Slashdot Mirror
X11R6.4 And Apache On Mac OS X Beta
Adam Attarian writes: "Tenon has released public betas of two new products for Mac OS X, XTools, an X-Window Server/Desktop based on X11R6.4. They integrate it with Aqua using both Cocoa and QuickDraw, with OpenGL support planned (the screenshots are cool, very cool). Also out in public beta is iTools, an Apache Port, that looks to have all the goodies like sendmail, DNS, and SSL. Now, if only we could get an AltiVec dnet client for X..." I should have been more clear - you can order this stuff right now - but their site so slashdotted, you can't really get through.
The X Server doesn't magically Aquify your X apps.
You will still get plain old X apps running in an
Aqua window manager.
Now, they could Aquify the GUI toolkits (ie: GTK+ or Qt) and allow you to port apps to MacOSX (as opposed to just using an Aqua theme XClient-side).
They could Aquify the Athena widgets too.
Could they capture Xlib calls and output PDF based on that?
Could they provide optional Anti-Aliasing of text and line-art on a per-program basis (in case its screws up some programs)?
DCMonkey
Rapid porting of all existing X11-based apps ?
--
The world is divided in two categories:
those with a loaded gun and those who dig. You dig.
--
The world is divided in two categories:
those with a loaded gun and those who dig. You dig.
I do other things with computers, but the main one is to save me from dragging myself into work on a weekend or after hours.
I'm sure Bill Gates is laughing his ass off.
Where do you get your acid?
If there's one inconsistent platform out there, it's certainly Windows. Even in the same version (ommit the pletora of conflicting UI from their pletora of versions of Windows), applications are always different.
This would be a good project to put up on cosource.com.
Until then, most folks will just run it in Terminal.app.
All X applications still have to formulate their drawing commands into X drawing commands and pipe them through the high-speed local socket. The local X server then has to read the command from the input, decode and interpret them, and then call its API calls to draw them.
Actually, no, that doesn't need to happen. What happens is this:
The application code uses the API to put a call to an X11 library's code, and the library takes implementation-specific steps to get carry out the application's drawing requests.
All of the overhead that you've mentioned is really an implementation issue; there's no reason that an X library can't be written that will forgo X11-protocol-usage and just draw to the screen.
-rozzin.
... Mac OS X will eventually displace Linux and *BSD and GNU and other Free Software that chooses to duplicate rather than evolve....
As we can see, the Free Software groups just attempt to reimplement what's been done in commercial software.
Have you read anything about the HURD?
GNU really isn't just a reimplementation of any OS that I've ever seen--it may be POSIX-compatible, but it sure as hell isn't unix.
-rozzin.
Long enough to make me wonder why, to this day, Microsoft and Macintosh haven't implemented something as useful. X may not have a lot of pretty client apps or great games, but it's one of the most useful things I have on my boxen. Being able to use graphical apps from my laptop by displayed on my desktop or vice-versa is such a godsend that I don't want to live without it. That's one of the main reasons I use UNIXen at home rather than Windows or MacOS.
It really is neat stuff.
Ita erat quando hic adveni.
Which brings us full-circle again, because that's exactly what this app is: a sort of "MacX for Aqua". It doesn't replace the existing MacOSX desktop, but merely allows X11 clients to display on it. I'm sure we all agree something like that is handy.
There's room enough for both sorts of displays, I think.
Ita erat quando hic adveni.
As I recall, all program linking does is pull a copy of the application over the network and run it locally; it doesn't run on the remote machine.
the fact is you still have a video server on OS X
Yes, you do. No, it has no compatibility whatsoever with X11.
Unfortunately, there isn't mention of compatibility issues with "X" applications
That's because there isn't any.
it would seem that since Apple is claiming "Unix-savvy" for this OS, there would be at least some level of compatibility
For someone who nitpicks other people's grammar, your own grasp of the English language seems a little...lacking. Anyway, Apple isn't claiming that OS X is "UNIX-savvy" (whatever that's supposed to mean), but rather that it is (reasonably) POSIX-compliant, which comes from the BSD-based bottom layer of the system (Darwin), which has absolutely nothing to do with what variety of graphic interface they provide with it. Their implementation has ZERO support for X.
But please, don't let your lack of clues stop you from posting...
Program Linking allows a server to run on one machine, and a client to connect to that server. The client application is running on your Mac, the server application is running on the other Mac, and they talk.
There is no MacOS equivalent to the "Run on one machine, but display on another" model that the X Window System uses.
Pope Felix the Scurrilous.
Pope Felix the Scurrilous.
Computer Geek by day, religious Icon by night.
An excellent counter-analysis!
Given SGI's bargain-basement price, do you think that it would be cheaper to buy SGI outright and get the SGI Origin-class hardware outright or do the R&D yourself?
What it mostly comes down is what is Apple going to do to grow their business? Regardless of how cool OS X/Aqua/Carbon/Insert-Buzzword-Here is, they cannot compete on price with x86 and will not get back into suit-n-tie type businesses anytime soon, barring an unexpected MS implosion.
Apple has a hole in its product lineup -- they cannot expect to grow a niche market much beyond what they already control. They may grab the odd x86 Unix workstation in some environments, but without serious hardware support for servers and storage, they will remain a niche product.
Buying SGI would give them an immediate entree to the high end and enable them to fast-forward the Macintosh platform as a low-to-mid server market where vendors like HP, IBM and Dell dominate as well as giving them scalability on the same (long-term, anyway) OS *across* the entire product line.
It's not a perfect marriage. SGI is in tough shape. Apple's business is much more focused than it was, and adding high-end Unix to its traditionally low-tech userspace may be tricky.
But unless Apple are willing to get into more than just selling new boxes to the people that already own Apple boxes, their growth will remain rather stagnant.
On the one hand you have the definition of the API or protocol, this may be well thought out or not. On the other hand you have the implementation of the protocol or API and this may be well done or not.
