Earlier I wrote: Note to parent: do your research before jumping to conclusions and making false claims, it helps prevent you from looking silly. I know. I've learned this the hard way myself...
Damn. You can never do too much research, I guess.
Anyway, before anyone else points this out, I'm going to have to point out that what you can buy at the Apple Store is Version 5.2. What the story is about ( and what ships with Tiger and XServes ) is version 5.3. The interesting part is that both are available, I guess? What does that imply??
From what I can tell, though, this all doesn't look like even 5.3 stops you from deploying a.war on Apache using mod_webobjects on any ol' platform you can find mod_webobjects for.
I'm going to shut up now, though. I'm not a WO expert, I just noticed 5.2 still at the Apple Store. I'm just as curious as everyone else to find out what Apple's plans are. If anyone can explain what's going on, please do. I'm not sure this means 'goodbye, WebObjects', though.
To me, this looks more like "Apple includes $699 enterprise development product with every XServe"... am I reading this wrong somehow? How is this bad again, I'm confused...
It is true, though, that the Tech Specs on the WebObjects 5.3 page only list OS X Server as the deployment environment, what the heck is up with that?
Is there any official Apple word on what the thinking is here moving forward, how committed they are to the Windows2000 and Solaris versions ( well, I guess since MS isn't supporting Win2k much... hmmm ) ? Not providing WO cross-platform does seem to call into question the purpose of having migrated to Java and having dropped Objective-C. Are they going to bring back WO Objective-C ? What the heck are they doing?
I guess the important question is how long the 5.2 version will be sold and supported, and how compatable 5.3 is with 5.2 and popular.war/mod_webobjects deployment methods. I'm not sure I see how they could easily _stop_ that from working, frankly, short of going back to native code or introducing some forced incompatability. Odd.
Sorry about the caps, just doing that to point out the icky use of caps by the incorrect-information-offering parent post. I'm also sorry if I appear combative, but incorrect information in a Slashdot post is almost worse than a troll, really. At least the average reader knows a troll when they see it. You have to look to know the parent is wrong on at least two counts.
Here's the truth: the article should read "Apple gives away $699 software package with every copy of OS X Server!"
Development platforms:
Mac OS X v10.2.2
Windows 2000 Professional SP3
Deployment platforms:
Mac OS X Server v10.2.2
Windows 2000 Server SP3
Solaris 8
just as it's been for some time. The only new thing is that the developer tools are free ( for OS X ) and the entire package is free ( for new OS X Server purchases ). Now it only costs money ( exluding developer time, of course ) to develop and deploy WebObjects if you want to do so entirely on Windows 2000, or if you want to avoid buying an XServe. This is actually a brilliant move by Apple, although it is one likely triggered in part by low sales due to increased competition from J2EE, LAMP, and.Net ( and probably other ) solutions.
Note to parent: do your research before jumping to conclusions and making false claims, it helps prevent you from looking silly. I know. I've learned this the hard way myself...
Pretty much. A fancy text editor. Where word falls flat on it's face is if you want to do things with graphics, or advanced multi-column newspaper-style layout, where different columns are different heights and widths. Page layout, like you said, is a problem with Word. If you just want text, paragraph layout, that kind of thing, it's about as feature-rich as you could ask for, if a bit difficult to use for all of the features.
What does a word processor do that Pages does not?
Pages is not your typical word processor. In many ways, it's no threat to the dominance of Microsoft Word. For instance, Pages isn't for you if:
You need a form letter to send to hundreds of contacts, with each contact's name and address substituted into the letter.
You often need to count the number of words in a selection of text (Pages will only give you full-document totals);
You have multiple users updating documents and need the ability to track the changes that each makes;
You're an advanced user who relies on macros to automate your word processing tasks.
But if you want a GPU that's targeted to supporting DirectX, I don't know if using it for a DSP is really the right idea. Maybe it is. Or maybe Intel's own vector processing capabilities could improve. Or maybe you could get an extra board and actually make use of those empty, fast PCI slots. Your call. TFA does a good job of pointing out that GPUs aren't exactly 'general purpose', and that's why the article contains:
why aren't we seeing it in the PC space yet?
Xenos's particular range of features are going into a closed box environment, hence the API can be tailored to expose all of the features of the chip, however on the PC space graphics processors really need to be tailored to the capabilities of the current DirectX release. This is where Xenos has an issue in that its features and capabilities are clearly beyond the current Shader Model 3.0 DirectX9 specification while it lacks features that are expected to be a requirement for WGF2.0.
So, what's the story even talking about here? Something that's possible except for Microsoft? Or something that's possible, in a completely different form? What's the point they're trying for?
So does this mean that apple can rewrite vecLib to take advantage of future vector processors?
This has nothing to do with that. Although, they can and are. They've been calling it the "Accelerate Framework" and it's been available ( well, most of it, I guess ) since 10.3. But we're not talking about a vecLib re-write. If you use vec_foo, you'll need to look for _mm_foo and re-write your own code. But vDSP, vImage, and other higher-level functions outlined on the linked documentation page, those should 'just work', as long as there's supported platform to target, as likely will be the case for Intel chips moving forward...
