Probably both. There's a difference between being fired and being forced to resign. A legal difference.
This kind of thing happens fairly often. Companies avoid possible lawsuits about early termination of contracts or general unfair dismissal by making an unwanted management-level employee an offer they can't refuse (usually something approximating to a golden handshake and help finding a new job.)
Win95 really zipped along in 32 megs until the advent of Windows-integrated IE.
15 windows being opened caused problems for you? I remember my Amiga used to get a little slow if you opened 15 windows on one screen, but that was on a 7MHz 68000. And it was a little slow, not unusably.
There are some things that suck CPU usage that really weren't practical back in the days of single or low double digit CPU speeds - outline fonts would be the big example I can think of. But generally, these weren't slow machines, not in the way you're describing anyway. Take a look at that recent NeXT demo that was linked to from Slashdot, circa 1992 or so, Steve Jobs demonstrating a complete, Unix based, OO-GUI (Objective C no less, a language with a high run-time overhead), on a 68030 running at, at most, 33MHz. Still think people who remember things being no slower were dreaming?
Things have improved. We've gone from bitmap fonts to beautiful anti-aliased outline fonts, and two to eight bit colour to 24/32 bit colour (admittedly, that's more of a memory thing, eight bit colour is often slower than 32 bit because you have to do shading), but it doesn't feel like we've actually improved in tune with the technology. The PC in front of me ought to feel around much faster than the Amiga I used in the early nineties even given the amount of new stuff it's doing (it's physically around 100x faster after all.) It doesn't.
It's interesting how everyone who wants to jump on anyone pointing out the flaws in the "IIS proves that unpopular systems are targets too" argument rarely bother to consider the context, and usually just accuse anyone making the argument of being Microsoft apologists. A quick look at my posting history would tell you otherwise.
The fact is: until those of you who ignore the degree to which platform marketshare affects virus writers actually open your eyes and look at what's happening, you'll continue to avoid attacking the problem. You'll install GNU/Linux and Apache in the belief that by doing so, you're some how securing yourself because, some how, those products are better written.
One bug is all it takes. I don't care if NIMDA exploited six million, the fact is if there's a viable root exploit, it can be exploited. And once the market figures change, you'll be just as vulnerable.
Right now I'm looking most of all not at the web servers, who at least benefit from the fact that the majority of Apache admins tend to be relatively clued up compared to IIS admins (see point 1), but at Apple users. As an OS X user, I'm seriously concerned about the number of people who have bought your argument and really belief OS X isn't going to have just as many attacks once it gains marketshare - as it probably will do now the Mac mini is out. Those people are going to make it harder for everyone. Just as the "experts" who used to claim you couldn't get a virus via email made sure nobody was prepared for these kinds of things, so you who argue that there's something about Microsoft's products that makes them vulnerable that just isn't true about anyone else's are going to leave non-Microsoft users vulnerable.
Stop it. Think about what you're arguing. I don't like Microsoft either, but I'm not going to spend my entire life in some permanent 1984ish hate, arguing advocacy for the sake of it, as you are.
Usenet clients is perhaps a bad example. They have new, bad, features, and have lost many old, good, ones.
I miss "nn". I think the end of being able to use tools like nn because there was so much totally unnecessary MIME/HTML on Usenet marked the point at which I ceased to care about the network.
I'm pretty sure there's a half-way house somewhere. I seem to remember the early nineties were full of fully functional does-everything-I-want applications that had nice GUIs and used a megabyte or two of memory when they ran.
I wouldn't categorize MacWrite or early Mac OS as that, but, I don't know, Lotus AmiPro and Microsoft Word for Windows 2 were pretty good, to give two examples.
First, let me say I think the GP hit the nail on the head.
Second, I project real life. Virus writers are, given the MASSIVE market share of Windows, likely to be Windows users themselves.
And generally, most people I know, regardless of whether they're geeks or not, are not especially fond of Microsoft. Most are neutral, many dislike them. They dislike Microsoft because they use Windows.
And geeks, in my experience - which naturally is going to include most virus writers - are especially negative towards Microsoft.
No, I don't know any virus writers personally, but I'd have to be blind not to notice the way the world is.
IIS has orders of magnitude more worms and viruses than Apache, not security holes. (If you know different, please report all these security holes to the respective developers.)
