Typing out 1000 different switches and options by hand...
Doesn't happen. You type "make" and it does it all for you. The first program I wrote for VAXC on VMS was an implementation of Make... and this was back before the Macintosh existed.
In the rare case where you need command line options, you put them in a Make variable and it uses them everywhere.
Then for the output, there's this great program called "error" that I first ran into at Berkeley in 1981. You pipe your output through that, and it puts all the errors in the files as comments and can even toss you into the editor right at the first one.
Most of the great features of GUI compiler tools already existed in command line tools back before there really was a general purpose GUI to implement them in.
I can't imagine how anyone could have demoed any of the Smalltalk systems without ever overlapping any windows. Smalltalk-72 had them, right from a very early stage, and I honestly can't see how someone giving a demo could have deliberately lined up two windows without overlapping them, and never actually overlapped any windows, unless they were trying very hard to avoid revealing that capability.
Which would be really weird, because the oldest published photos of Smalltalk-72 screens already showed overlapping windows.
Of the three items you listed, only one is actually an Apple innovation... and that one is an innovation we'd be better off without: pull-down menus.
Pull-down menus were a hack to let them have a single-button mouse. Everyone else used contextual menus, and even Apple has in a backhanded way adopted them... and before you go all Fitt's Law on me, don't forget that there are *5* "best targets" on the screen for Fitt's Law, and "right under the mouse" is one of them.
Overlapping windows were NOT an Apple innovation. Smalltalk-72 had overlapping windows Smalltalk-76 had overlapping windows. Smalltalk-80 and Interlisp-D and the Xerox Star office system had overlapping windows. The Star came out before the Lisa! I don't know where the whole urban legend about Apple inventing overlapping windows came from, but it's not true.
The trashcan is just a special case of the Xerox Star Office System's document targets. You printed a document on the Star by dragging it to a printer icon. You sent mail by dragging the filled-out letter to a mailbox. The Star, as shipped, didn't have a trashcan, but they had supposedly considered and rejected the idea.
Bill Atkinson left the two PARC visits with a mistaken understanding of the capabilities of the Smalltalk software. While Tesler was giving a demo, Atkinson thought that he had seen two windows layered on top of each other, though they were just bordering each other.
I doubt the reliability of this story... This must have been a very early version of Smalltalk if it didn't have overlapping windows.
Yeh, I remember the Xerox Star at NCC in 1982, it came out just before the Lisa. The problem with both of them was that they were too expensive for the growing home computer market. Apple responded with the Mac, Xerox responded with a really good CP/M-80-based computer.
Is it actually "spatial". It looks like the Apple tree view, but the appearance isn't why the Classic Finder is described as "spatial". The Finder has the annoying property (annoying to me anyway, the "spatial Finder" fans love it) that the location on screen of any Finder window is fixed, and there is a single view of any object in the file system. In the classic Finder, if you open an icon-view window of an expanded directory in the tree-view, it actually collapses, and if you expand it again, the icon-view closes. Every time you open a folder, it shows up in the same place, and even if you use cmd or opt (I don't recall which) to force it to keep the same window when you open a subdirectory it closes and re-opens in the "right place".
Is all that actually happening here? If not, they shouldn't call it "spatial".
Re:Why FreeBSD is not good for most businesses
on
Why FreeBSD
·
· Score: 1
I haven't EVER seen a production FreeBSD box that uses core.
Every production FreeBSD box you've ever seen has used core. You can't run without the core.
At my workplace the first thing the BSD folk do post install is get what they need from ports to achieve their tasks.
I'm not sure what your point is. This is a good thing.
On the one hand, I have a Red Hat system. It's a pile of packages. If I want to find out what's been added to the install, I need to start by figuring out what was installed originally, what's been added, and what came in later as a result of upgrades but isn't part of the local software. I spent six weeks earlier this year coming up with a list of RPMs, custom RPMs, and an install script to put it all together... just to install one Java application. Installing the same application on FreeBSD took half a day, and that included installing FreeBSD... and when I was finished "pkg_info -aI" gave me a complete list of just the components that were installed.