A good protocol can feel bad if nobody has taken time to write good code for it. At the same time a bad protocol can feel good if it has been optimized to the point of perfection.
This also goes to explain why one implementation of X11 is fast compared to another implementation on the same machine. If you aren't happy with the implementation improve it.
Also as long as it doesn't effect the specification an implementation can always be optimized with causing too much grief. Now try optimizing the specification and just watch who screams!
Jumpstart the tartan drive.
I guess it would depend: do X apps run unmodified on OS X? (My guess would be probably not.) So if you wanted to run X apps, you would need an X server, right?
Only Mac people would think it would be silly to run non-Macintosh apps on a Mac.
My journal has hot
I got the screen shot to load finally. They're running GIMP and EMACS on OSX. This will definetly increase the usefullness and power of OS-X, because now X apps won't have to be rewritten to use the Aqua interface.
Its making the machine much more workstation like. Its good to have all that scripting power like perl etc on a box. Scripts and remote control are the biggest failures of non unix OS (I know applescript , visual basic, etc.. but they're weak)
"Also out in public beta is iTools, an Apache Port, that looks to have all the goodies like sendmail, DNS, and SSL"
Not quite. Apache, sendmail, dns, OpenSSH; it's already there in the OS X public beta standard installation. iTools is merely a configuration tool for the aforementioned software.
And besides... this isn't 1975 anymore. I think we can do better than blasting bitmaps across the network using a broken security scheme.
Where did you get your information? The X Protocol sends raw bitmaps only in the case of a client server bitmap x-fer request (seminal transfer of an XPixmap, or an XImage XPixmap copy). Otherwise it uses other primitives like lines, rects, font glyphs, and server-side bitmaps (XPixmaps).
The ratio of people dogging X11 to people who actually understand X11 hovers around 100:1. There is so much uninformed drivel regarding X11 floating around that zero productive criticism of X11 can actually be accomplished, these days.
X11 would benefit from a new imaging model, something akin to SVG or PDF, much like Quartz or Display Postscript. Unfortunately, the bulk of graphical computing applications on *all* platforms is adapted to the old, limited model of the original X11. What you don't seem to recognize is that X11 survives precisely because of the abominable state-of-the-art in GUI technology (again, on all platforms).
Your precious Quartz might have changed that, except that it's owned and developed by Apple Computer. That, in itself, is a death warrant for promising technologies.
MJPDon't try that "protecting the children" shit you people use to keep the tits and bad words off my TV. --Seanbaby
when i typed 'where perl'
goodbye Sherlock, hello Grep !!
Now i just gotta hijack those Darwin tools.
i've been tooling around with OSX at work for the last few days, and i've noticed something distressing which i have heard no news or indignation about: gcc is missing, gone, completely not there. as far as i can tell, there is no way to compile on this thing. why has there been no press about this? has no one noticed this besides me? comments? thoughts? death threats?
strange things are afoot at the Circle K...
Or like saying "Linux runs on Intel", and "Windows runs on Intel", so that implies "Linux runs on Windows". Except that's actually sort of true, since there is VMWare...
:-)
To really confuse things, now you can (theoretically) get MS Office on MacOS X by running it under Wine under X under MacOS X. An interesting alternative to simply running the native Mac version of Office. To confuse things even more, just add SoftWindows and VMWare somewhere in the above chain.
Oh, I never thought MacX for Aqua was a bad thing...I was arguing the "goodness" of X...I just think it can be done better ;-)
--------------------------
And since X is a protocol, it is neither slow, buggy or holding anything back any more than HTTP is slow, buggy or holding the web back.
Good lord, man, have you used X11?
--------------------------
X has a cool concept. But if it ever wants to be used by the masses, in needs some reworking.
Windows and MacOS are significantly better (IMHO, YMMV, yaddah yaddah yaddah) than X because they simply do things faster. X with all its networking overhead makes an inefficent desktop.
Do I need to display remote apps? Not often. Even then, most of the time a terminal will do. I would like to see something for Linux much more Win/Mac like - a fast, no-network-bullshit display system. If I want to run remote apps, I'll start an X server similar to X-Win32 or MacX.
--------------------------
Yes, I have used it. Aqua is pretty, but I'd prefer a different theme.
But I'm wondering what the point of your comment is....
--------------------------
VNC (there's a java client which should run under MacOS X as well)
Actually, it runs suprisingly well under Mac OS 9 with IE5.
- Scott
------
Scott Stevenson
Scott Stevenson
Tree House Ideas
You have your facts backwards La-Z-Bastard.
Tenon actually had their iTools product BEFORE Apple introduced their iTools shwack.
Why not the Macintosh (OS X) Foundation Classes ?
Right...
Yeah, that was the joke. I had a written in, but forgot it was HTML formatted.
Right...
You can EASILY recompile apache under MOSX. YOu can either just drop your version into /usr/sbin/apache or you can work for it a little and compile it so it exactly replaces the built-in installation (the target directories are a little different than in the stock install). Neither option is particularly hard if you're worth your paycheck as a hacker.
As for adding mod_perl and php support: it's trivial for the first option I described and it should be *fairly* trivial for the second option (I am doing it this afternoon, if all goes as planned)
You mean like Gimp? I guess there is truth in some advertising...
An Altivec GCC is available from here. It works, but you may need to put your own headers in place in some instances. You have to sign up for a mailing list to get access to the archives.
but - don't bother buying a G4 just for Linux. I have a G4 Cube at 450MHz. Although the system is good, reasonable video, silent except for the hard disk, the much-hyped G4 is at most 60% of the speed of my 800MHz Athlon, and of course cost more. Also, unless someone knows of a driver, the cool speakers it comes with don't work except under MacOS. Great keyboard though.
Apple's much-quoted benchmarks may be true for Photoshop, but not for compiling a kernel.