Altivec-optimized code is going to be the single largest issue for programmers in the switch to Intel. Meaning, it's actually going to be a pretty darn smooth transition for most.
True, but before "Pages" there was the ugly beast called "AppleWorks"... which clearly couldn't compete with MS Word.
They still have AppleWorks. I think it even still ships with every Mac. Hey, check it out, can it really run on Windows?? It appears it can.
It's definitely still useful, though it's rudimentary spreadsheet is probably the weakest link, it's Carbon of course, and badly needs an update... although, now that I mention it, it looks like it has actually bumped a few version numbers since I last looked- interesting, huh?!? It does seem to be in fairly active development for something we'd all written off.
Pages doesn't really replace a word processor, I don't think you'd use it to write a report or something, it's really geared towards making a newsletter with ( somewhat ) fancy graphics or something. It's more of a niche app, like a end-user Illustrator or something.
No, AppleWorks doesn't have half the features of word. Then again, do you use half the features of Word ? It occupies that niche for folks who aren't going to pay for Office. It's $79 new, and though I doubt they sell a lot of copies that way, it's still a hell of a lot cheaper than Office.
Of course, it's entirely possible that Numbers is something different/more than a spreadsheet. Maybe it's a student-version Mathematics package. Maybe it's just a common word Apple thought they could snap up. We won't know until a product is shipped.
As to am I trolling or just uninformed? I an uninformed. It is good to learn.
I thought so. For whatever odd reason, I like to inform. Probably becasuse I like to learn, too, I guess.
Currently we are using MFC as the frame work.... And IT SUCKS to be blunt....We offer our software bundled with notebook computers. We currently sell IBM Thinkpads....
Seems like a good time to find a new vendor.
Ouch and double ouch.
You ask what is different now? Well when Apple moved from 68k to PPC the emulated 68k programs actually ran faster then the native machines that they replaced. Also there was no investment in AltaVec yet. There is no way that a P4 is going to emulate a ppc as fast as a G5 at least for a few years. I feel that is really going to hurt Apple.
It would hurt Apple, IF they hadn't made it so easy to recompile PPC code to x86. I say 'recompile', not 'port', because ( unless you're dealing with endian issues, hand-optimized Altivec, or a very, very small host of other issues ), it really is just a recompile. NeXTStep had cross-compiled 'fat' binaries a decade ago, and Apple's just leveraging that existing technology. QT/C++ for OS X Intel is likely to be no different than QT/C++ for OS X PPC. Most people won't be running emulated code. For those that will... reports are Rosetta ( the emulation engine ) doesn't run on those 3.6Ghz P4 systems at G5 speeds, but they do run faster than 800Mhz G4 speeds... which is pretty damn impressive. I'm still using my 800Mhz G4 iMac at home, and I never think "damn this is slow". I think maybe Rosetta pulls some Cocoa-lib-related tricks, so a C++/QT combo might not find *quite* the same speed ( I'm just speculating, it could be the same ), but still... for some integer-op-heavy code, based on initial reports, a dual-core intel chip running emulated code might easily outperform a 1.8Ghz G5 as found in most iMacs, running that same native code. Most importantly, those apps will be fast enough to use. Also, keep in mind, Apple will be using Intel chips released over the next two years- few are predicting that we'll see P4s in anything but these development machines we currently have.
Better for developers, many MacIntel users won't realize this, or won't be upgrading from G5s, or will want the extra performance anyway, so they'll be inclined to buy new code from us developer types. Code which we'll likely be able to provide as a 'cheap' upgrade, because, well... it took 2 hours to 'update'. It's an upgrade cycle that I think will actually be a boon to many OS X software merchants...
Freescale is not making the e600 in quantity yet, and I doubt Post-6/6 Apple will design a new MB for a chip they're only going to be shipping for a year or two anyway.
They may have already done a design, but you make a good point... especially since the low-end and mobile machines will be the first to be Intelized, it's not likely.
Maybe Apple can negotiate an anti-competitive exclusive OS provider contract with Dell... I guess, since they're not a convicted monopolist, they could actually do that... but why don't we just talk about Microsoft going bankrupt and Bill Gates living in a van down by the river, as long as we're talking about things that just aren't going to happen ?
I mean, really, it's about as likely as Apple using Intel chips for its... oh, wait...
What is worse is Apples switch has put the effort to port on hold.
What is the rationale for that ? There isn't really much of anything development-wise that's really affected terribly by the hardware switch. Especially if you're using C++ and QT... what's changed now that Apple's announced a slow, 2-year migration to a different hardware platform that would keep you from doing a port? The port you're talking about shouldn't be that hard.
Frankly I am afraid of the Osborne effect killing Apple.
I likely can't convince you of it now, but that effect is way overblown, and Apple has weathered such a change *twice* before, and is in a stronger position now than it was either with the 68k/PowerPC switch or the OS9/OS X transition. I don't see why this should be sucha concern- people who were going to hold off on buying new hardware were going to hold off anyway; people who need a new OS X-running machine are still buying them, knowing that their machines will be outdated soon. Steve Ballmer was actually right on this one: "What's different?"
