Slashdot Mirror

Consider this, the old NeXT display postscript and NeXT Step code are still proprietary even though neither technology is currently used by Apple.

I'm not sure they could release the source to DPS, even if they wanted to. They had to pay a licensing fee to Adobe for every copy of an OS that shipped with it. NextStep code as a whole, though, is very much in use at Apple. However, some parts of it have been released, such as NetInfo.

Despite the misleading hype, Apple is still closed source. [...] If one has any illusions that Apple is an Open Source company, one need only to speak to the developers of GNU Step who will greet your query with a hearty laugh. Apple open source? No, don't kid yourself.

Who is this attitude going to help?

This is not a black and white issue. It's multilayered and I think it's likely you need more information before you can make a judgement as to the worth of Apple's efforts.

Apple took a big first step with release a fully-functional Unix-based OS in Darwin. True, many of the components (Mach, BSD, etc.) were already available to the public through various other sources and licenses; but this is the foundation for both Apple's consumer and server operating systems. Why is this significant, you ask? It's significant because Apple's engineers are actively developing and improving the core OS on a daily basis. Any time they do this, you benefit. You don't even have to do anything. You just get free code, and Apple writes the check. And furthermore, this code is released under a license that has already had a few revisions, and pretty much everybody seems to be happy with at this point.

Complaining about Apple not releasing the rest of OSX is pointless. Apple, being a publicly-traded, for-profit company, needs to make money. Its positioning as an easy-to-use platform means it cannot possibly justify basing profits entirely on support (not that this business model is flawless in general), and Apple as another x86 box maker just isn't logical. The reason the company could justify the three year development of Mac OS X to its shareholders is that it knew it would gain it back in hardware and software sales. If there was no profit to be made by selling the product, OSX would simply not exist, and the advancement of all operating systems would be affected accordingly. Like it or not, Apple has been and continues to be a is a major influence in evolution of personal computers. As much as slashdotters slam Apple, it's doing pioneering work in Unix usability. This benefits everyone.

Essentially, you're expecting Apple to sit there and spin straw into gold for you. That's not the goal of open source. It's about give and take. Apple is giving you half the kingdom for free, and you just turn around and demand the other half? This attitude only discourages other companies from participating in open source. If you insist on being inflexible and taking an all-or-nothing standpoint on OS source release, then you're going to end up with a whole heap of nothing. You can't just expect Apple to abandon its current revenue streams all at once. That lacks balance and forethought.

The development of every platform cannot be structured identically to Linux. That's just as bad as everything being based on Windows. To avoid inbreeding, you need a variety of concepts, organizations and even business models in order for software to continue to advance. Apple is doing the best thing it can possibly do for the community while still keeping its product line intact.

- Scott

--
Scott Stevenson
WildTofu

I'm sure its been posted here before, but can someone update me with what unix apps will be able to run on OSX when its done? I think I remember apache and a few other things being available but what about other main stream apps. Is it just a matter of a recompile or is the BSD part too nonstandard to port to.

Most command line apps should recompile either out of the box, or with a little nudging. Apache comes with the OS, and I believe Fred Sanchez is making sure new builds continue to work. MySQL has built some OSX binaries, but most of them appear to be for the older OSX Server 1.x (Mach 2.5-based, no Aqua). Somebody is working on PostgreSQL.

The X11 apps are a different story because you have to get your hands on a X server first. There are several different efforts/approaches to this. The most seamless is a relatively expensive ($300) commercial product from Tenon. It runs X apps alongside native OSX apps. You can exit to raw Darwin and run something like XFree86, but you cannot run any OSX native GUI apps until you launch back into the window server.

Overall, anything that runs on *BSD should be able to make its way to OSX pretty effortlessly. Some of the remaining kinks should be worked out by the time the shrinkwrapped product hits shelves on March 24, and I'm sure things will continue to evolve via Darwin. Work done on Darwin is routinely synced with the OSX tree and vice-versa.