You can compile Java natively on Mac OS X, can't you? And do something to it (Swing?) to get it to use Aqua? I think when you do this it looks and acts just like any Mac OS X native app.
I'm not that into Java, but I keep seeing things about Mac OS X's Java support that sound like it goes above and beyond the call with Java. Maybe that's what Tenon is referring to.
Also, there may be Mac users who want to run this stuff, but are happier to pay Tenon a modest fee to deliver it already working, rather than have to mess with anything themselves.
The previous poster is right - Aqua is just a theme. But if you substitute MacOS X for Aqua...you're still wrong.
It's kinda like saying Netscape Navigator runs under Windows, and Microsoft Office runs under Windows, so that implies Microsoft Office on Netscape Navigator.
The crimes of eBay are a disgrace to it's pig latin heritage!
Why would X running under MOSX be any more of a threat than Linux running on the PowerPC? I think the previous poster hit the nail on the head - there is a significant difference in skill set.
The majority of Mac users don't give a damn about the techy details slashdotters hold so near and dear - they just want to get their email, surf the web, print pictures of their newest nephew, and maybe edit a newsletter or two. Without crashing. Whatever gets them that is fine.
I can't see masses of Mac users suddenly moving over to a platform that requires so much more of them simply because python is available. Remember - MacOS 9 already offers power tools (perl, AppleScript, Webstar server) - but *very* few Mac users ever delve into them.
The crimes of eBay are a disgrace to it's pig latin heritage!
Um....nothing runs on Aqua. Aqua's just the name of a theme. A very deep, well integrated theme, but a theme nonetheless.
Anyone in need of some karma?
/.ed, someone mirror the d/l's please 8^)
Tenon is being
-----------------------
Moderator's essentials
Well, that's a perfect opportunity to create a new a/c and start whoring all over again! Isn't it?
-----------------------
Moderator's essentials
Once X becomes a commonly installed piece of software on the Mac (say like Winzip on the PC), Microsoft will win even bigger. Why? Because the Mac users will be subjected to some of the worst interface designs ever. Every program having it's own quirks and not being consistant at all. Something the Mac is supposed to have over all other platforms is suddenly gone.
I'm sure Bill Gates is laughing his ass off.
*Real* Mac Crumudgeons barely accept the novelty of OS 7.1 and have no intention of moving beyond it!
Some part-real part-synthetic Mac Crumudgeons accept 7.5. Only simulated Mac Crumudgeons accept OS 8.1 and/or 8.6. *NO* Mac Crumudgeons of any stripe accept OS 9!
I don't diagree with the general tone of your statement, but please get your facts straight! OS 9 is the up-to-date latest and greatest release OS from Apple. Crumudgeons wouldn't touch it with a ten foot pole.
Mac OS X already has its own GUI framework, and there is absolutely no way that X could improve upon that. After all, the Mac GUI has always been its main selling point, whereas X is roundly, and deservedly, slated as being slow, buggy and holding back Linux on the desktop.
I honestly don't see the point in this. Surely it's not progress, it's going backwards?
The point is to give OS X users the ability to run legacy X11 apps, and to display X11 apps that are running on remote machines.
Just to make it nice and sparkling clear: This is not a replacement for Aqua. This X11 server will run under Aqua, and allow you to display X11 apps side by side with normal Mac applications.
X11 is nasty compared to a real GUI, but the network transparency is damned nice when you need it, and there are a few decent X11 apps around.
As one can imagine, this is both good news and bad news for Apple They undoubtedly welcome this product as it gives the connectivity they so sorely need; particularly in their important scientific research market where they're already fairly popular. On the other hand there is likely a fear that developers might 'port' to MacOS X by simply running their apps in a X term session (possibly a cut-down server licensed from Tenon) and not go native.
I'm not so sure that Apple has much fear that this will happen. The reason? Apple isn't including the X Server with OS X. If you try to sell software to someone and require a 3rd party product (especially one from a fairly small company) you're asking for trouble. First, people are hesitant about installing more stuff on their computer (maybe OS X will make this better, but I doubt it), second, their support department will be driven crazy by compatibility/version problems (are you running Tenon X? XFreePPC? Which version?) and third, companies worry that the 3rd party might go out of business leaving them in a bit of a bind. It's hard enough to deal with the foibles of different versions of an OS (and, yikes, MS makes it truly annoying by upgrading the OS everytime you install a new browser or word processor).
I'm not saying that there won't be companies that take this route, but that the vast majority won't and this is enough to make Apple not worry so much about this.
Anyway, I admit that I stand corrected regarding the X server; I mistook "X server" to mean "video server," which is (now that it's been pointed out to me) obviously not the same thing. My apologies to any I offended; I'll go flog myself with a wet noodle now. ;)
Your Friend,
Your Friend,
D
Apple did a lot of work to get their X server working, and from what I've seen at a couple of expos (don't have a Mac at home to test it myself, yet... :( ), it's damn smooth.
If Apple would only open the source for their video server... Now that would be cool.
Your Friend,
Your Friend,
D
Mac OS X uses DisplayPDF. At least IBM's AIX uses (or used last time I touched it...) DisplayPostscript. We all know that Postscript is a subset of PDF. So a recompile of IBM's X server would run under the Quartz layer of OS X.
So we just have to wait for IBM to open source their X server. *grin*
Yeah, I missed the disclaimer ("I don't just mean commercially") the first time around... :)
I imagine, though, that some of the same reasoning applies. Since OS X will ship on every Mac at some point, people willing to go to the trouble of replacing the UI are probably a pretty small subset of Apple consumers. As I pointed out below, the real curmudgeons will run OS 9 for the next fifty years if they can. I mean, nothing can please those people.
At any rate, from what I can tell, the state of the GUI art in the *nix world has a long way to go before it produces something that can compete for Apple's core market (the non-hacker). And even the hackers will have to contend with the fact that Aqua goes into certain depths that may or may not be accessible to third parties, making UI replacements that much less desirable.