In any case, that's not why IIS is a target. IIS is a target for a number of reasons:
1. The default web server (once you installed the Option pack) on NT4 was a version of IIS that had a couple of security holes.
2. IIS is almost certainly more popular than Apache on Windows-type servers. Yes, Apache runs on it too, but there's a mentality that goes with Microsoft products. With Windows being the most "popular" (well, most installed) operating system in existance, you'd expect it to be the home for most virus writers, and virus writers aren't going to want to write for platforms they consider obscure.
3. IIS always runs on Windows which almost always has the same CPU architecture (ie ix86) and APIs. Apache can run on a variety of platforms. Exploiting, for example, buffer overflows, is relatively difficult if you don't know the architecture, or even something as simple as how you can make system calls.
4. People hate Microsoft. Most people's views of Apache and similar systems are neutral at worst. I've never met or talked to anyone who hates Apache.
5. One serious hole is enough. Both Apache and IIS have suffered these in the past and probably will do in future (well, until Gosling rewrites Apache in Java;-)
IIS is usually used as the proof to counter the argument that marketshare matters when viruses are written. I think this is always a bogus argument: IIS's marketshare is enough (it's not like writing a virus for GNU/Linux), and IIS is tied to an operating system that is the most installed in the world. It's fair to say it's the exception that, quite literally, proves the rule.
Just tried it on Safari, and unfortunately none of the "bugs" that occur in Firefox that give the game away are present. Mouseover looks like Paypal. Connecting message shows nothing untoward. Renders as "www.paypal.com" in the regular font and location on the URL bar.
I wonder if there's a quick and easy fix for this for Safari users, like there is for Firefox (about:config, network.enableIDN -> false.)
The pre-linking that occurs while OS X is "optimizing the system" is to do with regular C symbols, not dynamic Objective C objects. Objective C, being dynamic, really doesn't need "pre-linking".
This is why the "Optimization" occurs regardless of the app being installed.
That's not true. NextStep used Display Postscript. DPS was removed from the operating system at around the Rhapsody stage of the OS and replaced with Quartz, which is a more traditional compositing engine. Nothing like Quartz existed in NextStep, and nothing like Display Postscript exists in OS X.
Quartz is modelled on, but not the same as, PDF (it isn't "Display PDF and I've been corrected by Apple employees when saying it is.)
I used to say that too and then was corrected by an actual Apple engineer. As I understand it, while the designers of Quartz used PDF as its model, it's not "Display PDF" in the same way as "Display Postscript" is related to PostScript. You don't write PDF, you make relatively mainstream looking API calls in the same way as you would Win16, Quickdraw, X11, or graphics.library.
Also, are you sure you have to pay royalties to Adobe to use Display Postscript? How does GNUStep work around that, then?
GNUStep doesn't use Display Postscript. (In fact, GNUStep doesn't use any DPS type system to the same degree of intensity that NextStep does, when NextStep used DPS, GNUStep, for the most part, uses an abstraction that ultimately goes straight to X11 bypassing DPS type stuff completely.) What GNUStep includes is Display Ghostscript, a complete rewrite that's compatable but doesn't contain Adobe code.
That's what Apple had and didn't want to use. Their options were: 1. Continue to license DPS from Adobe. 2. Write their own clone of DPS (or license such a thing.) or 3. Write something they considered more efficient and appropriate that could be done in less time. They chose 3.
As for the Dock, it seems to me that they just glued the two pieces together, moved it around, and made it transparent. Other than eye-candy (the icon scaling and whatnot), what additional functions does it perform?
They're pretty dissimilar beyond the "I can launch tasks and switch to launched tasks" thing of the original. You can minimise windows, navigate entire directory trees, and attach documents, to this dock, none of which could be done with the NeXT version. The NeXT version only showed certain apps, not all running apps (they appeared as icons on the bottom of the screen instead of in the dock, to dock them - indeed - you'd drag them to the dock.) You can get a good feel of what the old NeXT version did by running WindowMaker.
The file-viewing part of the Finder really is different from the NeXTStep file manager (thank goodness!), but the window manager part is still similar. They took out the virtual desktops, added a "close" button to the title bar, moved the application menu from the right mouse button to a bar at the top of the screen, and made pretty graphics.