So I will most likely be stuck using C++. Now if Troll tech would come out with Objective-C bindings and Intel would come put an Objective-C compiler for the Mac and for Windows.....
So how long before the Intel Compiler is available for Mac OS/X?
Are you trolling me, or just uninformed? Sorry... I just have to ask, because none of the things you're asking about are really needed, AFAIK. You can mix Objective-C and C++ code if you feel you must, so an Objective-C/QT binding is something that, well, you can probably work out a controller layer for what you need without too much effort, but you shouldn't need to if your code is already in C++. Leave it there. The GNU compiler ( which is what's used on OS X ) compiles C++ just fine, and there's plenty of support and documentation for using C++ in OS X. Cocoa is only *one* of many ways to write an app for OS X. It's the way that provides the cleanest user experience, but if you want to port your QT/C++ app to OS X, your limitations are likely going to be from QT rather than any other part of OS X.
The announced plan to use the Intel compiler is really just an optimization; they're building Intel-compatable OS X programs already without it. Actually, the compiler itself isn't likely to change much... maybe Apple will help Intel throw in some Objective-C magic if it's not already there, but that's not all that much, really, it's portable C code and a fairly well-known quantity.
The bottom line for you is that what you need to do is already easy, so much so that Apple's usually extensive documentation boils down to just a page for QT. I'm puzzled as to why you don't already have an OS X port. If you have a QT app that already works under Windows and KDE... it should really 'just work' under OS X, with not much effort at all. I mean, that sort of portability is why you're paying TrollTech the big bucks, right??
hey, I wasn't trying to slam you, just correct your very-nearly-correct information. You nearly had it right. I had to look it up.
The NeXT/OS/X stack is very flexible but I am not so sure about Darwin/Mach.
Interesting distinction, isn't it ? The fact that the Foundation and Application libraries of the Cocoa/OpenStep/GNUStep stack are so easily separated from the underlying 'kernel' is part of what will keep it possible to move forward as unforeseen changes occur. The difference between the NeXTStep Mach underpinnings and the Darwin core are already terribly significant. Darwin is more BSD than Mach.
I have heard that it has poor threading performance which in the new multi core world is going to hurt.
Specifically, you've heard that thread initialization is expensive. After that, the threading performance is actually very good. Proper thread pooling techniques utilize the system quite well. I've not studied how the current OS X core fares on multiple-CPU systems as compared to other modern OSs, but I've heard it does quite well in such comparisons. Real-world, full multiprocessor utilization on dual-CPU OS X systems is very common.
The big question now is do I learn objective c or stick with c++:)
If you already know C++, it won't take you more than a couple of weeks to pick up the basics of Objective-C. It'll shock you what a simple, clean syntax it has. What might take more time is becoming familiar with the AppKit and Foundation APIs and more 'advanced' time-saving features like bindings and the Document architecture, and a few related tools ( like Interface Builder ). But if you just want to do a quick application development project on OS X, by all means... you shouldn't find learning Objective-C to be much of an issue at all. Pick up "Cocoa Programming for Mac OS X" by Aaron Hillegass. I don't *really* agree with his assessment that "mastering [Objective-C] will take two hours if you already know C and an object-oriented programming language like Java or C++", but if you are any kind of programmer, the language is not an obstacle, it's a blessing.
The only real problem might be that you learn Objective-C and start getting mad at issues in C++ that you didn't even notice before.
If there's no significant forseeable profit in doing so, however, why would they bother ?
Don't a good number of companies give what are essentially specs ( possibly bundled with a few easily reproduced tools ) under the guise of a licensed developer program or some such?
Um, yea. I think NeXT was not *quite* there yet by 1985. Almost, not quite. At least not publicly. From the wikipedia entry:
NeXTSTEP 1.0 was released on 18 September 1989 after several previews starting in 1986, and the last release 3.3 in early 1995, by which time it ran not only on Motorola 68000 family processors (specifically the original black boxes), but also generic IBM compatible x86/Intel, Sun SPARC, and HP PA-RISC.
It's really not difficult to imagine at least *some* NeXTStep/OS X dirivative lasting another 20 years... it's a remarkably flexible software stack, and it's already been shown that swapping out different bits of that stack is quite doable...
Me, I miss CP/M... I haven't seen it since 6th grade...
Get a small air conditioner for about $129. That way, you won't spend an afternoon bending tubes, you won't spend lots of time carrying ice from your freezer, and you won't end up using your freezer as a heater.
In the long run, you'll probably end up cooler and save money on your electric bill. Seriously...
This story is just here to make us all go "OMG! ROTFL! What an idiot!", because the talks about his knowledge of thermodynamics, then uses ice water from a fridge *in his house* to cool that same house, right?