I've got two colo systems I maintain, one's FreeBSD 4.8, one's FreeBSD 5.something. When I set up the FreeBSD 5 box, I took the 10 lines I got from pkg_info on the 4.8 box, installed those ports, and copied over the config files... and I was done. Doing the same thing with Red Hat, well, I've had my dose of Red Hat this year.
And you didn't answer either of my questions.
Sure I did. Why is the Linux kernel comparable to the FreeBSD core? Because anything more than the kernel has "Separate source trees, separate build tools, separate configuration tools, separate development teams." Perhaps I should respond to the argument? I've been doing that all along, explaining the advantages of having a complete working operating system at the core, rather than just a kernel.
If anyone's entitled to run off in a huff because of "non-responsive" disputant, it's me.
If you know you'll get HD-DVD for the same price if you just wait, and you know that there will be games that need HD-DVD, seems like the smart thing to do will be to wait a while.
Microsoft needs to come up with some kind of coupon or rebate deal or otherwise give the early adopters a way to upgrade.
You're a lucky guy, you actually had a customer report the problem, instead of flaming you about it on some web board you never heard of. That's easily worth a few hundred bucks.
(personally, I hate GUI compilers, I much prefer to work with a Makefile and "vi"... REAL "vi", too, not this horrid "vim" thing Apple started shipping with Panther... luckily "nvi" builds easily on OSX)
Atlantis is certainly a great variant of games like Pengapop, but surely most of the time in development was spent on those cool steampunk graphics?
Re:Have you even looked at Apple's website?
on
Why FreeBSD
·
· Score: 1
No I'm not confusing the two. Accelerated graphics drivers have been buried so deeply into OSes nowadays that they are extremely hard to separate them from all internal APIs.
That is not in fact the case. Both NeXTSTeP and X11 have a very carefully defined formal interface between applications, and the graphics drivers. Cocoa takes this even further, even in the very first verions of OS X it cleanly separated rendering and compositing, and limited acceleration to what could be provided by OpenGL.
If your driver allocates memory
That's a standard Mach call. It doesn't require access to the kernel.
Or accesses the PCI bus
That's memory-mapped.
And Mach is *too* a microkernel.
L4 is a Microkernel. QNX is a microkernel. The Amiga Exec was a microkernel. Mach is far too large and complex to be considered a microkernel. It implements far too many capabilities that microkernels normally delegate to servers to qualify.
[HFS+ is] one of their more significant contributions.
If you were familiar with HFS+ you wouldn't say that.
There's a lot more stuff that's a lot more interesting than HFS+, but I guess it's going to be as hard for you to find as HFS+.
Such as the class of problems falling between wristwatches and the world's top supercomputer?
The Linux Watch isn't exactly practical, and a wristwatch isn't a particularly hard or tight real-time or embedded problem these days... and systems like Blue Gene are really only useful for "embarassingly parallel" problems. Yeh, that's a pretty narrow band.
NT became successful after they dropped their devotion to a pure microkernel design.
NT was never a pure microkernel design. Like Mach, it's always implemented an awful lot of stuff explicitly rather than through services. I assume what you're talking about was them moving GDI into the kernel in NT4.
1. Whether a server runs in the same address space or a separate address space is an implementation detail: having the X server running in user space on UNIX doesn't make it a microkernel design, and there's nothing in the microkernel model that requires you to cross a protection boundary on a message send. Many real time microkernels don't run on protected-mode systems, and others make the space a server runs in entirely arbitrary.
2. Moving GDI into the kernel came at a terrible cost in reliability and stability for servers. We kept most of our servers on NT 3.51 and jumped straight to 2000.
3. What "made NT a success" wasn't anything to do with the kernel design, it was the shims they added that let them run Windows 9x device drives in 2000 and (even more) in XP, which let them drop the 9x line. They could really have done it with Windows 2000... Windows Me was a mistake on all fronts.
4. Loadable kernel modules have nothing to do with microkernel design.
Forget what academia and the popular press have to say about microkernels. Ask anyone in the real-time controls industry, where we've been using what came to be called "microkernel design" since the '70s, they'll set you straight.