I don't recall NeXT having virtual desktops. Close buttons were always on the title bars (the new buttons are the "change size" button (yellow) and the pill on the right that hides and shows toolbars, amongst other things.) Window management is similar (click on an app icon - of course, they're all in the dock now - and all of the windows come to the front, and you can hide apps, but beyond that everything NeXT did was being done by everyone anyway.)
But that doesn't mean they threw out big chunks of the OS!
I think Display Postscript is the biggie. I do think that people who talk about this tend to exaggerate on both sides - there are those who are convinced that OS X has little in common with NeXT, and those who think it's just a themed NeXTStep with a bit added. It's neither, of course. The UI, for all intents and purposes, is entirely different. But behind it, NeXT programmers find it familiar, the command line is familiar, and there's a Services menu...
As a Kerry supporter, I don't see a problem with what you're saying. Why should it be an issue that if you search for Kerry, you get Bush ads?
It's free speech. If the ads were obtrusive and prevented you from getting the information you want, that'd be one thing, but that's not the case with Google.
Anyway, I still don't see the issue with respect to Google. You can't buy "all the advertising space" related to your rival with Google, all you can do is have it so your advertising comes up if you search for that rival.
From the point of view of an end user, this is far from evil or misleading. This isn't about people trying to pass themselves off as a company, it's simply marketing to people interested in a particular product. If I search for "Biro", the chances are I want a pen. If I'm searching for something specific to Biro (like "How do I contact them about repairing this pen"), then the relevent links will appear on the left, as always. If I'm looking to buy a biro, then I'll be interested in the unobtrusive links on the right from companies that feel I may be interested in their product. Maybe I would prefer a Parker. Maybe I want to buy the Biro from WHSmiths.
I think the fairest comparison is actually Parker Pens going to a newspaper and saying "When you next do a story concerning Biros, can you put our ad next to it?"
That shouldn't be considered trademark infringment, and it's certainly not evil business practices. Trademarks were intended as a way to guarantee against fraud, as one group passing its products off as the other's. It strikes me that the laws here are being taken to go far beyond that remit, to the point that it threatens speech instead. If they carry on this way, just as the DMCA is undermining the legitimacy of copyright, and software patents the legitimacy of the patent system, they risk undermining trademarks too. Legislatures need to intervene and stop this abuse.
I guess there's Display Postscript. Apple didn't want to pay Adobe royalties so they replaced the technology with Quartz. There's also a lot of stuff that was replaced in a more strict sense (eg old, not Mac-like, stuff replaced with new stuff that was more Mac like and a little nicer, rather than the removal of a technology) - the Workspace Manager (the NeXT equivalent of the Finder or Microsoft's Explorer) was replaced with something resembling the Finder (thank goodness, NeXT people may disagree but I've always thought it was abysmal), for example. The old dock was replaced with something that performs additional functions and has a different kind of look to it.
I think rewriting OS X to run over L4 would be possible but not as trivial as some people are arguing. You could probably modify Linux, if you really wanted, swapping out XNU and the Mach-tied device drivers and putting in the modified Linux and user-space module management tools to support it. However, would you want to?
OS X's kernel does exactly what Apple wants it to do. Unless there's a compelling reason to switch, Apple would be undermining much of the development and testing work its done so far for what'd amount to a largely academic exercise.
And don't encourage them too much: As Mac OS X is a proprietary operating system, they might even have better choices available than L4. For example, if they could consider replacing Mach with L4, why not replace Darwin with QNX?
The original poster's dates were off. Apple started looking at next generation operating systems in 1989, Copland started in 1994. So the comment "it was looking at next-generation operating systems before Copland development actually started." is true but the first half of your sentence is thus undermined by the GP's dates.
What Apple started in 1989 was Taligent, which was released a few years later after being spun off.
So we have NextStep, started in 1987 or so, which was successfully released within a couple of years. Taligent, started in 1989, which shipped within about four years of development. Copland, which was badly managed and arguably not the right design, which was cancelled after two years.
And then OS X, the first version of which was called Rhapsody and released to developers within a year or so of the NeXT/Apple merger. It then took several releases (Rhapsody DR2, Mac OS Server, Mac OS X Preview) before Apple felt they had something they could market as a consumer level operating system. But before that, it certainly was a full blown powerful usable operating system, it just was too un-Maclike and unrevolutionary to fit into their marketing strategy.
I think, in short, the GP was off-base. I agree with you the first versions of the consumer level operating system "OS X" weren't terribly good, but that's a whole different discussion. NEXTSTEP through Mac OS Server were, by all accounts, very nice and very usable. It was just Aqua changed everything.