Or is it just some elaborate troll meant to make Canadians look stupid ? To his credit, it does cool the spot near his 'heat pump'. You can actually buy similar devices here in the states at your local hardware store ( put ice in the top, a fan blows over rotating fabric band which melting ice coats w/ cool water ), so it's not *quite* as dumb as it sounds. Well. Ok. It is, actually, unless you're just interested in cooling a single spot, and not interested in doing so efficiently. In which case, uh... small air conditioners are in fact not all that expensive.
Why, I guess it is. I couldn't remember the name, honestly, just the latex outfit. If you look, you'll notice that there is at least one image of Ceren on the page I referenced above, though.
Now I can only question why you'd choose an image without said red latex suit... I'm going to guess you're just showing off/trolling. Typical BSD geek behavior;-)
So no, I didn't know her by name, nor did I recognize her without the outfit. You're clearly a much bigger BSD geek than I. I bow before you in shame.
But really, how famous do you expect the 'most famous BSD chick' to be ?
On the other hand, BSD is cool because it has a hot chick.
I mean, you've got to be able to come up with a better BSD daemon girl than that without even trying. What, is that your girlfriend or something? Pathetic.
Honestly, doing a google search didn't give me _just_ the image I wanted, but there are some pretty impressive examples in this collection, even if what is perhaps the best one is animated. ( Warning: not entirely work-safe, *and* contains flamefest-inducing images of penguins impaled on pitchforks ). You've been warned, now let's see that server melt...
IBM has announced a dual-core G5, and Freescale this nice new laptop-friendly chip. These don't seem like the old cold shoulder to me.
Announced. Shipping in quantity? The e600 is dual-core, but not at any higher clock speed. I'd be curious if they can ship it in quantities Apple needs at this point.
Yea, Apple supposedly has the dual-core G5 chip in testing. I'm willing to bet they'll actually ship some of those on the high-end, actually, this is a two-year plan after all. I'm much less sure about the e600, though, unless Freescale is already making them in quantity.
What's really weird is that the linked article only mentions zealots in passing, almost.
And to answer the story's question ( not the article's ), zealots and evangelists can both hurt and help the thing they're trying to support. A more balanced, rational approach is always more likely to convert the unwilling. The more zealous approach is always more likely to convert the willing.
Either approach is likely to make someone wonder what your motives are, and make them want to defend the choice they've made. I think that's the real mistake; you shouldn't be making someone feel stupid because they chose Windows, you should be asking if they've had problems because of that choice, and helpfully offering alternatives that you can show solve those problems... same goes with political parties, religion, or any other product you'd like to get others to support.
1. Software vendors will abandon support for PPC machines as soon as it makes business sense to do so.
Of course, but here's the thing : in the very, very, very vast majority of cases, it won't make sense. Comparing the situation to NeXT is a bit silly ( unless you'd like to compare numbers of installed users, see my point yet? ). If someone decides to stop supporting PPC, it means their product has somehow become Intel-specific ( how? ) and they have a small PPC install base, _and_ don't want the existing PPC customers. That should be pretty rare, really. I'd like to see the numbers of NeXT vendors that dropped 68k support and kept Intel support, anyway; I don't think that number is as high as this guy claims.
2. Software vendors will charge you money for Intel (or PPC) versions of software when it makes business sense to do so.
They'll happen either as upgrades or as new versions of software. You'd pay for those anyway. Yes, people buying Intel macs and wanting native performance will end up buying lots of new software. Duh. Guess what happens with every OS switch? Of course, if performance is acceptable, you can probably put off that software purchase for quite a while... initial reports show Rosetta getting pretty decent performance, so this could be less of an issue. Even if it is, is it a reason to not buy a Mac today? It sounds like more of a reason to not buy an Intel Mac if you already own a PPC Mac.
3. Apple has shown it drops support for old products regularly because it makes business sense to do so.
Like every other company on the planet? Yet, with the tools in place to make builds for PPC a single checkbox away, it should be some time before simply using that check box doesn't make business sense. Let's see... when exactly did NeXT stop supporting 68k hardware, since that's what he's comparing this to? Was there ever an NeXTStep for Intel release without a matching 68k release? Ok, I know... there weren't many of the first, but it was pretty easy for them to keep 68k support, and they did.
4. Macs tended to have a far longer life-span than the average PC.
Yea... and this changes how ? Right now, a dual G5 is right up there with a top-of-the-line PC ( well, except maybe for the graphics card, maybe ) and will still compare to that same PC... his prediction is that in 2 years today's PCs will be obsolete? Wow. Stunning.
5. The new Intel machines promise to be much faster than current machines.
Huh. That's the reason for the switch right there in a nutshell, isn't it?
6. People do not buy computers only for how it will serve them today, but for how capable it will be in serving them for their desired term of use.
uh... as compared to _other_ computers available to buy _today_. If I _can_ put off buying a computer, I'm going to; tomorrow will always provide faster computers at lower prices. If I can't ( say, I'm a student starting college this fall... or a business which just hired a new employee... or I just need a computer to handle my digital photos ), well, I'm likely to buy one now rather than wait a year. Better ones will be here in the future, but that's not terribly relevant today. If I want to run OS X, this doesn't deter me from buying a PPC mac unless I can wait anyway. In which case, I was going to maybe wait anyway for a year. For that relatively small group of users that can wait for a year, sure, Apple will parlay those users into pent-up demand for their first Intel machines. That's somehow bad for Apple?