Also note that the revamped version of Mac OS X Server, which will come up several weeks after March 24, will be based on the same core as consumer OSX. It will add server-specific packages and some very cool GUI tools for things like Apache and IP filters. It comes with PHP, Tomcat, and MySQL preinstalled. WebObjects with an unlimited license is also included (previously, a 50 client connection per minute limit), as is a mail server, ftp server, and samba. The last three have UI interfaces for them.

A have a full write-up of this new version of OSXS2 on my site. It's slick.

- Scott

--
Scott Stevenson
WildTofu

I'm sure its been posted here before, but can someone update me with what unix apps will be able to run on OSX when its done? I think I remember apache and a few other things being available but what about other main stream apps. Is it just a matter of a recompile or is the BSD part too nonstandard to port to.

Most command line apps should recompile either out of the box, or with a little nudging. Apache comes with the OS, and I believe Fred Sanchez is making sure new builds continue to work. MySQL has built some OSX binaries, but most of them appear to be for the older OSX Server 1.x (Mach 2.5-based, no Aqua). Somebody is working on PostgreSQL.

The X11 apps are a different story because you have to get your hands on a X server first. There are several different efforts/approaches to this. The most seamless is a relatively expensive ($300) commercial product from Tenon. It runs X apps alongside native OSX apps. You can exit to raw Darwin and run something like XFree86, but you cannot run any OSX native GUI apps until you launch back into the window server.

Overall, anything that runs on *BSD should be able to make its way to OSX pretty effortlessly. Some of the remaining kinks should be worked out by the time the shrinkwrapped product hits shelves on March 24, and I'm sure things will continue to evolve via Darwin. Work done on Darwin is routinely synced with the OSX tree and vice-versa.

Also note that the revamped version of Mac OS X Server, which will come up several weeks after March 24, will be based on the same core as consumer OSX. It will add server-specific packages and some very cool GUI tools for things like Apache and IP filters. It comes with PHP, Tomcat, and MySQL preinstalled. WebObjects with an unlimited license is also included (previously, a 50 client connection per minute limit), as is a mail server, ftp server, and samba. The last three have UI interfaces for them.

A have a full write-up of this new version of OSXS2 on my site. It's slick.

- Scott

--
Scott Stevenson
WildTofu

Apples and NeXT boxes are extremely proprietary, difficult to service and poorly scalable.

I'm going to lose my mind. Will you please visit this page at Apple and tell me how Apple's G4 tower is not the easiest machine to service on the planet? Other than the chipset, what component in this box is proprietary? The PCI slots? The AGP slot? The PC133 DIMM slots that accept up to 1.5GB of RAM? The USB ports? The NVIDIA card? Perhaps it's the gigabit ethernet controller?

- Scott
--
Scott Stevenson
WildTofu

Hey AC, you already posted this stuff 4 days ago in the "Jobs Plays it Frank" article. It doesn't make any more sense now than it did then. I don't suppose it would be worth asking why you are doing this?

- Scott
--
Scott Stevenson
WildTofu

to have that viewpoint that Apple keeps coming out with cool things even though they sell like ice cubes in Alaska

~5 million Macs a year. If my math is right (and it's requently not), that boils down to an an average of 9 or 10 Macs sold every minute. I think that's pretty good, however massively bigger the wintel figures may be.

- Scott
--
Scott Stevenson
WildTofu

One of the things that made NeXT work so well was their "developers" app (whose name I can't remember)

Apple is now distributing a much-updated Interface Builder for free, along with all the other OSX dev tools like Project Builder.

- Scott
--
Scott Stevenson
WildTofu

Apple will never ship OS X for Intel until it's too late? The hardware is too expensive, and there's already a HUGE market out there for OSes running on Intel. They could truly give Microsoft a run for their money.

Some questions:

[1] How do you expect Apple to make money?

[2] How do you expect them to take on Windows on its home turf?

[3] How will OSX succeed without any apps (like Office, Explorer, etc)? The only ones that would run on x86 would be the old OpenStep apps. The others are far too processor-specific. Sure, they could come up with a Windows emulator, but what's the point? If you're going to do that, you might as well run Windows.