That assumes that the Windows kernel (not other ring 0 services and subsystems, like Win32K) is a problem for Windows.
Can you elaborate a bit on what you're getting at here, because I don't think you're actually disagreeing with me.
When I said it would be stupid to replace the NT Kernel with Linux, that's because I don't think there's anything fundamentally wrong with the NT kernel, at least not for a desktop OS.
If it was based on money, Microsoft would already have followed IBM's model and diversified, or even voluntarily split up into separate companies under an umbrella group, rather than crippling almost everything they've produced to make it part of the support structure for Windows.
Instead, they remind me more of DEC in the VAX era, "all the wood behind one arrow". We've seen how well the strategy of organizing a company around one single product line works out.
Given the way MS operates, it doesn't make sense. Now to provide a better Unix on Windows environment (like better POSIX compliance, a version of BASH, etc) in the form of a good Services for Unix so that applications that are cross-platform can be run easily on Windows, that could help them (making it easy to run Unix/Linux/BSD programs on Windows opening up tons of applications and such). Out of the two, that would be FAR more likely.
It's so likely that it's actually happened.
Porting software to Interix is mostly just a matter of typing "./configure --options...; make; make install". It's got a kind of funky directory structure, with "C:" mapped to "/dev/fs/c" and so on, but it's far from the weirdest UNIX environment I've had to deal with.
They could do the same thing that Apple did, and run their Win32 subsystem on top of the Linux kernel, and not release any of that subsystem's source code.
It would be stupid, but they could do it.
Re:Why FreeBSD is not good for most businesses
on
Why FreeBSD
·
· Score: 1
GNU doesn't generally use that source tree.
I didn't say any of them included any of the source.
But Linux, more than most, faithfully emulates it. The Linux kernel is actually closer in design to the old Version 6 kernel in the Lyons book than Darwin/OSX is.
The kernel is maintained along with the userspace tools quite effectively
Separate source trees, separate build tools, separate configuration tools, separate development teams.
Anything that ain't the kernel or libraries is an applications.
The kernel, libraries, utilities, they aren't separate packages. Or ports. They're a single coherent source tree, like the Linux kernel is.
Last time I checked, there were apps that came in BSD, apps that were installable y 'packages', and apps that installable by ports.
In FreeBSD, "packages" and "ports" are parallel views of the same thing. A "package" is just the result of applying "make package" to a port. You can go either way... I prefer to use the port because it's more flexible, and some ports aren't packaged by default because of issues like copyright reasons (like with Sun Java, or qmail).
But thinking of "ports" and "packages" as separate things, or confusing them with the FreeBSD core, is just a mistake.
The cautionary tale is that NO ONE in Open Source can afford to allow splintering to destabilize the resilience and independence of Linux, Open Source/FLOSS and the movements behind them.
The Open Source movement doesn't even need Linux.
The biggest FOSS win of the past decade, Apple going to an OS operating system, didn't even involve Linux.
Re:In some ways it's inferior to IE5 and IE6...
on
IE7 Bugs and Reviews
·
· Score: 1
I'm just saying that after use for years in two alternative browsers, you're the first person I've seen ever complain about that specific issue with regard to the combined button.
You haven't neem paying much attention to browser design, then. This misfeature was actually introduced in one of the early Netscape 4 betas, and there was so much opposition it was backed out. This is not a new or unusual issue, it's a design that has the potential of causing problems for enough people, and that has so few benefits (frankly, I can't see any), there shouldn't be any question of at the very least making it an option.
The thought of internet ads becoming a battle ground like tv ads during the 2000/2004 US presidential campaign is a nightmare.
I can just see the senate debate in 2007 or 2008 over the use of popup-blockers and ad-killing proxies.
Re:In some ways it's inferior to IE5 and IE6...
on
IE7 Bugs and Reviews
·
· Score: 1
but how many of those pages you're hitting stop on just finished loading in the split second you're clicking on the stop button?
It's hardly a split second. The human reflex loop is about 1/5th of a second at best. You can train yourself to do better, but normally when you're really paying attention to a task at takes about 0.2s before you can react to a change and press a button. If you're already pressing a button, and have to avoid pressing it, the inhibitory reflex can take even longer to work.