7. Potentially no Classic support.
I thought he was looking for reasons why people won't buy more PPCs? If you want Classic support, you'll run out and stockpile PPC Macs. Believe me, though... few people want it. This publishing industry he's talking about don't represent that many computers... and it does represent a business that sets up systems and never, ever replaces the
And Apple may have overstated the Megahertz Myth, but not by much. They were basically correct, and I wish they'd remember that. Maybe they will, once they've finally killed off "Classic" Mac OS and completed the ten-year transition to NeXTstep.
That's very insightful. There's plenty of wiggle room in Apple's two-year transition plan for them to say "oh, hey, look at this fantastic new chip from Freescale... well, let's put it in a laptop and see who buys it". If it turns out to be a barn-burner in performance/watt, it's probably not too late. But does Apple really want to be making PPC and Intel versions of their product for a long time? Isn't that a little expensive and confusing for customers? I have to expect Freescale and IBM just said "we're more interested in these other, more profitable chip areas, Steve, sorry", and Steve said, "ok, I'll call Intel".
Slashdot Mirror
Note to parent: do your research before jumping to conclusions and making false claims, it helps prevent you from looking silly. I know. I've learned this the hard way myself...
Damn. You can never do too much research, I guess.
Anyway, before anyone else points this out, I'm going to have to point out that what you can buy at the Apple Store is Version 5.2. What the story is about ( and what ships with Tiger and XServes ) is version 5.3. The interesting part is that both are available, I guess? What does that imply??
From what I can tell, though, this all doesn't look like even 5.3 stops you from deploying a .war on Apache using mod_webobjects on any ol' platform you can find mod_webobjects for.
I'm going to shut up now, though. I'm not a WO expert, I just noticed 5.2 still at the Apple Store. I'm just as curious as everyone else to find out what Apple's plans are. If anyone can explain what's going on, please do. I'm not sure this means 'goodbye, WebObjects', though.
To me, this looks more like "Apple includes $699 enterprise development product with every XServe"... am I reading this wrong somehow? How is this bad again, I'm confused...
It is true, though, that the Tech Specs on the WebObjects 5.3 page only list OS X Server as the deployment environment, what the heck is up with that?
Is there any official Apple word on what the thinking is here moving forward, how committed they are to the Windows2000 and Solaris versions ( well, I guess since MS isn't supporting Win2k much... hmmm ) ? Not providing WO cross-platform does seem to call into question the purpose of having migrated to Java and having dropped Objective-C. Are they going to bring back WO Objective-C ? What the heck are they doing?
I guess the important question is how long the 5.2 version will be sold and supported, and how compatable 5.3 is with 5.2 and popular .war/mod_webobjects deployment methods. I'm not sure I see how they could easily _stop_ that from working, frankly, short of going back to native code or introducing some forced incompatability. Odd.
Here's the truth: the article should read "Apple gives away $699 software package with every copy of OS X Server!"
You can buy WebObjects from the Apple store just like always, and
just as it's been for some time. The only new thing is that the developer tools are free ( for OS X ) and the entire package is free ( for new OS X Server purchases ). Now it only costs money ( exluding developer time, of course ) to develop and deploy WebObjects if you want to do so entirely on Windows 2000, or if you want to avoid buying an XServe. This is actually a brilliant move by Apple, although it is one likely triggered in part by low sales due to increased competition from J2EE, LAMP, andNote to parent: do your research before jumping to conclusions and making false claims, it helps prevent you from looking silly. I know. I've learned this the hard way myself...
Pretty much. A fancy text editor. Where word falls flat on it's face is if you want to do things with graphics, or advanced multi-column newspaper-style layout, where different columns are different heights and widths. Page layout, like you said, is a problem with Word. If you just want text, paragraph layout, that kind of thing, it's about as feature-rich as you could ask for, if a bit difficult to use for all of the features.
What does a word processor do that Pages does not?
I'm going to let MacWorld handle that one :
Yea, occasionally I'd like one, but I'd rather it didn't take away from what my graphics card can do, wouldn't I ?
Like, maybe I'd like to have a generalized DSP chip like the one in my NeXT machine here, or one of these specialized DSP boards ?
But if you want a GPU that's targeted to supporting DirectX, I don't know if using it for a DSP is really the right idea. Maybe it is. Or maybe Intel's own vector processing capabilities could improve. Or maybe you could get an extra board and actually make use of those empty, fast PCI slots. Your call. TFA does a good job of pointing out that GPUs aren't exactly 'general purpose', and that's why the article contains :
So, what's the story even talking about here? Something that's possible except for Microsoft? Or something that's possible, in a completely different form? What's the point they're trying for?