But who knows? It is for sure an interesting shift, regardless of the eventual impact.
I don't quite see how, at least given your scenario.
If a decent cheap/free X server is running on top of OS X, then it is running on top of an OS built by Apple (revenue source number one) on a PPC box built by Apple (revenue source number two). That changes exactly nothing. If someone decides to forego OS X and, say, run XFree86 on top of Darwin, this removes only revenue source number one. If they decide to forgoe OS X and a PPC box built by Apple and run XFree86 on top of an Intelized Darwin, this removes both revenue sources, but is so far removed from the mainstream Mac user (in terms of skill-set and desired outcomes) that it still fails to be a threat. Anyway, at that point, there are plenty of solutions that would probably make more sense for most people looking to do such a thing.
If you keep reading through this conversation, you'll eventually come across several instances of this link.
There's been plenty of press on the subject. Most of it pointed toward an article at MacAddict which explained how to strip the dev tools out of Darwin.
I believe less drastic measures are being shipped to developers as we speak, or at least will be in the near future.
XFree86 on Darwin and OS X
OS X comes w/apache 1.3.12. You set it up with a similar interface to the web sharing control panel under OS 9. The problem with Apple's version is the modlules it comes with - while webdav is there, there's no mod_perl, no php, and no mod_ssl. God knows how you'd add them, I'd be happier just compiling the whole thing from scratch. Tenon's value added is mostly from the perl/cgi it ships. Htdig, a webmail client, and a web based admin setup.
Not sure what their bind distribution is like. OS X comes with bind and sendmail, but you'll have to work out netinfo to get wither to work properly as netinfo overrides the hostname.
what's the big deal? You can get this stuff for free here:
http://mrcla.com/XonX/
Duh. Why pay when you can get the crappy x-window interface for free?
--- only for the squeamish
XFree4 *IS* based on X11R6.4
It makes a lot of sense if XFree 4 is based upon X11R6.4, I hadn't realised they had opened it up soI assumed (wrongly) it must still be based on R6.3. Anyway chalk another duff headline up for
Hemos also changed the Or to an And
Paul Priest
Vita Nuova
Paul Priest
Vita Nuova
http://www.vitanuova.com/
This is not open-source, and certainly not free!
And to everyone who has said that this will not improve upon Cocoa, think again. It would be slightly disappointing if Mac's become *nix based only to find they don't have an X server, Remote logins anyone?
I don't speak in any capacity for Vita Nuova, I'm just the tea-boy
Paul Priest
Vita Nuova
Paul Priest
Vita Nuova
http://www.vitanuova.com/
OSX is different, though. Being *nix underneath, there's a whole batch of existing alternative UIs which can be ported onto it and bring users with it. I've seen plenty of comments from people who want to use OSX but also have the option of their favourite window manager.
A significant number of people are going to do this. If the practice spreads, will it reduce Aqua to just one interface among many?
Maybe it isn't a realistic scenario - but it could be.
They won't be the only kind of OSX user. There will be others wanting to use other interfaces on OSX - and they will. This will have an impact. Question is, how much.
I just downloaded Tenon's Xtools and installed it on MacOS X PB. Apart from the fact that In the course of literally less than 5 minutes it crashed twice (OK, so it's beta software, but really - the X server died just launching xcalc!) I now see clearly how Apple was very wise to not go the X server route for MacOS X.
I was among those who though that Xtools might be useful, even a threat to Aqua (if enough X apps were ported, etc.).
However, my brief experience with it only served to remind me just how ugly X is, what a throwback to the 80's in terms of GUI sophistication, and how buggy the whole X experience is (I've used XFree on Linux ix86 and LinuxPPC too).
Bottom line, X will never be a serious contender for the Mac desktop because X itself is so damned ugly, flaky and slow. Xtools might however be useful for running remote X apps on a local mac, or running open source X apps locally whre no Aqua port yet exists.
"if someone does provide a decent cheap/free X server for OSX then a huge number of apps suddenly become instantly accessible to Mac users."
Which no one will want to use, since there are already existing Mac apps which fill that need.
Put it another way. Tenon has had an X server for MacOS (7, 8, 9) for many years (Xtools is news only because it runs under MacOS X which has a BSD layer). They even ported numerous X apps to run with it. Yet there was no stampede in the Mac community to use Tenon's X server to run X apps.
Why? Because if you're used to native Mac apps, then X apps look really ugly, inconsistent in GUI, flaky and slow. The only real use for an X server on MacOS (7, 8, 9 *and* X) is to run remote X apps. For local apps, there are already existing Mac applications that do the job as well or better.
In addition, there are other alternatives for running remote X apps as well, such as VNC (there's a java client which should run under MacOS X as well).
If X runs on Aqua, and MS Office runs on Aqua, doesn't that imply MS Office on X?
With all of the litigations these days iTools is an awful choice for a name. Ever heard of that set of "internet tools" that Apple offers free to all new Mac owners. It's called something like, hmm .... let me think. Oh, it's iTools !
Let's extend the example:
Photoshop is installed on Mac #1. You want to display on Mac #2. If you are running something compat with X, you can launch Photoshop on Mac #1 (i.e. running on Mac #1's CPU, using Mac #1's RAM, etc) but have the input/output on Mac #2. Mac #2 needs little or no CPU power, because al the actual work is done on #1.
AFAIK, only X supports this fully. PCAnywhere and VNC are addons to give partial functionality. Windows Terminal Server does some of it (I don't know how much) but I haven't heard anything about that software in over a year.
--
Non-meta-modded "Overrated" mods are killing Slashdot
(Hey Ryan! Here's your proof!)
All in all I can't see how you can seriously ask this from Apple. X-server under MacOS X is a perfect example of software for a select group of users(surely less than 5%). Third parties should create a distro with or without support. Otherwise you can compile it yourself.
Why not ask Apple to create the best platform for using software. You can't ask them to program and support your particular piece of software.