And that's if you're expecting it to change on you.
So that "split second" can easily be half a second or more if you're distracted.
And it only has to happen once in a while to be annoying, and it only has to happen on the wrong page, once, to be a right pain in the neck. And... this is Safari on a Macintosh we're talking about now. It's supposed to be designed to be best human interface standards, not "hey, it'd be cool if we could save half a centimeter of toolbar space by putting both buttons in the same spot".
Even if it only causes problems for impatient people on slow machines visiting slashdotted websites, that should be enough for Apple to re-think the damn thing.
Oh, who am I kidding, after Metal Finder, why should I expect Apple to be any better than Microsoft when it comes to human interface design?
I leave an unsecured AP with throttled bandwidth open for general usage in downtown Boston. Obviously that's in front of a firewall behind which my network lives.
What's your SSID? "linksys" or "free wifi"?
Children who pick blackberries and swim in a creek on a many-acre rural farm are not commiting any ethical violation.
Children get a lot of slack in many communities. When I was a kid we used to run all over the neighborhood, didn't pay much attention to fences, if we met the property owner they might chase us off, or they might talk to us. There were several old retirees who we were on regular speaking terms with.
But I don't think they'd be as happy about college students or middle aged geeks pottering around their yards. They might be birdwatchers or they might be casing the joint.
When I was crewing for hot-air balloon pilots, we were always careful to try and contact the landowner before attempting to land a balloon and walk it out, and this WAS out in the country and people generally like HABs. Because you can't assume they know you're chasing, because it's only common courtesy.
In reality, all those linksysers don't care. If they cared, they'd say something. They don't.
If they knew, they'd probably care. If they don't say anything, you can't assume they know.
If someone sets up an access point with an SSID like "free wifi" then it's reasonably to assume they are happy for you to leech off them.
If someone sets up an access point and leaves everything set to the default, then you can't assume they're willing to let you use their access. That's like someone not bothering to lock their backyard gate: if you come in and lounge by the pool... you're trespassing even if you never use anything but the sunshine.
I don't know what the law specifically says about this kind of thing, but it's only common courtesy to ask... or to refrain if you can't find out who's responsible for the AP. If you're not prepared to extend that basic consideration to people, then you're a pretty low kind of worm in my book.
Re:In some ways it's inferior to IE5 and IE6...
on
IE7 Bugs and Reviews
·
· Score: 1
And, really - how often are you hitting stop on a page?
I probably hit "stop" on slashdot several times an hour when I'm actually reading it... either because I jumped to some slashdotted page or because my proxy is having problems resolving some banner from some advertiser's website so it's sitting there waiting on something like "sirius-cybernetics.megadodo.biz".
And that's just slashdot. Cisco's got some really nasty-slow webservers too, and those can be really painful over dialup, plus there's sites with a dozen or more "flash" plugins, though those I usually hit "back" on.
Re:In some ways it's inferior to IE5 and IE6...
on
IE7 Bugs and Reviews
·
· Score: 1
The thing is, the Quicktime play/pause control doesn't break the rules, because it's not a pair of separate functions that alternate on the same control, it's a toggle, like a menu button, like maximize/minimize, like mute/unmute. Toggles are a normal feature of user interfaces.
But toggles only work if you're really turning something on and off when you hit the control. The model is a state change, not a pair of actions. Hitting the toggle again is always safe, it will just switch you reliably between two states... the music/movie is playing/is paused, the menu is displayed/hidden, sound output is enabled/inhibited, the window is maximized/normal. In the real world, the light is on/off, the CD is playing/paused, and so on. The mental model is "on" versus "off".
If the action is not a toggle, you don't use a toggle control. When you get into a lift, even if it's a two story lift, you don't have a button that switches between "1" and "2", you have two buttons.
Stop/reload is not a toggle. It's like accelerator/brake... these are opposite functions, but you would never build a car that had a single "stop/accelerate" pedal that switched state whenever you hit the clutch, even though that's a reliable indicator of whether you're intending to brake or accelerate.
This isn't just a technical difference, it's a functional one. If you're not toggling between complementary states by using the control, then use separate controls.