This has nothing to do with that. Although, they can and are. They've been calling it the "Accelerate Framework" and it's been available ( well, most of it, I guess ) since 10.3. But we're not talking about a vecLib re-write. If you use vec_foo, you'll need to look for _mm_foo and re-write your own code. But vDSP, vImage, and other higher-level functions outlined on the linked documentation page, those should 'just work', as long as there's supported platform to target, as likely will be the case for Intel chips moving forward...
Altivec-optimized code is going to be the single largest issue for programmers in the switch to Intel. Meaning, it's actually going to be a pretty darn smooth transition for most.
Very funny, r jensen. Everyone knows I was talking about the Microsoft variety Office, not the Star or Open variety.
I mean Open Office *is* free, is there a reason *anyone* buys Star Office??
They still have AppleWorks. I think it even still ships with every Mac. Hey, check it out, can it really run on Windows?? It appears it can.
It's definitely still useful, though it's rudimentary spreadsheet is probably the weakest link, it's Carbon of course, and badly needs an update... although, now that I mention it, it looks like it has actually bumped a few version numbers since I last looked- interesting, huh?!? It does seem to be in fairly active development for something we'd all written off.
Pages doesn't really replace a word processor, I don't think you'd use it to write a report or something, it's really geared towards making a newsletter with ( somewhat ) fancy graphics or something. It's more of a niche app, like a end-user Illustrator or something.
No, AppleWorks doesn't have half the features of word. Then again, do you use half the features of Word ? It occupies that niche for folks who aren't going to pay for Office. It's $79 new, and though I doubt they sell a lot of copies that way, it's still a hell of a lot cheaper than Office.
Of course, it's entirely possible that Numbers is something different/more than a spreadsheet. Maybe it's a student-version Mathematics package. Maybe it's just a common word Apple thought they could snap up. We won't know until a product is shipped.
I thought so. For whatever odd reason, I like to inform. Probably becasuse I like to learn, too, I guess.
Currently we are using MFC as the frame work.... And IT SUCKS to be blunt....We offer our software bundled with notebook computers. We currently sell IBM Thinkpads.... Seems like a good time to find a new vendor.
Ouch and double ouch.
You ask what is different now? Well when Apple moved from 68k to PPC the emulated 68k programs actually ran faster then the native machines that they replaced. Also there was no investment in AltaVec yet. There is no way that a P4 is going to emulate a ppc as fast as a G5 at least for a few years. I feel that is really going to hurt Apple.
It would hurt Apple, IF they hadn't made it so easy to recompile PPC code to x86. I say 'recompile', not 'port', because ( unless you're dealing with endian issues, hand-optimized Altivec, or a very, very small host of other issues ), it really is just a recompile. NeXTStep had cross-compiled 'fat' binaries a decade ago, and Apple's just leveraging that existing technology. QT/C++ for OS X Intel is likely to be no different than QT/C++ for OS X PPC. Most people won't be running emulated code. For those that will... reports are Rosetta ( the emulation engine ) doesn't run on those 3.6Ghz P4 systems at G5 speeds, but they do run faster than 800Mhz G4 speeds... which is pretty damn impressive. I'm still using my 800Mhz G4 iMac at home, and I never think "damn this is slow". I think maybe Rosetta pulls some Cocoa-lib-related tricks, so a C++/QT combo might not find *quite* the same speed ( I'm just speculating, it could be the same ), but still... for some integer-op-heavy code, based on initial reports, a dual-core intel chip running emulated code might easily outperform a 1.8Ghz G5 as found in most iMacs, running that same native code. Most importantly, those apps will be fast enough to use. Also, keep in mind, Apple will be using Intel chips released over the next two years- few are predicting that we'll see P4s in anything but these development machines we currently have.
Better for developers, many MacIntel users won't realize this, or won't be upgrading from G5s, or will want the extra performance anyway, so they'll be inclined to buy new code from us developer types. Code which we'll likely be able to provide as a 'cheap' upgrade, because, well... it took 2 hours to 'update'. It's an upgrade cycle that I think will actually be a boon to many OS X software merchants...
They may have already done a design, but you make a good point... especially since the low-end and mobile machines will be the first to be Intelized, it's not likely.
Yea? Really? On who's hardware?
Did Dell sell NeXTStep for Intel?
Keep dreaming...
Maybe Apple can negotiate an anti-competitive exclusive OS provider contract with Dell... I guess, since they're not a convicted monopolist, they could actually do that... but why don't we just talk about Microsoft going bankrupt and Bill Gates living in a van down by the river, as long as we're talking about things that just aren't going to happen ?
I mean, really, it's about as likely as Apple using Intel chips for its... oh, wait...
What is the rationale for that ? There isn't really much of anything development-wise that's really affected terribly by the hardware switch. Especially if you're using C++ and QT... what's changed now that Apple's announced a slow, 2-year migration to a different hardware platform that would keep you from doing a port? The port you're talking about shouldn't be that hard.
Frankly I am afraid of the Osborne effect killing Apple.