[4] The Mac loses a lot of its core value once you take it off native hardware. What is it that you like about OSX? Just the interface?

A lot of OSX's value is due to a wide variety of applications and tight intergration with hardware. You lose both of these with OSX86. But at least you'll have seventh OS to select from in lilo.

This simply doesn't make any sense if you look at the details.

- Scott
--
Scott Stevenson
WildTofu

Secondly, I wouldn't use OS X if I had nothing else! The System requirements are huge

You're clearly referring to the beta because the GM isn't out yet. The latter will have lesser requirements. And furthermore, the requirements are largely so outlandish because they have to take into account the Classic environment -- which is, more or less, emulating Mac OS 9 on Mach. If you stick to Carbon (OSX native port of Mac API), Cocoa, BSD and Java apps, then you'll be stylin'. Classic does work very well -- a true achievement in emulation -- but it is a resource hog.

- Scott

--
Scott Stevenson
WildTofu

Things like quartz which contains things (I am assuming) licensed from Adobe.

As far as I know, this is not the case. One of the goals with the transition from NextStep to OSX was to ditch licensing fees that raised the price of the product. This included getting rid of Display Postscript in favor of the PDF-based Quartz engine.

- Scott
--
Scott Stevenson
WildTofu

Apple's stock is in the tank.

I don't suppose you've looked at the Nasdaq composite recently.

Preliminary indications are that Apple users are not particularly interested in the complexity and sluggishness of Apple's latest operating system.

Didn't realize you were conducting a survey.

Processor speed is stuck at 500 MHz.

No need to waste time on research before posting.

Alternative architectures and software are killing Apple on features, price, and performance.

You're certainly right on price, but I'm not going to let you get away with features and performance. Apple is currently kicking ass in terms of desktop video editing and DVD production. And in terms of performance the G4 is, at worst, competitive with x86, and at best, is noticeably faster in things like certain types of encoding and Photoshop performance. Comparing two chip families by megahertz alone just doesn't cut it, otherwise Sun and IBM would be using P4s in their super servers.

There are legions of corporations and individuals who have been disrespected by Apple--from the BeOS community to the Apple clone industry

I don't understand how they "disrespected" the BeOS community except deciding not to buy them. I heard something about them not releasing system specs at some point, but LinuxPPC seems to be doing fine.

The Apple clone industry threw down the first gauntlet by marketing within the fold, rather than attempting to grow the platform user base. They were reaping all the profits, but eliminating much of the core value of the Mac in the process. That would have been fine if they had actually advertised somewhere other than Macworld.

Today Motorola announced 2500 layoffs.

Yes, at the MOBILE PHONE plant.

Everyone does GUI and mice nowadays.

Everyone makes cars with steering wheels and a gas pedals too. So what?

Apple is left marketing decor.

More great research. You're conveniently forgetting the three years of hard work by Apple's software engineering to create OSX. Not to mention all the pro/consumer video stuff they've been doing.

The most reasonable solution would be for Apple to open up. Open up its hardware specs and software so that where now exists little more than a corporate cult, there might exist a vibrant autonomous industry of developers, hackers, and alternative hardware vendors.

Yee-haw! Now just to figure out the part about how to make money. SGI forgot that step.

- Scott

--
Scott Stevenson
WildTofu

Playing devil's adovcate...

What makes Linux cool (popularly usable) to the masses? A Web appliance whose software is mostly open source and free

How many users that make up "the masses" will actually ponder the software license as part of their purchase decision? Hackers will, sure, but the masses aren't hackers.

whose programmers are somewhat accessible/responsible to user feedback

Which programmers? Nokia or the Linux/Mozilla people?

and the ability to upgrade and update without repeatedly paying fees and buying licenses

I don't know any set top box maker that charges for software updates. You generally have to pay for the ISP, or in the case of Tivo, the programming service. But software updates are free last time I checked.