Copland development started in 1994. The project was initiated because Apple went into a panic about Windows 95, whose release was imminent. I suspect the 1999 year you quote is confusion with Taligent. Taligent was spun off and released three or so years after it was initiated and it did ship, but it bombed. Had Microsoft not been developing a pre-emptive multitasking 32 bit version of DOS/Windows and hadn't been developing NT, it's unlikely Copland would have happened.
Mac OS X is not Copland. Copland was cancelled after a couple of years of development, and NeXT bought to replace it. OS X is, essentially, the latest version of NEXTSTEP.
"Aha!", I pretend to hear you cry, "But it took until 2003 before there was a viable version of OS X!". Well, that's not true either. Within a year or so of the takeover, Apple released, internally and to developers, Rhapsody, the first NEXTSTEP-MacOS combination operating system. This was, by all accounts, excellent, and the second release of it had most of the functionality we associate with OS X today. However, while this was a complete operating system, Apple decided there were two problems with it: it didn't feel right, they wanted the operating system to feel more revolutionary, and developers weren't happy, they felt they needed APIs compatable with the older platform to help them port their application.
The result is this working, fully functional, many people would have killed to get it, operating system was kept off consumer machines, essentially, until Apple felt they had something more revolutionary. That's not to say you couldn't buy it, sometime after Rhapsody was released, the first "commercial" release happened in the form of "Mac OS Server." I can't remember, for the life of me, the year that came out, but I believe it was prior to 2000.
In short, when Apple had a right design, they did have working products in a reasonable space of time, they just felt they didn't have commercially successful products, and avoided selling them (hard) for that reason. But they've been releasing or selling OPENSTEP based operating systems, in some shape or form, for much longer than most people realise.
The project from 1994, Copland, failed because, well, it probably wasn't the right design, and it certainly wasn't managed properly. Everyone makes mistakes.
I don't this is really comparable to the HURD. The HURD has suffered delay after delay, but much of this is because of Linux. The fact Linux exists means a free operating system based on GNU running over Linux, can exist, and the pressure is off of HURD. HURD, therefore, can be a test bed, with the aim towards making it a perfect kernel.
I honestly believe that if it wasn't for Linux, we'd probably have had HURD 1.0 appear over a decade ago.
That's something we can safely rule out. Microsoft is a company that has only ever produced software, and while they may have flirted with Linux based systems in the same way Corel and Caldera did, ultimately they'd have given up. Again, in the same way Corel and Caldera did.
The big companies that have to do with GNU/Linux and remain supportive of it are either start-ups, consultancies, or hardware manufacturers, or a bit of everything. Microsoft is not a consultancy, not a hardware manufacturer, and is established.
There are, of course, two other issues with Microsoft becoming Linux's saviour in this parallel universe. The first is that Microsoft is owned by Bill Gates, a pioneer of proprietary software who believes, strongly, in using copyright to restrict redistribution. The other is that Microsoft would probably be weaker than Novell is today if IBM did this. They'd be little more than an application vendor tied to OS/2, with no way of changing their business model without rebuilding their entire business.
Of course, if you live in this parallel universe, you have to ask if OS/2 Warp would have been as decent as it was, had Windows 3.0 not happened.
When OS/2 2.0 came out, IBM had made impressive steps at improving the operating system to a point that people could actually like it (1.x, even with Presentation Manager, was widely considered to be awful.) Even then though, it was released after the IBM/Microsoft split, and much of it appeared to be an attempt to play catch-up with Windows. By 3.0 (Warp), IBM was marketing it as the solution to all Windows problems.
Unquestionably, it was a better operating system by that point, but it was several years of development in competition with Windows that lead to that point. Back in 1990, it sucked. And, with no competitor, I wonder if it'd have improved as fast as it did. OS/2 2.0 might have continued to be an 80286 compatable operating system, development would have remained slower than intended, with warring and not exactly harmonous groups developing different parts working for the bickering "allies" (IBM and Microsoft) that made the OS/2 alliance.
I suspect the HURD would have been out by 1995 had Linux not arrived. I suspect Linux basically changed things, making a kernel available, and so allowing users to "rip MP3s, encode video, write documents", etc, and thus taking the pressure off the HURD people who could then focus on making something they considered to be truly great.
