte think people will want to run on the OLPC? Sugar on Windows would require a lot of porting, and it's unlikely that it would work particularly well. If you want an educational software environment on Windows, get Squeak and eToys.
This is not even taking into account the fact that Microsoft would likely take advantage of any alliance with OLPC to destroy OLPC, like Intel tried, and like they have done with so many other business partners; Microsoft simply isn't a trustworthy business partner. Furthermore, it is reasonable and justifiable for volunteers to have the goal of exposing children to an alternative to the Microsoft Windows monopoly, rather than to further Microsoft's business interests; that's not "fundamentalism", it is long-term rational, economic self-interest. Few people would have volunteered if it had meant developing a free educational software platform for Windows.
So, Windows on the OLPC just doesn't make any sense, and Sugar on Windows also makes little sense. And an alliance with Microsoft doesn't make sense either. I certainly am not going to develop free software for some kind of get-them-hooked-early Windows educational platform. There are plenty of other projects that help children that I can volunteer for. Negroponte either needs to make a more convincing argument (good luck), or he can expect a mass exodus of volunteers; nobody is obligated to work for him or his vision.
In principle, an element doesn't have to be unstable in order to be able to participate in chain reactions (uranium is almost stable). If it were completely stable, you could probably use something else to start the chain reaction.
The main issue with the desktop experience is that the geeky programmers and designers assume too much from the average user.
:-)
In all fairness, so do Windows and Macintosh: there is plenty of software on those systems (even in the base system) that also asks questions that only geeky programmers can answer. People solve that by having geeky friends.
And what's wrong with that anyway? If it weren't for obscure, user-hostile software, nobody would talk to us geeks, so don't dare fix this
Adobe should really update their web page. For Ubuntu, to install flash, you install the package through the package manager (you may or may not need to enable non-free). Ubuntu could try to work around this by handling Flash player like they do non-free drivers.
I am getting somewhat concerned, however, about the ubiquity of Flash... we really need to make the free versions work.
Actually, Google has something much better than Pages, namely Google Sites. Unfortunately, you only get it with Google Apps, and you still get Pages for your domain's home page.
I think they should scrap Pages, replace it with Sites, and add subversion access, like they do with the Code Wiki.
Speaking of the Code Wiki, that should probably also be replaced with Google Sites...
It's good that Congress recognizes the importance of reforming law on orphaned works, but this is the wrong way.
Congress should simply require copyright registration for all works every 10 years, for $1/registration, with submission of an electronic copy (or approximation thereof when that's not feasible). Once you don't renew, your work falls in the public domain. The list of all currently registered works should be published.
It's simple, reduces bureaucracy, is self-funding, and it creates legal certainty.
You claim that today's programming field is about big systems with many programmers. I claim that this is because management and technical leadership in most places isn't competent enough to divide up a big system in modular fashion and allow smaller, more flexible teams to solve the little problems before multiplying them all up to solve the big ones.
You're stating the obvious: of course management doesn't have the competency and leadership to do that. Who do you think does? You? God? Some mythical guy that's the company may be able to hire in 5 years?
In real-world projects, there's deadlines, there's a given group of managers and programmers that one has to work with, and there are ill-defined and constantly changing requirements. Nobody knows how to divide up such problems correctly among teams or how to solve many of the subproblems.
This explains how a small company with a few dozen employees can produce software that is better in every way than the competing offering from a larger company with hundreds of developers. [...] This elementary mathematics seems to be beyond a lot of senior management in the software business, and that has far more to do with the need to build monolithic systems maintained by zillions of developers than any actual project requirements do.
Senior management understands that just fine. What you don't seem to understand is that, although by sheer luck, sometimes a small company can get just the magic combination of really smart programmers that are 10x more productive than the average (you don't hear about the other 99.9%), but that one can't run a company assuming that that's the case.
If the product is due in 18 months, you need to get started now, with the programmers and managers you have. And they are going to be average. You can't hire really smart programmers because there simply aren't enough of them to go around.
You claim that today's programming field is not about clever tricks and fast algorithms. I claim that if more people understood these old-fashioned concepts, we would have much better software today.
Most software today isn't slow because of lack of clever tricks or lack of fast algorithms, it's slow because people get basics like memory management, communications, and system design wrong. And the most serious problem with software is not speed anyway, they are usability, crashes, data loss, and security holes. Your attitudes of "let's get out the C compiler and the algorithms textbook to speed this up" is quite common, and it's completely the wrong way to go, and (no offence) to me is the mark of an average and fairly inexperienced programmer.