Overall, I think this is probably step in the right direction. But I think the concept needs some more focus. Right now, it sounds as if Nokia is positioning it as an "everything" box, which makes it very hard to market (except on the Home Shopping Network). You need to define boundaries of the product. You need to define who you think will buy it. And it's not exactly cheap. $400 for a 366 Celeron?

Maybe I'll be a bit less skeptical if they had some screenshots of the UI.

- Scott

--
Scott Stevenson
WildTofu

When Apple can justify the loss it will ensue by not selling some hardware versus the gain it will generate by further hardware market penetration, they will release an X86 version. It's simple business sense.

It's a little more complex than that. Apple is about the entire experience, not just individual parts of it. Jobs has said this time and time again. This is beyond business, this is what they do. Case in point: if some generic video card doesn't work on OSX86, it's Apple fault. This kind of thing doesn't mesh well with the brand.

- Scott

--
Scott Stevenson
WildTofu

It's easy to fault them for taking so long with OSX, until you see the newest build in action (the one at Macworld). Then you see what they've been doing all this time.

And in reality, OSX shipping last fall wouldn't have done anyone any good. There were virtually no native apps at the time. The press would have slammed Apple hard for that one. At MWSF, though, there were plenty of booths running native software.

- Scott

--
Scott Stevenson
WildTofu

Look at his history. He always and ever acts solely in his own interest.

I bet he also eats spotted owls and pushes old ladies into oncoming traffic.

This is just silly. The guy doesn't even take a salary, and didn't even have stock for some time after he came back to the company. He doesn't need money. He works at Apple because it's his baby. Anything eccentric thing he has done as the head of Apple's has been a function of that. I wish the CEOs of other large corporations cared half as much about their company's products as Jobs does about Apple's. It's an extremely personal issue to him.

The worst thing you can say about him is that he is fanatical about people experiencing Apple's work as he intends. He holds contempt for retailers that do a poor job of displaying Macs, or rumor sites that announce products outside of the context of an event. Though, if I was involved in such a project for 12-18 months, I would probably feel the same way.

- Scott

--
Scott Stevenson
WildTofu

If Apple dies in the process

You make this sound as if this could actually happen anytime soon. They have $4 billion in the bank

Steve is dealing with a changing world that is not bending to Apple the way it used to because the younger generation doesn't remember the "old" Apple and, frankly, couldn't care less

I would agree with you, to a point. The catalyst, though, is Mac OS X. It's drawn interest from all sorts of people that didn't give a thought to Apple before. Maya is certainly one of the most visible, and had quite a sizeable/packed booth at Macworld Expo last week. But I was also quite surprised to see Roxen there!

With Mac OS X and software like iDVD, Apple has a much more compelling story that it has had in some time. And except to see the advertising change accordingly. After March 24, there will actually be very good reasons to own a Mac beyond it just sucks less than Windows.

- Scott

--
Scott Stevenson
WildTofu

Maybe OS X client is not meant to complete with Linux, whereas OS X Server 2.0 is....more-so W2K Server I'd think though;)

http://www.apple.com/macosx/server/

See a preview here...

http://wildtofu.com/resources/osx/osxs2-preview.ph p

What is wrong with tape?

You can't put it in a computer, it's linear, and it's more prone to damage.

And while I don't think you can put a DVD-RAM disc in consumer DVD player (anyone know for sure), this is a step in the right direction.

- Scott
--
Scott Stevenson
WildTofu

Seriously, just what kind of an ass-hole are you? You've got the processor wrong. Its a G4 not a G3.

I'm pretty sure that post was a joke.

- Scott
--
Scott Stevenson
WildTofu

Even my untrained troll detectors see through this one. You make it way too obvious with the 500MHz limit bit (which is false).

- Scott
--
Scott Stevenson
WildTofu

or $500 you can put together a really nice system. High quality Abit motherboard, AMD Thunderbird, Micron memory, IBM ATA hard drives, etc.

Like I said, if all you care about is price...

Most people don't want to build a PC from scatch. I did that for years, and just got sick of dealing with all the problems.

If Apple had had any brains they would've made a sub $1000 "iMac" without the space-wasting monitor built into it.