I don't know if that's a good thing or not. I guess we'll know when the first viable HURDs start to appear. But, given the major change in direction from the project leaders (dumping Mach, switching to L4), I guess we can see already that HURD wasn't "perfect", and would have been far from it had it been released in the first half of the 90's.
Several companies have released code under the GPL after being threatened with a lawsuit. But no Judge has ever stood up and said "Abide by the license! Release the source, or go to jail!" - to the best of my knowledge, no lawsuit has ever gotten that far before the parties choose to settle.
While the GPL is "unmodifiable", there's nothing stopping you from adding additional permissions to a GPL'd project whose copyright belongs to you and other consenting individuals.
For example, Linus explicitly allows non-GPL'd software to run over Linux, though an addition to the LICENSE file. In this case, Torvalds wasn't modifying the GPL, he was essentially adding an additional license.
This is allowed because a license (as opposed to an EULA) is just a set of permissions. Each set of permissions adds to any you already have (including your default set of "fair use" privileges.) You can license any project you own under as many licenses you wish, and end users can pick and choose which (complete) licenses they want to agree to. (The word "complete" in that sentence is important.)
Also, while the GPL is unmodifiable for existing projects that do not belong to you, if you have a strong enough case you can persuade the FSF to agree to a modified version for projects you own, on occasion even if the result is a license incompatable with the GPL. For example, the Affero General Public License, whose history you can read about here.
From my personal point of view, *this* is the most valuable thing OSI does. If I come across some oddball license, I don't need to sort through the legalese, I can just check if it's OSI approved.
I generally agree except in that their efforts are redundant, the FSF has been maintaining a list of licenses for a while too. I've also always felt that while the FSF's process is less formal (and could do with being more so, and more transparent), it ends up with a list that's usually more useful. The FSF always describes the licenses, giving information on whether it's a copyleft, the stength if it has one, its compatability with the GPL (the original neutral copyleft license), whether there are any subtle bombs hidden in the license you might need to know about before using it, etc.
If OSI helped get them to embrace open source, we all owe OSI quite a bit. If OSI, throughout the entire history of it's existance convinces one company on the scale of IBM to make a comment to OS as large as they have, then there efforts will have been justified.
That's the problem though isn't it? We really don't know. It's far more likely that companies like IBM have been persuaded by the success of GNU/Linux and through their own use of FOSS projects like Apache than it is that OSI lobbying helped to any degree. The major campaigning I've seen has been to get Sun to open source Java, and (a) they haven't been terribly successful at that and (b) if Sun does, in the end, do so, there's unlikely to be evidence it'll be because of the efforts of the OSI.
Now, to some extent, that doesn't invalidate your point. Whether we know about it or not, if it does persuade companies to do embrace Free Software, then that's a good thing. It's just hard at the moment to believe that's what's happening.
Slashdot Mirror
This kind of thing happens fairly often. Companies avoid possible lawsuits about early termination of contracts or general unfair dismissal by making an unwanted management-level employee an offer they can't refuse (usually something approximating to a golden handshake and help finding a new job.)
And how quickly would all the people who'd installed the previous version upgrade to the latest (especially if it involves an eight meg download...)?
Win95 really zipped along in 32 megs until the advent of Windows-integrated IE.
15 windows being opened caused problems for you? I remember my Amiga used to get a little slow if you opened 15 windows on one screen, but that was on a 7MHz 68000. And it was a little slow, not unusably.
There are some things that suck CPU usage that really weren't practical back in the days of single or low double digit CPU speeds - outline fonts would be the big example I can think of. But generally, these weren't slow machines, not in the way you're describing anyway. Take a look at that recent NeXT demo that was linked to from Slashdot, circa 1992 or so, Steve Jobs demonstrating a complete, Unix based, OO-GUI (Objective C no less, a language with a high run-time overhead), on a 68030 running at, at most, 33MHz. Still think people who remember things being no slower were dreaming?
Things have improved. We've gone from bitmap fonts to beautiful anti-aliased outline fonts, and two to eight bit colour to 24/32 bit colour (admittedly, that's more of a memory thing, eight bit colour is often slower than 32 bit because you have to do shading), but it doesn't feel like we've actually improved in tune with the technology. The PC in front of me ought to feel around much faster than the Amiga I used in the early nineties even given the amount of new stuff it's doing (it's physically around 100x faster after all.) It doesn't.