And if you haven't seen this in Safari, you've had the good luck to never have hit any flakey websites at the wrong time, and you've got a good fast Mac.
Slashdot Mirror
Typing out 1000 different switches and options by hand...
Doesn't happen. You type "make" and it does it all for you. The first program I wrote for VAXC on VMS was an implementation of Make... and this was back before the Macintosh existed.
In the rare case where you need command line options, you put them in a Make variable and it uses them everywhere.
Then for the output, there's this great program called "error" that I first ran into at Berkeley in 1981. You pipe your output through that, and it puts all the errors in the files as comments and can even toss you into the editor right at the first one.
Most of the great features of GUI compiler tools already existed in command line tools back before there really was a general purpose GUI to implement them in.
Or else he really was seeing overlapping windows.
I can't imagine how anyone could have demoed any of the Smalltalk systems without ever overlapping any windows. Smalltalk-72 had them, right from a very early stage, and I honestly can't see how someone giving a demo could have deliberately lined up two windows without overlapping them, and never actually overlapped any windows, unless they were trying very hard to avoid revealing that capability.
Which would be really weird, because the oldest published photos of Smalltalk-72 screens already showed overlapping windows.
Of the three items you listed, only one is actually an Apple innovation... and that one is an innovation we'd be better off without: pull-down menus.
Pull-down menus were a hack to let them have a single-button mouse. Everyone else used contextual menus, and even Apple has in a backhanded way adopted them... and before you go all Fitt's Law on me, don't forget that there are *5* "best targets" on the screen for Fitt's Law, and "right under the mouse" is one of them.
Overlapping windows were NOT an Apple innovation. Smalltalk-72 had overlapping windows Smalltalk-76 had overlapping windows. Smalltalk-80 and Interlisp-D and the Xerox Star office system had overlapping windows. The Star came out before the Lisa! I don't know where the whole urban legend about Apple inventing overlapping windows came from, but it's not true.
The trashcan is just a special case of the Xerox Star Office System's document targets. You printed a document on the Star by dragging it to a printer icon. You sent mail by dragging the filled-out letter to a mailbox. The Star, as shipped, didn't have a trashcan, but they had supposedly considered and rejected the idea.
Bill Atkinson left the two PARC visits with a mistaken understanding of the capabilities of the Smalltalk software. While Tesler was giving a demo, Atkinson thought that he had seen two windows layered on top of each other, though they were just bordering each other.
l ltalk-72-1977.jpgl ltalk-72-simple.jpgp ng
I doubt the reliability of this story... This must have been a very early version of Smalltalk if it didn't have overlapping windows.
http://squab.no-ip.com:8080/collab/uploads/61/sma
http://squab.no-ip.com:8080/collab/uploads/61/sma
http://squab.no-ip.com:8080/collab/uploads/St-76.
Yeh, I remember the Xerox Star at NCC in 1982, it came out just before the Lisa. The problem with both of them was that they were too expensive for the growing home computer market. Apple responded with the Mac, Xerox responded with a really good CP/M-80-based computer.
Is it actually "spatial". It looks like the Apple tree view, but the appearance isn't why the Classic Finder is described as "spatial". The Finder has the annoying property (annoying to me anyway, the "spatial Finder" fans love it) that the location on screen of any Finder window is fixed, and there is a single view of any object in the file system. In the classic Finder, if you open an icon-view window of an expanded directory in the tree-view, it actually collapses, and if you expand it again, the icon-view closes. Every time you open a folder, it shows up in the same place, and even if you use cmd or opt (I don't recall which) to force it to keep the same window when you open a subdirectory it closes and re-opens in the "right place".
Is all that actually happening here? If not, they shouldn't call it "spatial".
I haven't EVER seen a production FreeBSD box that uses core.
Every production FreeBSD box you've ever seen has used core. You can't run without the core.
At my workplace the first thing the BSD folk do post install is get what they need from ports to achieve their tasks.
I'm not sure what your point is. This is a good thing.
