what was one block replicated hundreds of times now becomes hundreds of blocks exhausting all storage.
What? Why would that happen?
If you have a block and a hundred COW pointers to it, and you modify one, then you get two blocks, with 99 references to the old one and one reference to the new one.
The expense of adding the support personnel to handle the 9 million driver problems that pop up and the stupid little proprietary pieces of kit on the motherboards are not really worth the effort.
Well, back in the OpenStep days, NeXT published a shopping list of what hardware they supported. They wouldn't have to promise to support everything the way that MS does. IIRC, they sold OpenStep 4.2 for $800 for a user seat, and $1500 for a developer seat.
If Apple offered an OS X license for non-Apple hardware, and priced it at whatever their margin is for a mid-range Mac, they'd be able to break into a lot of businesses where the customers don't get to pick their hardware. A lot of companies have company-wide purchasing contracts with Dell or HP, and the typical user doesn't have the authority to buy anything different. A lot of those same people though, could spend $500 for a software package on their own authority.
That being said, Apple still has to consider Microsoft when deciding whether to do this. If Apple offered Mac OS X on generic hardware, you can bet that MS would pull the plug on all Mac products immediately. Maybe we'll see this happen when iWork is ready to replace MS Office, but not yet.
Some of the most active GNUstep developers (both in terms of core development and building on top of GNUstep) joined the project since 2004, after getting their first experience with Cocoa.
When I talk about the wind being taken out of its sails, I'm referring to interest beyond GnuStep's implementors. I think you guys have done some fine work, but the interest among the users just isn't there.
a bunch of senior engineers in the hardware devision were let go or put in such a horrible situation that they left.
Don't know what you mean by "a bunch", butI heard that when Apple went from PPC to Intel, they let some ASIC engineers go since they didn't have to make their own motherboard chipsets anymore. Is that what you're referring to?
how about a marriage of GUI and Unix-y pipe goodness, where you could connect applications together in a GUI and have them do data flow type work - take the Unix filters approach one (or more) steps further?
I think you'd like Apple's Quartz Composer app. It's visual data-flow system for generating motion graphics, and its diagram editor would be great for plugging UNIX pipes and filters together.
I'm not sure if you still can find it, but I contributed some code to the MiscKit, which was a collection of code for NeXTSTEP developers. If you're developing code on the Mac, you might have some use for the sample code I did when I was at Apple, which I described here.
More recently, I posted a couple of little iPhone hacks here and here.
Nothing wrong with being white, male, or rich. The problem with the list is that it claims to be the top "influencers", and I really can't picture anyone asking themselves "gee, what would those guys think?" when deciding how to proceed with any open-source project.
Well, thanks, but I don't see how my experience with NeXTSTEP and the Mac make me any kind of hero, let alone an "open source hero". I've given a little bit of code away in my time, but it's not like it's any kind of mission I'm on.
As for GnuStep, it's a nice try, but once Apple and NeXT merged and the danger of NeXTSTEP vanishing altogether was alleviated, that really took the wind out of GnuStep's sails. The Linux crowd doesn't care about it, and the Mac crowd doesn't need it.
they should be focusing on replicating the NeXT/Apple experience.
I have to disagree with you on that. Trying to match any existing system is shooting too low. I remember when Visix was very proud of bringing "the Mac level of UI to UNIX" back around 1987 or so. I interviewed with them, and told them that unless they were looking to substantially exceed what the Mac offered, they shouldn't bother.
What I'd love to see happen with the Linux desktop is some serious re-thinkng of how a UI should be done. Trying to make it like Windows is tragic, and trying to make it like the Mac is just never going to be good enough.
There are processes for removing councilors who are doing a bad job, acting illegally or who lose the trust of the people who voted for them.
And criticizing them so they resign in shame doesn't qualify, in your opinion? Isn't this precisely why the citizens have the freedom of speech, to criticize their government?
The BBC is heavily critical of the government's restrictions on freedoms such as control orders and all the other various "orders" the Blair government introduced....and when was the last time that the BBC criticized the government for disarming the British people and rendering them unable to defend themselves from criminals?
Sure, the BBC will quibble about the details of how usurped power is used, but never against the increase of power itself.
Slashdot Mirror
Virtualization has been highly useful ever since IBM first shipped it on the System/360.
-jcr
Yeah, and now NeXT is king of the computer world?
Not king, but alive and gaining market share at a pretty good clip, not to mention an enviable position in smart phones.
I mean they went bankrupt and had to be rescued by Apple
NeXT didn't go bankrupt, they were operating profitably and sold out for about $400M.
-jcr
Microsoft doesn't really care if you run Windows or Mac OS on your generic hardware
Guess again. They pulled out all the stops to kill BeOS. They're very concerned about alternatives to windows.
-jcr
what was one block replicated hundreds of times now becomes hundreds of blocks exhausting all storage.
What? Why would that happen?
If you have a block and a hundred COW pointers to it, and you modify one, then you get two blocks, with 99 references to the old one and one reference to the new one.
-jcr
If I wanted to make something up to make myself look good, "open source hero" isn't in my top ten accolades I'd like to have.