The Drowned and the Saved - Primo Levi
That said, X protocol is pretty brain damaged. NeXT had it right with Display Postscript, and now Apple has it right with DisplayPDF. Let's hope the GNUStep team completes their Display Ghostscript X server extension so we can all enjoy the benefits of DPS, for free. GNUStep is probably the most important and underappriciated Free Software projects around...
Er, insert the phrase "is faster" between LinuxPPC and Mac OS X server up there. I had a greater than symbol in there, but I should've realized that it's an HTML symbol. My bad.
Haaz: Co-founder, LinuxPPC Inc., making Linux for PowerPC since 1996.
-- haaz.
Now. This is not oS X final or the beta. Apple is probably going to spend a fair bit of money on the theoretical optimizations for the Mach microkernel that can (theoretically) make it faster. But, as far as the public knows, Linux holds the speed record right now.
The review is at:
http://macworld.zdnet .co m:80/1999/11/reviews/linuxppc.html
That said, I am not against Mac OS X. I wish it the best of luck. I also would like to sell Linux for the Mac. ;-)
Haaz: Co-founder, LinuxPPC Inc., making Linux for PowerPC since 1996.
-- haaz.
Why would Apple, fresh from it's own brush with death want to purchase another ailing company, particularly one with such poor prospects?
Apple has finally gotten it's hardware habit somewhat under control. The original lean line-up has now ballooned but at least almost everything is being based on common motherboards. SGI would bring in a completely different set of technologies with almost no common elements.
Apple has also pared down it's software holdings. No longer does it try to compete with it's own applications but has spun off/shut down/de-emphasized most of those distracting and ultimately non-profitable lines of business. SGI would pull it back into those markets in a big way.
Finally, Apple is once again focussing on profitable markets, one where it can sell either enormous quantities of hardware (iMacs) or fewer but highly marked up ones (Cubes.) SGI doesn't fit into either category particularly well, nor does it's own category seem much longer for this world.
This appears to be a case of looking-for-a-buyer-for-SGI & not what-could-Apple-spend-it's-bucks on. Sure if gifted with SGI I'm sure Apple would be happy but $ for $ they'd be a lot better off spending their money on R&D to extend their own stuff into what's left of SGI's market.
I don't read ACs: If a post isn't worth so much as a nom de plume to its author then I wont bother either.
Their application runs under BSD & 'till recently they've been requiring their customers buy a standard distribution. They've now expanded their support to the various free BSDs and are now considering MacOS X.
As their customer base is non-technical folks with limited support resources something like MacOS X would seem a logical step. However they're unwilling to commit resources and fork their product until MacOS X has been out in the market awhile, appears successful & enough customers seem willing to adopt it.
One interim strategy we discussed was simply continuing to run their application under an X Server and pretty-much ignoring the whole MacOS X-side of things until they've determined how far they want to go with this.
Generally one can get a significant discount from MSRP if one licenses only the required parts of a product, limits it's use to running the licensed application, handles product support & is willing to include an upgrade offer in the packaging. If the X Server vendor is open to licenses like this then the cost per unit could be acceptable.
Thus their product could ship almost immediately and with the exception of the added X Server install and some other "MacOS-X-ifications" (graphical installer, file locations, frameworks, icons) would be the same as their other releases. Indeed we guesstimated it would take 2 or 3 days to modify the unix side of things and another 2 or 3 to produce the MacOS X side.
This is exactly the sort of thing Apple probably worries about. Ultimately it might be a win for them but in the meantime they'll have a quasi-ported-application runnning on their shiny new OS.
As to concerns regarding the supplier, well, yours aren't new ones. Generally one can request an audit of the company to determine it's health, invest in things like a code-escrow policy, have alternative strategies in place (I expect there won't be one X Server vendor for MacOS X forever), etc.
Finally these concerns are less strong then they might be otherwise as this would intended as an interim step. Should the product prove popular then the full port to Mac OS X would take place with the Aqua GUI etc. and then the licensed X Server could be dispensed with.
I don't read ACs: If a post isn't worth so much as a nom de plume to its author then I wont bother either.
Software-wise they're not a particularly good matchup. Sure they overlap in some ways but then almost everyone in that part of the market does.
Hardware-wise they've very little in common. Where SGI does have 'kewl stuff' it's not particularly applicable towards Apple's current hardware base (that I'm aware of.)
Customers, contracts, service & support, development engineers - SGI is hemmoraging all of those and again, none of them are very good matches for Apple's business model.
What Apple needs now is to improve it's hardware and solidify MacOS X.
In hardware Apple needs faster, more flexible motherboards. Faster CPU's would be great too but they can get away with dual (or more) PowerPC's as long as they're competively priced. Should PowerPC truly be stuck at a roadblock Apple could consider moving to Alpha. In no case are they going to succeed (ie profit) competing in the consumer or workstation x86 market - at least not in the near-future. Once the transition to MacOS X is complete and it's established Apple would do well to consider a Darwin-based x86 subset specialized for server duty, much like the MacOS X Server release of two years ago is used today but on a x86 hardware platform. I expect they'd treat it much like they do their '486-based AirPort today - the CPU is x86 but they don't make a big deal of it.
OS X needs to get out in the market and establish itself as a Unix peer. Once it gets some creditability as a stable Unix vendors can start porting over Unix-apps to it and customers can start weaning off of the Classic/Carbon codebase. With this transition Apple could begin to move into some of the server market making the same arguments MS makes for NT: one OS scalable for desktop / workstation / server. They'd be able to come up to speed rather quickly in the server space as the developers would be simply performing another Unix port with some added UI work.
Were I Apple i'd be looking to spend my money on a few things:
I don't read ACs: If a post isn't worth so much as a nom de plume to its author then I wont bother either.
Damn - posted instead of previewing. Well, there it is, warts & all - you get the idea. Honestly I'm not terribly interested in pursuing this discussion much futher anyway. Apple buying SGI seems extremely unlikely IMHO and 'what ifs' aren't particularly engaging.