On the one hand, I have a Red Hat system. It's a pile of packages. If I want to find out what's been added to the install, I need to start by figuring out what was installed originally, what's been added, and what came in later as a result of upgrades but isn't part of the local software. I spent six weeks earlier this year coming up with a list of RPMs, custom RPMs, and an install script to put it all together... just to install one Java application. Installing the same application on FreeBSD took half a day, and that included installing FreeBSD... and when I was finished "pkg_info -aI" gave me a complete list of just the components that were installed.
I've got two colo systems I maintain, one's FreeBSD 4.8, one's FreeBSD 5.something. When I set up the FreeBSD 5 box, I took the 10 lines I got from pkg_info on the 4.8 box, installed those ports, and copied over the config files... and I was done. Doing the same thing with Red Hat, well, I've had my dose of Red Hat this year.
And you didn't answer either of my questions.
Sure I did. Why is the Linux kernel comparable to the FreeBSD core? Because anything more than the kernel has "Separate source trees, separate build tools, separate configuration tools, separate development teams." Perhaps I should respond to the argument? I've been doing that all along, explaining the advantages of having a complete working operating system at the core, rather than just a kernel.
If anyone's entitled to run off in a huff because of "non-responsive" disputant, it's me.
If you know you'll get HD-DVD for the same price if you just wait, and you know that there will be games that need HD-DVD, seems like the smart thing to do will be to wait a while.
Microsoft needs to come up with some kind of coupon or rebate deal or otherwise give the early adopters a way to upgrade.
You're a lucky guy, you actually had a customer report the problem, instead of flaming you about it on some web board you never heard of. That's easily worth a few hundred bucks.
... REAL "vi", too, not this horrid "vim" thing Apple started shipping with Panther... luckily "nvi" builds easily on OSX)
(personally, I hate GUI compilers, I much prefer to work with a Makefile and "vi"
Atlantis is certainly a great variant of games like Pengapop, but surely most of the time in development was spent on those cool steampunk graphics?
No I'm not confusing the two. Accelerated graphics drivers have been buried so deeply into OSes nowadays that they are extremely hard to separate them from all internal APIs.
That is not in fact the case. Both NeXTSTeP and X11 have a very carefully defined formal interface between applications, and the graphics drivers. Cocoa takes this even further, even in the very first verions of OS X it cleanly separated rendering and compositing, and limited acceleration to what could be provided by OpenGL.
If your driver allocates memory
That's a standard Mach call. It doesn't require access to the kernel.
Or accesses the PCI bus
That's memory-mapped.
And Mach is *too* a microkernel.
L4 is a Microkernel. QNX is a microkernel. The Amiga Exec was a microkernel. Mach is far too large and complex to be considered a microkernel. It implements far too many capabilities that microkernels normally delegate to servers to qualify.
[HFS+ is] one of their more significant contributions.
If you were familiar with HFS+ you wouldn't say that.
There's a lot more stuff that's a lot more interesting than HFS+, but I guess it's going to be as hard for you to find as HFS+.
Such as the class of problems falling between wristwatches and the world's top supercomputer?
The Linux Watch isn't exactly practical, and a wristwatch isn't a particularly hard or tight real-time or embedded problem these days... and systems like Blue Gene are really only useful for "embarassingly parallel" problems. Yeh, that's a pretty narrow band.
NT became successful after they dropped their devotion to a pure microkernel design.
NT was never a pure microkernel design. Like Mach, it's always implemented an awful lot of stuff explicitly rather than through services. I assume what you're talking about was them moving GDI into the kernel in NT4.
1. Whether a server runs in the same address space or a separate address space is an implementation detail: having the X server running in user space on UNIX doesn't make it a microkernel design, and there's nothing in the microkernel model that requires you to cross a protection boundary on a message send. Many real time microkernels don't run on protected-mode systems, and others make the space a server runs in entirely arbitrary.
2. Moving GDI into the kernel came at a terrible cost in reliability and stability for servers. We kept most of our servers on NT 3.51 and jumped straight to 2000.
3. What "made NT a success" wasn't anything to do with the kernel design, it was the shims they added that let them run Windows 9x device drives in 2000 and (even more) in XP, which let them drop the 9x line. They could really have done it with Windows 2000... Windows Me was a mistake on all fronts.