-jcr
They have licensed Mac OS before, so we know exactly what the outcome of that would be.
They licensed it before on terms that weren't profitable to Apple. Who says that's the only way it can be done?
-jcr
The expense of adding the support personnel to handle the 9 million driver problems that pop up and the stupid little proprietary pieces of kit on the motherboards are not really worth the effort.
Well, back in the OpenStep days, NeXT published a shopping list of what hardware they supported. They wouldn't have to promise to support everything the way that MS does. IIRC, they sold OpenStep 4.2 for $800 for a user seat, and $1500 for a developer seat.
-jcr
I don't have hard data on it, but it doesn't need to be a terribly high percentage to be worth doing.
-jcr
If Apple offered an OS X license for non-Apple hardware, and priced it at whatever their margin is for a mid-range Mac, they'd be able to break into a lot of businesses where the customers don't get to pick their hardware. A lot of companies have company-wide purchasing contracts with Dell or HP, and the typical user doesn't have the authority to buy anything different. A lot of those same people though, could spend $500 for a software package on their own authority.
That being said, Apple still has to consider Microsoft when deciding whether to do this. If Apple offered Mac OS X on generic hardware, you can bet that MS would pull the plug on all Mac products immediately. Maybe we'll see this happen when iWork is ready to replace MS Office, but not yet.
-jcr
I can't tell if you're serious or a brilliant satirist.
-jcr
Oh, the complexity!
Great, so do you want to explain to the non-hackers in your office what that line does? Your example makes his point.
-jcr
Automator is fine if what you want to do is a linear sequence. QC's approach is rather more flexible.
-jcr
Some of the most active GNUstep developers (both in terms of core development and building on top of GNUstep) joined the project since 2004, after getting their first experience with Cocoa.
When I talk about the wind being taken out of its sails, I'm referring to interest beyond GnuStep's implementors. I think you guys have done some fine work, but the interest among the users just isn't there.
-jcr
a bunch of senior engineers in the hardware devision were let go or put in such a horrible situation that they left.
Don't know what you mean by "a bunch", butI heard that when Apple went from PPC to Intel, they let some ASIC engineers go since they didn't have to make their own motherboard chipsets anymore. Is that what you're referring to?
-jcr
how about a marriage of GUI and Unix-y pipe goodness, where you could connect applications together in a GUI and have them do data flow type work - take the Unix filters approach one (or more) steps further?
I think you'd like Apple's Quartz Composer app. It's visual data-flow system for generating motion graphics, and its diagram editor would be great for plugging UNIX pipes and filters together.
-jcr
The UI is more or less a solved problem, sort of like the controls of a car.
If you believe that, then by all means, enjoy what you can buy today. Heck, I know people who still live in EMACS.
-jcr
I'm not sure if you still can find it, but I contributed some code to the MiscKit, which was a collection of code for NeXTSTEP developers. If you're developing code on the Mac, you might have some use for the sample code I did when I was at Apple, which I described here.
More recently, I posted a couple of little iPhone hacks here and here.
-jcr
Nothing wrong with being white, male, or rich. The problem with the list is that it claims to be the top "influencers", and I really can't picture anyone asking themselves "gee, what would those guys think?" when deciding how to proceed with any open-source project.
-jcr
No, it's what happened at Apple in the mid 90's, that caused them to go through a near-death experience. Since then, they take coding very seriously.
-jcr
Well, thanks, but I don't see how my experience with NeXTSTEP and the Mac make me any kind of hero, let alone an "open source hero". I've given a little bit of code away in my time, but it's not like it's any kind of mission I'm on.
As for GnuStep, it's a nice try, but once Apple and NeXT merged and the danger of NeXTSTEP vanishing altogether was alleviated, that really took the wind out of GnuStep's sails. The Linux crowd doesn't care about it, and the Mac crowd doesn't need it.
they should be focusing on replicating the NeXT/Apple experience.
I have to disagree with you on that. Trying to match any existing system is shooting too low. I remember when Visix was very proud of bringing "the Mac level of UI to UNIX" back around 1987 or so. I interviewed with them, and told them that unless they were looking to substantially exceed what the Mac offered, they shouldn't bother.
What I'd love to see happen with the Linux desktop is some serious re-thinkng of how a UI should be done. Trying to make it like Windows is tragic, and trying to make it like the Mac is just never going to be good enough.
-jcr
it is the business side of the company that matters the most.
That's the kind of thinking that destroys companies.
-jcr
So, no votes from anyone who's actually, you know, writing any open source code?
-jcr
How can these people be "influential" when nobody's ever heard of them?
-jcr
There are processes for removing councilors who are doing a bad job, acting illegally or who lose the trust of the people who voted for them.
And criticizing them so they resign in shame doesn't qualify, in your opinion? Isn't this precisely why the citizens have the freedom of speech, to criticize their government?
-jcr
The BBC is heavily critical of the government's restrictions on freedoms such as control orders and all the other various "orders" the Blair government introduced. ...and when was the last time that the BBC criticized the government for disarming the British people and rendering them unable to defend themselves from criminals?
Sure, the BBC will quibble about the details of how usurped power is used, but never against the increase of power itself.
-jcr