I don't read ACs: If a post isn't worth so much as a nom de plume to its author then I wont bother either.
This is all well and good, but this program costs quite a bit of money, and for, say, educational and other use, I think it'd be really nice for a free X Server to be available for OS X. Does anyone know if there are any efforts out there to create one? (Afaik, John Carmack's project is for a complete X environment on top of Darwin, and not one that meshes with Aqua, which would be infinitely more useful.)
Where, exactly, have I or anyone stated that X11 is without flaws? As I've said in another post, it has its share of warts but that's just the way life is. But the fact remains that it still just a protocol that anybody can implement with varying degrees of success. Perhaps one day a better windowing protocol (Berlin?) will supplant it like HTTP did to Gopher, but that day isn't here yet and the work continues.
And even when something better comes along, X11 may persist just like Gopher sites can still be viewed with web browsers. Network transparency guarantees non-exclusivity, and that's also a Good Thing.
Ita erat quando hic adveni.
As I recall, NeWS and Display Postscript were not open standards. That, more than anything, likely accounts for why neither is around any longer. Berlin is an open standard, but I'm discouraged by the slowness it's progressing and the lack of running client apps. A better "Kept Simple Stupid", network transparent, open, windowing environment would be a good thing to have, but it would have to be substantially better than X11 (probably not too difficult) and easier to use (both user-side and developer-side) in order to catch on. Unfortunately, we're not there yet.
Ita erat quando hic adveni.
HTTP has been improved. So has X11. When the need calls for it, protocols get improved. If HTTP really were in need of dire replacement, there'd be a call for FOOTP to replace it. As of yet, I haven't seen such a desire.
Nobody's saying X11 is perfect. But it's here now, has *lots* of clients and it does exactly what it's supposed to. And since it need not replace an existing GUI, there's no reason not to implement it and enjoy X11's benefits on top of an existing GUI scheme.
Perhaps Berlin will do an even better job, but it's not here yet and remains unproven. And in the meantime, X11 still works. To me, it seems like most of the "X11 is bad!" cries are from people who don't understand what it is and when benefits it has over the current schemes, which is unfortunate because neither Microsoft or Macintosh offers some of its greatest benefits. I'd hate to see the standard free graphical display replaced by something prettier but less featureful until the benefits of network-transparent GUIs are rediscovered all over again.
Ita erat quando hic adveni.
Tenon wants $199 for the product during the beta phase, and apparently plans on increasing the cost significantly thereafter. (The $199 will give you the release version at no extra charge if you pay when receiving the beta).
Now, $200 is a healthy chunk of money. Even at OEM discounts, I don't think it's going to be attractive to port many applications relying on this type of support. So I don't think Apple has anything to worry about there.
The big question is "When is carbonized Photoshop arriving?" - because until we get it, I think there will be at least some doubt surrounding the platform.
Incidentally, I'm posting this from my Mac running OS/XBeta. If it weren't for the keyboard dropping characters all the time, I'd say it's darn near perfect other than non-support of video editing applications.
D
----
As one can imagine, this is both good news and bad news for Apple They undoubtedly welcome this product as it gives the connectivity they so sorely need; particularly in their important scientific research market where they're already fairly popular. On the other hand there is likely a fear that developers might 'port' to MacOS X by simply running their apps in a X term session (possibly a cut-down server licensed from Tenon) and not go native.
As I've already said and further explained this would dovetail nicely with an Apple purchase of Silicon Graphics, who already have a very strong presence (albeit slipping somewhat) in the high-end scientific/visualization environment that even the glorious G4 hardware cannot compete in.
Mac interface zealotory aside, the ability to more-or-less seamlessly integrate MacOS & Unix X applications would make such a marriage an ideal thing -- high performance SGI enterprise-grade hardware for bigger jobs, iMacs or towers on desktops for local jobs and talking to the larger-scale machines.
I think Apple should rush to support this. I agree they might be a little scared of people just running Mac apps over X and not going native, but without a high-end server lineup in their product mix this is going to happen anyway.
So I don't have to fire up VNC to get a remote display for my linux box when I work on my powerbook. I don't even have a monitor on my linux box, it's tucked away in a closet, and if I ever want a GUI there, I use VNC. It would be nice to just be able to export my display to my powerbook...
>. To me, it seems like most of the "X11 is bad!" >cries are from people who don't understand what >it is and when benefits it has over the current >schemes, which is unfortunate because neither >Microsoft or Macintosh offers some of its >greatest benefits Don't get me wrong.... no 'X11 is bad!' cries here. X lets me get a lot done..... BUT at the same time i refuse to agree with the standard ./ mantra that if you have X (specifically Xfree86) then life is complete.
Nonsense. X has some huge holes in it, or at least limitations due to the era in which it was conceived. I don't advocate just throwing it away immediately, but the status quo isn't something that interests me.
>And since X is a protocol, it is neither slow,
>buggy or holding anything back any more than
>HTTP is slow, buggy or holding the web back. So
>can we please stop with the "X is bad" rhetoric?
huh? So a protocol in and of itself can't be good or bad? Come on.... (that is why HTTP *has* been improved, persistent connections, etc, because as things scale it *does* hold back the web)
And besides... this isn't 1975 anymore. I think we can do better than blasting bitmaps across the network using a broken security scheme.
(I'm not trying to upsell MacOSX by any means... but lets quit worshipping at the altar of X11. I mean, it gets the job done {mostly}, but it is LONG overdue for a replacement.)
j
Or does this seem a bit silly to port Xfree86 to OS X?
XFree86 was ported to Darwin, not Mac OS X. The difference is that Darwin had no graphics layer controlling the screen, much like Linux itself has no graphics layer controlling the screen.