The fact is: until those of you who ignore the degree to which platform marketshare affects virus writers actually open your eyes and look at what's happening, you'll continue to avoid attacking the problem. You'll install GNU/Linux and Apache in the belief that by doing so, you're some how securing yourself because, some how, those products are better written.
One bug is all it takes. I don't care if NIMDA exploited six million, the fact is if there's a viable root exploit, it can be exploited. And once the market figures change, you'll be just as vulnerable.
Right now I'm looking most of all not at the web servers, who at least benefit from the fact that the majority of Apache admins tend to be relatively clued up compared to IIS admins (see point 1), but at Apple users. As an OS X user, I'm seriously concerned about the number of people who have bought your argument and really belief OS X isn't going to have just as many attacks once it gains marketshare - as it probably will do now the Mac mini is out. Those people are going to make it harder for everyone. Just as the "experts" who used to claim you couldn't get a virus via email made sure nobody was prepared for these kinds of things, so you who argue that there's something about Microsoft's products that makes them vulnerable that just isn't true about anyone else's are going to leave non-Microsoft users vulnerable.
Stop it. Think about what you're arguing. I don't like Microsoft either, but I'm not going to spend my entire life in some permanent 1984ish hate, arguing advocacy for the sake of it, as you are.
I miss "nn". I think the end of being able to use tools like nn because there was so much totally unnecessary MIME/HTML on Usenet marked the point at which I ceased to care about the network.
I'm pretty sure there's a half-way house somewhere. I seem to remember the early nineties were full of fully functional does-everything-I-want applications that had nice GUIs and used a megabyte or two of memory when they ran.
I wouldn't categorize MacWrite or early Mac OS as that, but, I don't know, Lotus AmiPro and Microsoft Word for Windows 2 were pretty good, to give two examples.
Second, I project real life. Virus writers are, given the MASSIVE market share of Windows, likely to be Windows users themselves.
And generally, most people I know, regardless of whether they're geeks or not, are not especially fond of Microsoft. Most are neutral, many dislike them. They dislike Microsoft because they use Windows.
And geeks, in my experience - which naturally is going to include most virus writers - are especially negative towards Microsoft.
No, I don't know any virus writers personally, but I'd have to be blind not to notice the way the world is.
In any case, that's not why IIS is a target. IIS is a target for a number of reasons:
1. The default web server (once you installed the Option pack) on NT4 was a version of IIS that had a couple of security holes.
2. IIS is almost certainly more popular than Apache on Windows-type servers. Yes, Apache runs on it too, but there's a mentality that goes with Microsoft products. With Windows being the most "popular" (well, most installed) operating system in existance, you'd expect it to be the home for most virus writers, and virus writers aren't going to want to write for platforms they consider obscure.
3. IIS always runs on Windows which almost always has the same CPU architecture (ie ix86) and APIs. Apache can run on a variety of platforms. Exploiting, for example, buffer overflows, is relatively difficult if you don't know the architecture, or even something as simple as how you can make system calls.
4. People hate Microsoft. Most people's views of Apache and similar systems are neutral at worst. I've never met or talked to anyone who hates Apache.
5. One serious hole is enough. Both Apache and IIS have suffered these in the past and probably will do in future (well, until Gosling rewrites Apache in Java ;-)
IIS is usually used as the proof to counter the argument that marketshare matters when viruses are written. I think this is always a bogus argument: IIS's marketshare is enough (it's not like writing a virus for GNU/Linux), and IIS is tied to an operating system that is the most installed in the world. It's fair to say it's the exception that, quite literally, proves the rule.
I wonder if there's a quick and easy fix for this for Safari users, like there is for Firefox (about:config, network.enableIDN -> false.)
This is why the "Optimization" occurs regardless of the app being installed.
Quartz is modelled on, but not the same as, PDF (it isn't "Display PDF and I've been corrected by Apple employees when saying it is.)
That's what Apple had and didn't want to use. Their options were: 1. Continue to license DPS from Adobe. 2. Write their own clone of DPS (or license such a thing.) or 3. Write something they considered more efficient and appropriate that could be done in less time. They chose 3.