No offence meant, but I think your preconceptions may be clouding your judgement here.
My preconceptions are based on working in industry for 25 years and building products (in addition to teaching).
Remember that some things in engineering and geophysics are still fully CPU bound and that even a 1% improvement saves you more than an hour per process per week
And there are a small number of specialists that deal with that sort of thing, just like there are specialists for Thai localization or GPU design. You can be a programmer and be such a specialist as well, but programmers in general aren't.
Mathematicians should know better than to have arguments about terms that haven't been defined. "Invented" and "discovered" aren't well-defined terms, and debating about whether mathematics is one or the other is pointless until one has defined them.
As for Roger Penrose, despite his mathematical insights, he seems to be prone to ill-founded leaps of logic and unfounded generalizations. Being able to reason logically in one domain apparently doesn't generalize to the consistent application of logic in other areas.
Knuth's view of programming seems to be that of clever tricks and fast algorithms. That may have been true when he got into the field, but it isn't anymore. Today, it's about creating big systems that need to be maintained by many programmers, not all of whom are as smart as Knuth. Algorithms come in libraries and are implemented by specialists. So, maybe Knuth doesn't need unit tests for his projects, but real-world projects do.
To properly study the physics of whipped cream in zero G, I need the Swedish woman's volleyball team as... assistants.
The bottom line is, you appear not to know what you're talking about
You know, you can posture, attempt to change the subject, and insult all you want, the basic fact remains:
Compared to other big companies, Apple's contributions to the research literature, sponsorship of academic research, and hiring of research-oriented graduates is negligible.
Nobody has to take my word for it; it's something anybody can easily veriy for themselves by going to the apple.com website and looking at the job postings and academic sponsorship programs, and checking for peer reviewed publications by Apple employees in computer science literature databases.
The data does not suggest that the bursts of enrollment we saw in the early 80's and late 90's are or should be the norm. When there is another technology boom, there will likely be another enrollment boom in CS.
So, in effect, you're saying that it doesn't really matter that Apple doesn't support computer scientists because nobody really needs them anyway.
I guess that's not surprising for someone coming from a company that still basically ships the same UI as it did in the 1980's, plus a few visual effects. I happen to believe that we need to make progress, both in areas of usability and software dependability; OS X sucks in both those areas just as badly as all the other platforms out there.
As far as what Apple has contributed, how about consumer-priced machines shipping with: Unix, stereographic display support, Apple/Genentech BLAST libraries, optimized FFT and imaging libraries, vector libraries, XGrid, zeroconf, launchd, and, yeah, sure, a Smalltalk-80 variant that can communicate with C++, C libs, several scripting languages, AppleScript, etc.
Funny, I had those before OS X came out. And most of what you list are simply Apple derivatives or Apple reimplementations of software that previously was already widely available for Linux and UNIX.
Further, if you'd actually -used- the Unix of 30 years ago, or worked on a system of any scale, I don't think you'd be so down on the progress that has been made since.
Yeah, sure, UNIX has improved. Too bad that Apple/NeXT wasn't part of that improvement: anything Apple considers remotely valuable or distinctive, they guard jealously. It took years just to get NeXT to comply with the GPL on the Objective C compiler. Apple could have contributed by releasing a core Objective C runtime and libraries, but, no, nothing.
In any case, what does UNIX software engineering have to do with computer science?
I've given my credentials. What are yours?
I'm not going to get into a pissing contest with you over credentials. Stick to the facts, please.
The problem is that you can't tell the system when to collect garbage. So you may find that during a critical section your VM stops everything and starts collecting up the garbage and throwing it out.
That is completely false. Even with a regular garbage collector, you can disable the garbage collector when you want/need to. Concurrent garbage collectors don't stop the system at all. And real-time garbage collectors give you precise upper limits for every memory allocation.
If you manage memory yourself, you can prove that your system will not do this.
That, too, is false, in two ways.
First, having a garbage collector doesn't prevent you from doing manual storage management.
Second, malloc/free and new/delete can have arbitrarily high latencies, and they do in practice.
The vast majority of computer science research takes place at universities
Of course. And Apple uses the fruits of that research extensively; without it, Apple wouldn't exist.