I likely can't convince you of it now, but that effect is way overblown, and Apple has weathered such a change *twice* before, and is in a stronger position now than it was either with the 68k/PowerPC switch or the OS9/OS X transition. I don't see why this should be sucha concern- people who were going to hold off on buying new hardware were going to hold off anyway; people who need a new OS X-running machine are still buying them, knowing that their machines will be outdated soon. Steve Ballmer was actually right on this one: "What's different?"
So I will most likely be stuck using C++. Now if Troll tech would come out with Objective-C bindings and Intel would come put an Objective-C compiler for the Mac and for Windows..... So how long before the Intel Compiler is available for Mac OS/X?
Are you trolling me, or just uninformed? Sorry... I just have to ask, because none of the things you're asking about are really needed, AFAIK. You can mix Objective-C and C++ code if you feel you must, so an Objective-C/QT binding is something that, well, you can probably work out a controller layer for what you need without too much effort, but you shouldn't need to if your code is already in C++. Leave it there. The GNU compiler ( which is what's used on OS X ) compiles C++ just fine, and there's plenty of support and documentation for using C++ in OS X. Cocoa is only *one* of many ways to write an app for OS X. It's the way that provides the cleanest user experience, but if you want to port your QT/C++ app to OS X, your limitations are likely going to be from QT rather than any other part of OS X.
The announced plan to use the Intel compiler is really just an optimization; they're building Intel-compatable OS X programs already without it. Actually, the compiler itself isn't likely to change much... maybe Apple will help Intel throw in some Objective-C magic if it's not already there, but that's not all that much, really, it's portable C code and a fairly well-known quantity.
The bottom line for you is that what you need to do is already easy, so much so that Apple's usually extensive documentation boils down to just a page for QT. I'm puzzled as to why you don't already have an OS X port. If you have a QT app that already works under Windows and KDE... it should really 'just work' under OS X, with not much effort at all. I mean, that sort of portability is why you're paying TrollTech the big bucks, right??
hey, I wasn't trying to slam you, just correct your very-nearly-correct information. You nearly had it right. I had to look it up.
The NeXT/OS/X stack is very flexible but I am not so sure about Darwin/Mach.
Interesting distinction, isn't it ? The fact that the Foundation and Application libraries of the Cocoa/OpenStep/GNUStep stack are so easily separated from the underlying 'kernel' is part of what will keep it possible to move forward as unforeseen changes occur. The difference between the NeXTStep Mach underpinnings and the Darwin core are already terribly significant. Darwin is more BSD than Mach.
I have heard that it has poor threading performance which in the new multi core world is going to hurt.
Specifically, you've heard that thread initialization is expensive. After that, the threading performance is actually very good. Proper thread pooling techniques utilize the system quite well. I've not studied how the current OS X core fares on multiple-CPU systems as compared to other modern OSs, but I've heard it does quite well in such comparisons. Real-world, full multiprocessor utilization on dual-CPU OS X systems is very common.
The big question now is do I learn objective c or stick with c++:)
If you already know C++, it won't take you more than a couple of weeks to pick up the basics of Objective-C. It'll shock you what a simple, clean syntax it has. What might take more time is becoming familiar with the AppKit and Foundation APIs and more 'advanced' time-saving features like bindings and the Document architecture, and a few related tools ( like Interface Builder ). But if you just want to do a quick application development project on OS X, by all means... you shouldn't find learning Objective-C to be much of an issue at all. Pick up "Cocoa Programming for Mac OS X" by Aaron Hillegass. I don't *really* agree with his assessment that "mastering [Objective-C] will take two hours if you already know C and an object-oriented programming language like Java or C++", but if you are any kind of programmer, the language is not an obstacle, it's a blessing.
The only real problem might be that you learn Objective-C and start getting mad at issues in C++ that you didn't even notice before.
I think Sun Open-sourcing parts of Solaris is enough, thanks. I'd forgot to mention that one.
Don't a good number of companies give what are essentially specs ( possibly bundled with a few easily reproduced tools ) under the guise of a licensed developer program or some such?
Me, I miss CP/M... I haven't seen it since 6th grade...
It's official, we've crossed over into the Bizarro World.
In the long run, you'll probably end up cooler and save money on your electric bill. Seriously...
Or is it just some elaborate troll meant to make Canadians look stupid ? To his credit, it does cool the spot near his 'heat pump'. You can actually buy similar devices here in the states at your local hardware store ( put ice in the top, a fan blows over rotating fabric band which melting ice coats w/ cool water ), so it's not *quite* as dumb as it sounds. Well. Ok. It is, actually, unless you're just interested in cooling a single spot, and not interested in doing so efficiently. In which case, uh... small air conditioners are in fact not all that expensive.
Why, I guess it is. I couldn't remember the name, honestly, just the latex outfit. If you look, you'll notice that there is at least one image of Ceren on the page I referenced above, though.
Now I can only question why you'd choose an image without said red latex suit... I'm going to guess you're just showing off/trolling. Typical BSD geek behavior ;-)
So no, I didn't know her by name, nor did I recognize her without the outfit. You're clearly a much bigger BSD geek than I. I bow before you in shame.
But really, how famous do you expect the 'most famous BSD chick' to be ?