They're pretty dissimilar beyond the "I can launch tasks and switch to launched tasks" thing of the original. You can minimise windows, navigate entire directory trees, and attach documents, to this dock, none of which could be done with the NeXT version. The NeXT version only showed certain apps, not all running apps (they appeared as icons on the bottom of the screen instead of in the dock, to dock them - indeed - you'd drag them to the dock.) You can get a good feel of what the old NeXT version did by running WindowMaker. I don't recall NeXT having virtual desktops. Close buttons were always on the title bars (the new buttons are the "change size" button (yellow) and the pill on the right that hides and shows toolbars, amongst other things.) Window management is similar (click on an app icon - of course, they're all in the dock now - and all of the windows come to the front, and you can hide apps, but beyond that everything NeXT did was being done by everyone anyway.) I think Display Postscript is the biggie. I do think that people who talk about this tend to exaggerate on both sides - there are those who are convinced that OS X has little in common with NeXT, and those who think it's just a themed NeXTStep with a bit added. It's neither, of course. The UI, for all intents and purposes, is entirely different. But behind it, NeXT programmers find it familiar, the command line is familiar, and there's a Services menu...It's free speech. If the ads were obtrusive and prevented you from getting the information you want, that'd be one thing, but that's not the case with Google.
Anyway, I still don't see the issue with respect to Google. You can't buy "all the advertising space" related to your rival with Google, all you can do is have it so your advertising comes up if you search for that rival.
From the point of view of an end user, this is far from evil or misleading. This isn't about people trying to pass themselves off as a company, it's simply marketing to people interested in a particular product. If I search for "Biro", the chances are I want a pen. If I'm searching for something specific to Biro (like "How do I contact them about repairing this pen"), then the relevent links will appear on the left, as always. If I'm looking to buy a biro, then I'll be interested in the unobtrusive links on the right from companies that feel I may be interested in their product. Maybe I would prefer a Parker. Maybe I want to buy the Biro from WHSmiths.
I think the fairest comparison is actually Parker Pens going to a newspaper and saying "When you next do a story concerning Biros, can you put our ad next to it?"
That shouldn't be considered trademark infringment, and it's certainly not evil business practices. Trademarks were intended as a way to guarantee against fraud, as one group passing its products off as the other's. It strikes me that the laws here are being taken to go far beyond that remit, to the point that it threatens speech instead. If they carry on this way, just as the DMCA is undermining the legitimacy of copyright, and software patents the legitimacy of the patent system, they risk undermining trademarks too. Legislatures need to intervene and stop this abuse.
I think rewriting OS X to run over L4 would be possible but not as trivial as some people are arguing. You could probably modify Linux, if you really wanted, swapping out XNU and the Mach-tied device drivers and putting in the modified Linux and user-space module management tools to support it. However, would you want to?
OS X's kernel does exactly what Apple wants it to do. Unless there's a compelling reason to switch, Apple would be undermining much of the development and testing work its done so far for what'd amount to a largely academic exercise.
And don't encourage them too much: As Mac OS X is a proprietary operating system, they might even have better choices available than L4. For example, if they could consider replacing Mach with L4, why not replace Darwin with QNX?
What Apple started in 1989 was Taligent, which was released a few years later after being spun off.
So we have NextStep, started in 1987 or so, which was successfully released within a couple of years. Taligent, started in 1989, which shipped within about four years of development. Copland, which was badly managed and arguably not the right design, which was cancelled after two years.
And then OS X, the first version of which was called Rhapsody and released to developers within a year or so of the NeXT/Apple merger. It then took several releases (Rhapsody DR2, Mac OS Server, Mac OS X Preview) before Apple felt they had something they could market as a consumer level operating system. But before that, it certainly was a full blown powerful usable operating system, it just was too un-Maclike and unrevolutionary to fit into their marketing strategy.
I think, in short, the GP was off-base. I agree with you the first versions of the consumer level operating system "OS X" weren't terribly good, but that's a whole different discussion. NEXTSTEP through Mac OS Server were, by all accounts, very nice and very usable. It was just Aqua changed everything.
Mac OS X is not Copland. Copland was cancelled after a couple of years of development, and NeXT bought to replace it. OS X is, essentially, the latest version of NEXTSTEP.
"Aha!", I pretend to hear you cry, "But it took until 2003 before there was a viable version of OS X!". Well, that's not true either. Within a year or so of the takeover, Apple released, internally and to developers, Rhapsody, the first NEXTSTEP-MacOS combination operating system. This was, by all accounts, excellent, and the second release of it had most of the functionality we associate with OS X today. However, while this was a complete operating system, Apple decided there were two problems with it: it didn't feel right, they wanted the operating system to feel more revolutionary, and developers weren't happy, they felt they needed APIs compatable with the older platform to help them port their application.