Tenon's solution is very different. It goes through Mac OS X's drawing APIs, while XFree86 controls the screen directly. The XFree86 port was to give people running Darwin, and not Mac OS X, a GUI (of sorts).
Geez. This is only brought up in every single discussion thread about Mac OS X or Darwin since Carmack announced his intention to port XFree86 over a year ago.
Apple did a lot of work to get their X server working...
This X server is not made by Apple. It's made by Tenon. Also, Apple's current Display PDF system is not the same as an X server, if that's what you meant.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
X11 is a protocol
Why, yes. X11 is a protocol. So is Windows' SMB protocol.
Being a protocol doesn't mean that it isn't inherently flawed. X11 and SMB are very much alike in the fact that they are a series of hacks, forced inconsistencies, and special cases for backwards compatibility. They are both dangerous to run in kernel mode because there is no clean way to implement a broken standard. You may eventually reach stability, but there will still be occasional "gotchas," and your code will never be clean and well designed.
If you don't believe me, try actually reading some of the ICCCM standard sometime.
Try reading some of the sections of the Unix-Hater's handbook dedicated to X, like this one.
If you honestly think that just because X11 is a protocol that it's not inherently flawed, then you obviously have never tried to implement someone else's protcol before. There is a major difference between implementing something clean and simple and something that is a complex mass of special cases. Yes, you still want X for a Mac, just the same as you want SAMBA for Unix boxes, but don't go thinking it's not a flawed system. X is an albatross hanging about the neck of the Unix world. It's a great example of how superior products can be crushed by flawed but free alternatives.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Yes, there is an overhead. All X applications still have to formulate their drawing commands into X drawing commands and pipe them through the high-speed local socket. The local X server then has to read the command from the input, decode and interpret them, and then call its API calls to draw them.
The alternative is to have the application go straight to the graphics APIs and do it itself. There may be no network slowdown, but the above is a major slowdown. Even DPS had a slight slowdown on the local machine compared to straight rendering. Personally, I hope GNUStep takes off too. It would be good to see the UNIX world with a free, well written, exportable graphics layer.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Where, exactly, have I or anyone stated that X11 is without flaws?
"And since X is a protocol, it is neither slow, buggy or holding anything back any more than HTTP is slow, buggy or holding the web back. So can we please stop with the 'X is bad' rhetoric?"
That sounds an awful lot like you're saying X is good, which it blatantly isn't. X is bad. It's incredibly bad.
Perhaps one day a better windowing protocol (Berlin?) will supplant it like HTTP did to Gopher, but that day isn't here yet and the work continues.
What, you mean like NeWS or Display Postscript? Both lost badly to X in the days before it became a standard. Superior code does not equal inevitable domination. Just look at Windows. Just look at X. I really hope the Berlin project takes off, but don't hold your breath. X has shown an amazing ability to crush better written competitors because of its wider adoption and better availability. We'll see.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Sure, as long as you don't want to run Mac apps and Unix apps at the same time. Might as well dual boot into Linux at that point.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
Duh. Why pay when you can get the crappy x-window interface for free?
You can't run it at the same time as the Quartz graphics layer. This means you can't run Carbon, Cocoa, and Classic Mac OS X apps at the same time as X apps. You might as well dual-boot into Linux at that point. What's the point of having Mac OS X at that point?
Duh. <g>
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
And since X is a protocol, it is neither slow, buggy or holding anything back any more than HTTP is slow, buggy or holding the web back.
Actually, some people do believe this to be true. Who says a protocol can't be bad? And who says there isn't room for a better protocol than X?
- Scott
------
Scott Stevenson
Scott Stevenson
Tree House Ideas
Apple is probably going to spend a fair bit of money on the theoretical optimizations for the Mach microkernel that can (theoretically) make it faster. But, as far as the public knows, Linux holds the speed record right now.
Mac OS X public beta is considerably faster than even DP4, both of which run Mach 3.0. OSX Server has Mach 2.5 and display postscript, amongst many other differences. Mac OS X public beta and Mac OS X Server can't really be reasonably compared in terms of performance.
No doubt that LinuxPPC is fast, though.
- Scott
------
Scott Stevenson
Scott Stevenson
Tree House Ideas
Actually, all that good scripting power is already there in the default install. Well, at least TCL is for sure, I beleive Perl is as well.
Perl, TCL, Java, tcsh, all preinstalled. Along with OpenSSH, OpenSSL, Java2, WebDAV and the newset version of Apache.
- Scott
------
Scott Stevenson
Scott Stevenson
Tree House Ideas
I have Tomcat running under OS 9
But Mac OS 9 doesn't hava Java 2 support, which is required for JSP 1.0+. How'd you do it?
- Scott
------
Scott Stevenson
Scott Stevenson
Tree House Ideas
And why the hell can't Apple ever name anything that gives you an idea of it's purpose instead of giving it some cute nickname?
Like GNOME?
I'd rather have a cute nickname than a widly out-of-control buzzword fest like:
Java 2 and Objective-C API Set for Mac OS X
...
Pro Enterprise Edition.
- Scott
------
Scott Stevenson
Scott Stevenson
Tree House Ideas
Is webdav there? How do you turn it on?
/Library/WebServer/Configuration/apache.conf:
/System/Library/Apache/Modules/libdav.so
Apparently, you can mount WebDAV volumes on the desktop by just entering the address (http://webdav.company.com) in to the "Connect to server" dialog under the Go menu in the Finder. As far as turning on your own WebDAV server, I suppose you can turn it on by just uncommenting these two lines in
#LoadModule dav_module
#AddModule mod_dav.c
Save the file, restart Apache. I haven't tried it, but assume it will work. All the appropriate files are there.
- Scott
------
Scott Stevenson
Scott Stevenson
Tree House Ideas
Why not the Macintosh (OS X) Foundation Classes ?
;)
MFC is already taken.