4. Loadable kernel modules have nothing to do with microkernel design.
Forget what academia and the popular press have to say about microkernels. Ask anyone in the real-time controls industry, where we've been using what came to be called "microkernel design" since the '70s, they'll set you straight.
That assumes that the Windows kernel (not other ring 0 services and subsystems, like Win32K) is a problem for Windows.
Can you elaborate a bit on what you're getting at here, because I don't think you're actually disagreeing with me.
When I said it would be stupid to replace the NT Kernel with Linux, that's because I don't think there's anything fundamentally wrong with the NT kernel, at least not for a desktop OS.
It's based on money.
If it was based on money, Microsoft would already have followed IBM's model and diversified, or even voluntarily split up into separate companies under an umbrella group, rather than crippling almost everything they've produced to make it part of the support structure for Windows.
Instead, they remind me more of DEC in the VAX era, "all the wood behind one arrow". We've seen how well the strategy of organizing a company around one single product line works out.
Whether they went BSD or GPL, they wouldn't do it like Apple.
Given the way MS operates, it doesn't make sense. Now to provide a better Unix on Windows environment (like better POSIX compliance, a version of BASH, etc) in the form of a good Services for Unix so that applications that are cross-platform can be run easily on Windows, that could help them (making it easy to run Unix/Linux/BSD programs on Windows opening up tons of applications and such). Out of the two, that would be FAR more likely.
It's so likely that it's actually happened.
Porting software to Interix is mostly just a matter of typing "./configure --options...; make; make install". It's got a kind of funky directory structure, with "C:" mapped to "/dev/fs/c" and so on, but it's far from the weirdest UNIX environment I've had to deal with.
They could do the same thing that Apple did, and run their Win32 subsystem on top of the Linux kernel, and not release any of that subsystem's source code.
It would be stupid, but they could do it.
GNU doesn't generally use that source tree.
I didn't say any of them included any of the source.
But Linux, more than most, faithfully emulates it. The Linux kernel is actually closer in design to the old Version 6 kernel in the Lyons book than Darwin/OSX is.
The kernel is maintained along with the userspace tools quite effectively
Separate source trees, separate build tools, separate configuration tools, separate development teams.
Anything that ain't the kernel or libraries is an applications.
The kernel, libraries, utilities, they aren't separate packages. Or ports. They're a single coherent source tree, like the Linux kernel is.
Last time I checked, there were apps that came in BSD, apps that were installable y 'packages', and apps that installable by ports.
In FreeBSD, "packages" and "ports" are parallel views of the same thing. A "package" is just the result of applying "make package" to a port. You can go either way... I prefer to use the port because it's more flexible, and some ports aren't packaged by default because of issues like copyright reasons (like with Sun Java, or qmail).
But thinking of "ports" and "packages" as separate things, or confusing them with the FreeBSD core, is just a mistake.
The cautionary tale is that NO ONE in Open Source can afford to allow splintering to destabilize the resilience and independence of Linux, Open Source/FLOSS and the movements behind them.
The Open Source movement doesn't even need Linux.
The biggest FOSS win of the past decade, Apple going to an OS operating system, didn't even involve Linux.
I'm just saying that after use for years in two alternative browsers, you're the first person I've seen ever complain about that specific issue with regard to the combined button.
You haven't neem paying much attention to browser design, then. This misfeature was actually introduced in one of the early Netscape 4 betas, and there was so much opposition it was backed out. This is not a new or unusual issue, it's a design that has the potential of causing problems for enough people, and that has so few benefits (frankly, I can't see any), there shouldn't be any question of at the very least making it an option.
The thought of internet ads becoming a battle ground like tv ads during the 2000/2004 US presidential campaign is a nightmare.
I can just see the senate debate in 2007 or 2008 over the use of popup-blockers and ad-killing proxies.
but how many of those pages you're hitting stop on just finished loading in the split second you're clicking on the stop button?
It's hardly a split second. The human reflex loop is about 1/5th of a second at best. You can train yourself to do better, but normally when you're really paying attention to a task at takes about 0.2s before you can react to a change and press a button. If you're already pressing a button, and have to avoid pressing it, the inhibitory reflex can take even longer to work.