But computer scientists need jobs and funding, neither of which Apple provides. Maybe you haven't been paying attention, but academic computer science is in trouble, with student enrollment down about 50%. And why should people go into CS, if the only perspective companies like Apple give them is to become coding monkeys?
Personally, I think Apple is better off doing what it does well
Of course Apple is better off in the short term not investing in the next generation of computer scientists: it saves a lot of money.
which is making a great platform for others to do their work on
Oh, please, spare me the Apple marketing spiel, it's getting tiresome.
If Microsoft had done that, rather than investing so heavily in research, I dare say they would have done a lot more good for the world
And what good has Apple done for the world? What has Apple actually contributed other than make nice looking boxes for people with above-average incomes?
Take OS X: except for design and some tinkering, it's basically Smalltalk-80 on Mach. Where is the innovation? Where is the perspective for computer scientists if Apple's interfaces, programming, networking etc. are essentially 30 year old technology invented by other people?
While it's true that Apple does a lot less pure research than it did in the 90's
Apple didn't do much "pure research" even in the 90's, and there has been essentially nothing since. That's particularly galling because Apple has been using so much of the results of academic and other people's corporate research.
Microsoft is evil when it comes to their business practices, but they at least employ a lot of computer scientists and have some programs to support academic research. Apple just leeches off other people's results. Any academic computer scientist who supports Apple is a fool: if all computer companies were as stingy as Apple, computer science would cease to exist.
Show me a PC company that has a similar track record of visionary planning and impeccable execution.
What are your criteria? If it's market success, Microsoft is clearly better. If it's technology, there are plenty of companies that had better technology than Apple but that Apple harmed and killed just like Microsoft harmed and killed Apple and other competitors.
Incompetence? Dtrace? ZFS? Zones / Containers? Ultra SPARC T1, T2, T2+?
Yes, thanks for listing those examples for me. Why don't you add Java, OpenLook, NeWS, and NFS to that while you're at it. There are plenty more Sun engineering disasters.
Considering the numerous versions and variations, there's obviously some things that everyone just can't agree on for a licensing model.
The problem isn't that Sun uses a different license, the problem is that Sun reserves special rights for themselves that no other contributor has. They do that even with GPL'ed projects.
The chance of running into all kinds of memory problems is gradually outweighing the performance penalty you have to pay for garbage collection
It's a myth that there is a performance penalty for garbage collection; garbage collection usually has less overhead than doing the same kind of memory management with malloc/free.
The reason that garbage collection has a reputation for being slower is that once people have garbage collection, they get sloppy and wasteful. And languages using garbage collection often just aren't designed for efficiency (e.g., Java).
It would be nice to have a safe, garbage collected systems language, and such languages are possible. Unfortunately, all we get is Java and C#, two bloated languages that make writing efficient programs harder than eating tomato soup with chopsticks.
Apple invests in technology years in advance while the Dells and HPs are running their businesses on a quarterly basis.
Apple doesn't invest much in technology: they shuttered their research labs in the 1990's and outsourced a lot of the engineering. Since then, they are largely a design and marketing firm, with the occasional bit of software development.
The old Apple of the late 80s basically stopped the frantic pace of upgrades,
Well, it's not like they had a choice; their operating system was a mess.
Now the tables are turned, and Microsoft is the one coasting along on past performance, allowing Apple to catch up and surpass it.
I think Apple is also going to be less and less relevant.
Sun has the reigns to make certain that everything works
Yes, and that's the problem: Sun holds the reigns. Even if they were using that control well, it would be a problem. But Sun engineering and management doesn't know what they are doing, and that makes it worse. Of course, it doesn't matter for Solaris (who cares?), but Java has become so much a part of the computing infrastructure that Sun's continued screw-ups are a real problem.
Can you even download a copy of RedHat Enterprise Server and use it, with the ability to get patches for free? Last time I checked, you weren't allowed to do that - unless you count the CentOS project.
Who cares? The problem is not with companies making proprietary add-ons to open source software; if that was all Sun is doing (and they are doing plenty of that), I wouldn't complain.
The problem is with a company setting up projects such that they obtain special rights to contributions, and using tricks to retain control of projects that they should have lost long ago because of their incompetence.
Solaris >=2 is based on SunOS 5 which was derived from closed source system v. SunOS 4 was indeed based on some kind of BSD, but got killed by Solaris 2 long time ago
Solaris >=2 still contains plenty of BSD code. Furthermore, System V contains stuff derived from Berkeley as well.
Without BSD, Sun wouldn't even exist.