- Scott
------
Scott Stevenson
Scott Stevenson
Tree House Ideas
However, only X (to my knowledge) allows you to *display* applications that are being run on another networked machine. The distinction is that, while the application is running on my Intel-based system, it can be displayed on my Macintosh machine 1000 miles away. The benefits for remote-administration are substantial.
-----------
"You can't shake the Devil's hand and say you're only kidding."
You could always just run straight Darwin (by typing ">console" at the login window) and install XFree on top of that. Darwin binaries of XFree are at www.stepwise.com/Softrak, and FVWM2 and AfterStep have been ported to Darwin. Of course, you wouldn't be running Aqua underneath, but it's still cool, and it's free.
I think it's only a matter of time till there is a free X Window Server for Aqua.
The latest clients from distributed.net have Altivec support in both the OS X and OS 8/9 versions. 3.4 Mkeys/sec on my G4/400.
How to solve most of our problems: 1.Lots of nuclear plants. 2.Cure aging.
Take a look at the info about iTools. Tenon has claimed to have ported Tomcat to OS X. The funny thing is that Tomcat has been running on OS X since it had Java support. Also consider how odd it is to have "ported" a Java program to another platform. I have Tomcat running under OS 9 and I also had it running under DP 4 back in June. Now I'm sure that Tenon has created a nicer interface to Tomcat, but to call it a "port" is wrong. Does anyone know if this violates the license that Tomcat is released under?
The majority of Mac users don't give a damn about the techy details slashdotters hold so near and dear - they just want to get their email, surf the web, print pictures of their newest nephew, and maybe edit a newsletter or two.
:P
A few of us also actually make a living with our Macs. But we do all those other things too.
I imagine that the people disgusted enough with Aqua to even consider the kind of thing proposed by the original commentator will instead hold onto OS 9 until they drown with it. It won't be pretty, but so far it looks like they will be few and far between.
Mac OS X already has its own GUI framework, and there is absolutely no way that X could improve upon that. After all, the Mac GUI has always been its main selling point, whereas X is roundly, and deservedly, slated as being slow, buggy and holding back Linux on the desktop.
I honestly don't see the point in this. Surely it's not progress, it's going backwards?
This one article has caused the letter 'X' to move from the 24th most frequently used letter (in the English language) to the 22nd spot, surpassing 'J' and 'K'. If this keeps up, Huffman encoding will have to be completely re-written.
Admit nothing, deny everything and make counter-accusations.
I have avoided a flame such as this for as long as I reasonably can, because I'm well aware that you are all human beings, that geeks are stereotypically bad in the grammar department, and it's generally not that big of a deal anyhow. However, the headline:
XFree86 And Apache Or Mac OS X Beta
is so wildly different in meaning than the proper headline of:
XFree86 And Apache On Mac OS X Beta
that it merits a bit of a slap on the back of the wrist.
You, the /. Editors, run a much appreciated and much used online web service, and I am extremely glad that /. exists. I do not deny that you all have a whole lot of stuff that you need to tackle, and I do not think for one second that you are anything less than intelligent, inspired and driven individuals. However, the fact that single paragraph, front page articles regularly have the most basic of typographical and grammatical errors is cause for some embarassment, IMHO. Slashdot is recognized as one of, if not the premiere online geek news source. It's readership has grown to the point that only the strongest of web clusters can survive a fully-blown Slashdot Effect. Major international news sources (AP, Reuters) have begun referencing quotes and threads from Slashdot. Slashdot is now in the Big League(tm). Please, please put in that extra three miunutes of reading, re-reading and re-re-reading front page stories for typographical errors. Think of them as bugs in your code, and each front page post as a final version release.
I know that you're proud of Slashdot, and you should be. It's an amazing website. But please, take the next step and start making the content delivery a bit more professional. It's easy to do, and it'll help make Slashdot a better news source.
Trolls: If you must, go ahead and attack this post for typos and grammatical errors. I'm still in the process of learning this danged AZERTY keyboard, and there is in fact a difference between an editor-approved front page article and a comment in the attached thread.
Obliteracy: Words with explosions
Apple's MacOS X does not come with an X Windows server. Apple has no intention of developing or releasing such a beast. Apple instead chose to develop a PDF-based display system called "Quartz" upon which runs their UI named "Aqua".
Tenon, a respected developer of Mac software, has developed an X Server that runs on MacOS X. This is significantly different from John Carmack's port of X to Apple's open-source OS Darwin (upon which MacOS X is based.) Tenon's X Server is driving the Quartz PDF-based display and utilizes the Aqua toolkits to produce a fully MacOS X-integrated display. In short anything run under Tenon's X Server is immediately available to the rest of the MacOS environment as just another PDF (cut-'n-paste, etc.) as well as appearing as Aqua-like as possible.
Thus with Tenon's X Server one can run a generic X application and it will appear as simply another Mac OS X application obeying as many of the Aqua UI principals as possible. Indeed with many X applications trivial changes will be required to make them appear to be native Mac OS X applications (mostly menu placements, dialogue boxes, etc.)
As one can imagine, this is both good news and bad news for Apple They undoubtedly welcome this product as it gives the connectivity they so sorely need; particularly in their important scientific research market where they're already fairly popular. On the other hand there is likely a fear that developers might 'port' to MacOS X by simply running their apps in a X term session (possibly a cut-down server licensed from Tenon) and not go native.
As to Tenon's other 'ports' - are they ports? Likely yes. Getting a unix application, particularly one already native to BSD to run under Mac OS X isn't much effort, if any. Constructing a native user interface for it, while not very difficult, does take time and some (presumably) skill. Since we already talk of ports to other such brother/sister OS's the same would seem to apply here. The argument is strengthened with the added effort of creating an Aqua UI and integrating other Mac OS X conventions which Tenon has done.
-- Michael
No affiliation with any Mac developer currently, nor have I ever purchased any Tenon product. However I do look forward to running their OS X Server.
I don't read ACs: If a post isn't worth so much as a nom de plume to its author then I wont bother either.