The result is this working, fully functional, many people would have killed to get it, operating system was kept off consumer machines, essentially, until Apple felt they had something more revolutionary. That's not to say you couldn't buy it, sometime after Rhapsody was released, the first "commercial" release happened in the form of "Mac OS Server." I can't remember, for the life of me, the year that came out, but I believe it was prior to 2000.
In short, when Apple had a right design, they did have working products in a reasonable space of time, they just felt they didn't have commercially successful products, and avoided selling them (hard) for that reason. But they've been releasing or selling OPENSTEP based operating systems, in some shape or form, for much longer than most people realise.
The project from 1994, Copland, failed because, well, it probably wasn't the right design, and it certainly wasn't managed properly. Everyone makes mistakes.
I don't this is really comparable to the HURD. The HURD has suffered delay after delay, but much of this is because of Linux. The fact Linux exists means a free operating system based on GNU running over Linux, can exist, and the pressure is off of HURD. HURD, therefore, can be a test bed, with the aim towards making it a perfect kernel.
I honestly believe that if it wasn't for Linux, we'd probably have had HURD 1.0 appear over a decade ago.
The big companies that have to do with GNU/Linux and remain supportive of it are either start-ups, consultancies, or hardware manufacturers, or a bit of everything. Microsoft is not a consultancy, not a hardware manufacturer, and is established.
There are, of course, two other issues with Microsoft becoming Linux's saviour in this parallel universe. The first is that Microsoft is owned by Bill Gates, a pioneer of proprietary software who believes, strongly, in using copyright to restrict redistribution. The other is that Microsoft would probably be weaker than Novell is today if IBM did this. They'd be little more than an application vendor tied to OS/2, with no way of changing their business model without rebuilding their entire business.
Henry Ford's contribution, believe it or not, was to add conveyor belts. Not a lot of people know that.
When OS/2 2.0 came out, IBM had made impressive steps at improving the operating system to a point that people could actually like it (1.x, even with Presentation Manager, was widely considered to be awful.) Even then though, it was released after the IBM/Microsoft split, and much of it appeared to be an attempt to play catch-up with Windows. By 3.0 (Warp), IBM was marketing it as the solution to all Windows problems.
Unquestionably, it was a better operating system by that point, but it was several years of development in competition with Windows that lead to that point. Back in 1990, it sucked. And, with no competitor, I wonder if it'd have improved as fast as it did. OS/2 2.0 might have continued to be an 80286 compatable operating system, development would have remained slower than intended, with warring and not exactly harmonous groups developing different parts working for the bickering "allies" (IBM and Microsoft) that made the OS/2 alliance.
I don't know if that's a good thing or not. I guess we'll know when the first viable HURDs start to appear. But, given the major change in direction from the project leaders (dumping Mach, switching to L4), I guess we can see already that HURD wasn't "perfect", and would have been far from it had it been released in the first half of the 90's.
Several companies have released code under the GPL after being threatened with a lawsuit. But no Judge has ever stood up and said "Abide by the license! Release the source, or go to jail!" - to the best of my knowledge, no lawsuit has ever gotten that far before the parties choose to settle.
For example, Linus explicitly allows non-GPL'd software to run over Linux, though an addition to the LICENSE file. In this case, Torvalds wasn't modifying the GPL, he was essentially adding an additional license.
This is allowed because a license (as opposed to an EULA) is just a set of permissions. Each set of permissions adds to any you already have (including your default set of "fair use" privileges.) You can license any project you own under as many licenses you wish, and end users can pick and choose which (complete) licenses they want to agree to. (The word "complete" in that sentence is important.)
Also, while the GPL is unmodifiable for existing projects that do not belong to you, if you have a strong enough case you can persuade the FSF to agree to a modified version for projects you own, on occasion even if the result is a license incompatable with the GPL. For example, the Affero General Public License, whose history you can read about here.
It's not Jobs who's been the primary advocate of single button mice.
Now, to some extent, that doesn't invalidate your point. Whether we know about it or not, if it does persuade companies to do embrace Free Software, then that's a good thing. It's just hard at the moment to believe that's what's happening.