I mean, you've got to be able to come up with a better BSD daemon girl than that without even trying. What, is that your girlfriend or something? Pathetic.
Honestly, doing a google search didn't give me _just_ the image I wanted, but there are some pretty impressive examples in this collection, even if what is perhaps the best one is animated. ( Warning: not entirely work-safe, *and* contains flamefest-inducing images of penguins impaled on pitchforks ). You've been warned, now let's see that server melt...
Announced. Shipping in quantity? The e600 is dual-core, but not at any higher clock speed. I'd be curious if they can ship it in quantities Apple needs at this point.
Yea, Apple supposedly has the dual-core G5 chip in testing. I'm willing to bet they'll actually ship some of those on the high-end, actually, this is a two-year plan after all. I'm much less sure about the e600, though, unless Freescale is already making them in quantity.
And to answer the story's question ( not the article's ), zealots and evangelists can both hurt and help the thing they're trying to support. A more balanced, rational approach is always more likely to convert the unwilling. The more zealous approach is always more likely to convert the willing.
Either approach is likely to make someone wonder what your motives are, and make them want to defend the choice they've made. I think that's the real mistake; you shouldn't be making someone feel stupid because they chose Windows, you should be asking if they've had problems because of that choice, and helpfully offering alternatives that you can show solve those problems... same goes with political parties, religion, or any other product you'd like to get others to support.
1. Software vendors will abandon support for PPC machines as soon as it makes business sense to do so.
Of course, but here's the thing : in the very, very, very vast majority of cases, it won't make sense. Comparing the situation to NeXT is a bit silly ( unless you'd like to compare numbers of installed users, see my point yet? ). If someone decides to stop supporting PPC, it means their product has somehow become Intel-specific ( how? ) and they have a small PPC install base, _and_ don't want the existing PPC customers. That should be pretty rare, really. I'd like to see the numbers of NeXT vendors that dropped 68k support and kept Intel support, anyway; I don't think that number is as high as this guy claims.
2. Software vendors will charge you money for Intel (or PPC) versions of software when it makes business sense to do so.
They'll happen either as upgrades or as new versions of software. You'd pay for those anyway. Yes, people buying Intel macs and wanting native performance will end up buying lots of new software. Duh. Guess what happens with every OS switch? Of course, if performance is acceptable, you can probably put off that software purchase for quite a while... initial reports show Rosetta getting pretty decent performance, so this could be less of an issue. Even if it is, is it a reason to not buy a Mac today? It sounds like more of a reason to not buy an Intel Mac if you already own a PPC Mac.
3. Apple has shown it drops support for old products regularly because it makes business sense to do so.
Like every other company on the planet? Yet, with the tools in place to make builds for PPC a single checkbox away, it should be some time before simply using that check box doesn't make business sense. Let's see... when exactly did NeXT stop supporting 68k hardware, since that's what he's comparing this to? Was there ever an NeXTStep for Intel release without a matching 68k release? Ok, I know... there weren't many of the first, but it was pretty easy for them to keep 68k support, and they did.
4. Macs tended to have a far longer life-span than the average PC.
Yea... and this changes how ? Right now, a dual G5 is right up there with a top-of-the-line PC ( well, except maybe for the graphics card, maybe ) and will still compare to that same PC... his prediction is that in 2 years today's PCs will be obsolete? Wow. Stunning.
5. The new Intel machines promise to be much faster than current machines.
Huh. That's the reason for the switch right there in a nutshell, isn't it?
6. People do not buy computers only for how it will serve them today, but for how capable it will be in serving them for their desired term of use.
uh... as compared to _other_ computers available to buy _today_. If I _can_ put off buying a computer, I'm going to; tomorrow will always provide faster computers at lower prices. If I can't ( say, I'm a student starting college this fall... or a business which just hired a new employee... or I just need a computer to handle my digital photos ), well, I'm likely to buy one now rather than wait a year. Better ones will be here in the future, but that's not terribly relevant today. If I want to run OS X, this doesn't deter me from buying a PPC mac unless I can wait anyway. In which case, I was going to maybe wait anyway for a year. For that relatively small group of users that can wait for a year, sure, Apple will parlay those users into pent-up demand for their first Intel machines. That's somehow bad for Apple?
7. Potentially no Classic support.
I thought he was looking for reasons why people won't buy more PPCs? If you want Classic support, you'll run out and stockpile PPC Macs. Believe me, though... few people want it. This publishing industry he's talking about don't represent that many computers... and it does represent a business that sets up systems and never, ever replaces the
That's very insightful. There's plenty of wiggle room in Apple's two-year transition plan for them to say "oh, hey, look at this fantastic new chip from Freescale... well, let's put it in a laptop and see who buys it". If it turns out to be a barn-burner in performance/watt, it's probably not too late. But does Apple really want to be making PPC and Intel versions of their product for a long time? Isn't that a little expensive and confusing for customers? I have to expect Freescale and IBM just said "we're more interested in these other, more profitable chip areas, Steve, sorry", and Steve said, "ok, I'll call Intel".