Sounds good to me.

- Scott

--
Scott Stevenson
WildTofu

From what I can tell, he knows that OS X is the last chance for Apple.

I suppose... but only in the way that Earth is the last chance for humanity.

- Scott
--
Scott Stevenson
WildTofu

Try this experiment--go to Pricewatch and see what 500 dollars will buy you in Intel architecture hardware. Do your best shopping. Now see what 500 dollars will buy you in Apple hardware.

If all you care about is price, then yes. Go buy a eMachines or Dell. Though, if you can get past your preconceptions and actually figure out that there are advantages to Apple's approach to product development, then you may find yourself on the other side of the argument.

G4 performance is, at worst, reasonably competitive with x86; and at best, actually quite a bit faster in some situations. Though, I know a lot of people just refuse to believe that clock speed could actually be an inaccurate measurement of overall system performance this day and age.

At some point the Apple PR machine will run out of shit and the lie will be exposed for all to see.

Have you actually used OSX yet?

- Scott

--
Scott Stevenson
WildTofu

Does LinuxPPC run faster? Using similar software, which system runs the app faster? Which OS eats up more memory? etc.

It's hard to say until OSX goes GM, but in the end, the raw speed difference will probably be somewhat insignifcant compared to the functionality comparison.

- Scott

--
Scott Stevenson
WildTofu

Yep, all the second CPU s doing is keeping the first warm and cozy. :-) The machine cost $4k, so I hope they at least buy OS X when it is released so that the nice hardware won't go to waste.

Depends on what he's doing on OS9. Besides Photoshop (which is the poster boy) some apps do take advantage of the dual CPUs for encoding and video rendering and such. The point is that the second CPU only adds an additionl $300 or so to overall cost, and OSX (out March 24) will make proper use of it. Might as well toss it in.

- Scott

--
Scott Stevenson
WildTofu

No, what's really sad is to see a sweet machine crippled by a crappy OS.

I agree. Mac OS X ships on March 24. Some people may prefer LinuxPPC, which is what this article is, oddly enough, about.

- Scott
--
Scott Stevenson
WildTofu

This post has so many factual errors that it's comical. Ironically, he bashes "Classic" Mac OS when the story itself is about LinuxPPC. I'd love to hear the rationale from the people that marked it up as "Insightful" and "Interesting."

- Scott

--
Scott Stevenson
WildTofu

Are you just jealous that OSX is gonna do more to bring *nix to the masses than your 6+ years of obscure forum posting and mutual masturbation have?

Easy, we're all on the same side here for the most part.

I'm as big an Apple supporter as anyone, but I think you went a bit overboard. A lot of people would like to see two button mice on Macs -- including something people inside Apple, apparently, as OSX's contextual menus do work with the right mouse button on most USB mice.

- Scott


--
Scott Stevenson
WildTofu

Last I've heard, in the non-server edition, they've obscured the *nix parts enough that only someone who wanted to get to them could.

I don't know what this means, exactly. OSX hides things like /etc, /bin and such by default when the user is in the GUI. Just launch Terminal and you can see everything. You can change the default behavior of the GUI (at least in the beta), by flipping a bit in an XML file.

You should also get up to speed on the differences between Mac OS X (shipping in March) and the new version of Mac OS X Server (shipping April/May). They are from the same codebase, and are considerably different that Mac OS X Server 1.2 (essentially Rhapsody), which is currently available.

Bringing "Unix to the masses" isn't going to consist of expecting everyone to become comfortable with /etc and bash. It's going to consist of taking the functionality and features of Unix, and wrapping them in a consistent, approachable user interface. The masses aren't going to change for Unix. Unix will have to change for the masses (I know I sound like a broken record).

This thing also runs on Mach, which is SLOW. (MKlinux anyone?)

There is a big difference in between MkLinux and Darwin.

- Scott
--
Scott Stevenson
WildTofu

With /etc, I can just do a `ls *programname*` and easily find the configuration file.