And that's if you're expecting it to change on you.
So that "split second" can easily be half a second or more if you're distracted.
And it only has to happen once in a while to be annoying, and it only has to happen on the wrong page, once, to be a right pain in the neck. And... this is Safari on a Macintosh we're talking about now. It's supposed to be designed to be best human interface standards, not "hey, it'd be cool if we could save half a centimeter of toolbar space by putting both buttons in the same spot".
Even if it only causes problems for impatient people on slow machines visiting slashdotted websites, that should be enough for Apple to re-think the damn thing.
Oh, who am I kidding, after Metal Finder, why should I expect Apple to be any better than Microsoft when it comes to human interface design?
I leave an unsecured AP with throttled bandwidth open for general usage in downtown Boston. Obviously that's in front of a firewall behind which my network lives.
What's your SSID? "linksys" or "free wifi"?
Children who pick blackberries and swim in a creek on a many-acre rural farm are not commiting any ethical violation.
Children get a lot of slack in many communities. When I was a kid we used to run all over the neighborhood, didn't pay much attention to fences, if we met the property owner they might chase us off, or they might talk to us. There were several old retirees who we were on regular speaking terms with.
But I don't think they'd be as happy about college students or middle aged geeks pottering around their yards. They might be birdwatchers or they might be casing the joint.
When I was crewing for hot-air balloon pilots, we were always careful to try and contact the landowner before attempting to land a balloon and walk it out, and this WAS out in the country and people generally like HABs. Because you can't assume they know you're chasing, because it's only common courtesy.
In reality, all those linksysers don't care. If they cared, they'd say something. They don't.
If they knew, they'd probably care. If they don't say anything, you can't assume they know.
If someone sets up an access point with an SSID like "free wifi" then it's reasonably to assume they are happy for you to leech off them.
If someone sets up an access point and leaves everything set to the default, then you can't assume they're willing to let you use their access. That's like someone not bothering to lock their backyard gate: if you come in and lounge by the pool... you're trespassing even if you never use anything but the sunshine.
I don't know what the law specifically says about this kind of thing, but it's only common courtesy to ask... or to refrain if you can't find out who's responsible for the AP. If you're not prepared to extend that basic consideration to people, then you're a pretty low kind of worm in my book.
And, really - how often are you hitting stop on a page?
I probably hit "stop" on slashdot several times an hour when I'm actually reading it... either because I jumped to some slashdotted page or because my proxy is having problems resolving some banner from some advertiser's website so it's sitting there waiting on something like "sirius-cybernetics.megadodo.biz".
And that's just slashdot. Cisco's got some really nasty-slow webservers too, and those can be really painful over dialup, plus there's sites with a dozen or more "flash" plugins, though those I usually hit "back" on.
The thing is, the Quicktime play/pause control doesn't break the rules, because it's not a pair of separate functions that alternate on the same control, it's a toggle, like a menu button, like maximize/minimize, like mute/unmute. Toggles are a normal feature of user interfaces.
But toggles only work if you're really turning something on and off when you hit the control. The model is a state change, not a pair of actions. Hitting the toggle again is always safe, it will just switch you reliably between two states... the music/movie is playing/is paused, the menu is displayed/hidden, sound output is enabled/inhibited, the window is maximized/normal. In the real world, the light is on/off, the CD is playing/paused, and so on. The mental model is "on" versus "off".
If the action is not a toggle, you don't use a toggle control. When you get into a lift, even if it's a two story lift, you don't have a button that switches between "1" and "2", you have two buttons.
Stop/reload is not a toggle. It's like accelerator/brake... these are opposite functions, but you would never build a car that had a single "stop/accelerate" pedal that switched state whenever you hit the clutch, even though that's a reliable indicator of whether you're intending to brake or accelerate.
This isn't just a technical difference, it's a functional one. If you're not toggling between complementary states by using the control, then use separate controls.
And if you haven't seen this in Safari, you've had the good luck to never have hit any flakey websites at the wrong time, and you've got a good fast Mac.