The biggest problem isn't that the files are stored in etc, but rather they all have varying formats and aren't named consistently. The fact that "etc" is an exceptionally obscure title doesn't help matters either.

I understand why things were named the way they were (usr, bin, sbin, etc), but things have changed in the last few decades.

- Scott
--
Scott Stevenson
WildTofu

Linux is still hard to use, and problematiclly, disunified. [...] While the desktops are more or less easy enough to use, the system itself needs major retooling. Here are the problems I see

Interesting, for virtually every shortcoming you mention, Apple is working to address in Mac OS X. This isn't purely a OSX evangelist session (I'll save that for later). But perhaps rather that working so hard to emulate Windows, maybe Linux developers should look towards OSX for inspiration (though stopping short of making exact clones of Aqua). From a system design perspective, Linux is far more similar to Mac OS X than to Windows.

Apple has already solved a lot of the problems Linux distros are struggling with. Mac OS X stores a substantial amount of configuration information in NetInfo, a network-distributed database. This software is open sourced, incidentally. The remaining config information is stored in XML files in both the user's home directory and in system directories. Frameworks and Bundles greatly simplify software development and distribution. XML files (with DTDs defined by Apple) are used to store meta information about applications such as icons, localization information, and architecture-specific settings.

Most of the many directories are named normal things like /Applications, /System, /Users, and /Developer. The BSD directories like /etc are still there, but they are hidden from the GUI and the user is not expected to interact with them. They are mostly there for running BSD apps. There are a few different APIs available (Carbon, Cocoa, Java), but all native apps use the same base application services for rendering and such.

And to top it all off, the installation process is extremely easy and swift. Beats the pants off a Windows install. Don't be fooled by the casual appearance of Aqua. This is a brilliantly designed operating system, both in terms of architecture and interface.

Apple is even making strides in making unix-based server functionality accessible to just about anyone. I suggest taking a look at the completely revamped version of Mac OS X Server which will be available this spring. I've used things like Linuxconf before, and it just can't hold a candle to the first-hand demos of this new version of OSX Server I saw at Macworld Expo.

And in case anybody thinks I'm some sort of Mac bigot, please note that my servers currently run Linux, and I've been using various forms of Unix for around 6 years, including Solaris, SunOS and FreeBSD.

- Scott
--
Scott Stevenson
WildTofu

I know you're really excited about Linux, but sometimes it's important to see the problem for your critic's perspective:

I've long predicted that in the next recession, Linux use will grow by leaps and bounds. That's because licensing issues will be more important to people, and companies will be looking for more ways to save money.

That would be nice, but I think it's pretty optimistic. The difference is you see Windows as an acceptable (and in reality, a preferred) casualty. Many business and IT types do not. For them, ditching Windows for Linux to save money would be the equivalent of switching from light bulbs to candles because the electric bill is too high. Many people think of Windows as an essential part of what a PC is.

The best part is -- there's nothing MS can do about Linux! They can try FUD, but enough people know the truth by now to make it really effective.

If there's one thing that Microsoft has taught us, it's that FUD does, in fact, work with a significant quantity of the population.

- Scott
--
Scott Stevenson
WildTofu

it seems that you need the $3499 model to get it. its not even available as a choice to upgrade in the custom configurations of the other models...

This is because the drives are quite scarce right now. This technology will eventually filter its way down to the other product lines, as DVD and FireWire did.

- Scott

--
Scott Stevenson
WildTofu

Actually, IE on the Mac is notoriously slow at rendering Slashdot. iCab, OmniWeb, Netscape and Mozilla all render Slashdot in a few seconds.

Hmmm, I haven't had the same results as you have with complex pages in Mozilla. It's not surprising that iCab, OmniWeb and Netscape would render faster, though. For the most part, all they care about is HTML, and it one case - CSS (albeit poorly). You can't see it on Slashdot, but on sites that use CSS extensively (which are rapidly becoming quite common), you'll be wishing for the MacIE rendering engine. And that sophitication doesn't come without extra consumption of resources.

- Scott

------
Scott Stevenson