He makes the argument there that because 10.6 is so great, we don't really need to be disappointed at 10.7 being delayed. But, for me, it is precisely because 10.6 was so great compared to 10.5 (and 10.5 was so great compared to 10.4 and 10.4 was great compared to 10.3, etc.) that I am so disappointed by the delay of 10.7. Apple (often) makes great and innovative products that improve the way you interact with technology. I honestly didn't know I needed spaces/timemachine/spotlight, until I had them. I honestly didn't realize how long the computer took to start up and shut down until 10.6 came along and made it do those things pretty damned quick. Based on Apple's track record of innovation, 10.7 could be super freakin' awesome in ways I can't even think of. But instead... iPad only. The iPad is neat and all, but some of us still need a computer, and computers have a long way to go before they're good enough so that it doesn't make sense to make them better.
I thought 10.6 was basically just like 10.5, and I liked 10.4 enough that I pushed back on upgrading until I had to buy a new machine, whereas I ran out and bought a retail copy of 10.4 to get away from 10.3.
Basically, I agree with the article that progress (or "change", if you don't agree it's progress) has slowed significantly in the OS X line, and so my most recent computer is an Arch Linux box.
Interesting that you see such a huge improvement between 10.6 and 10.5. (To describe it as great)
In my day to day life, I haven't really noticed any practical difference. Maybe expose on the dock, which I haven't developed a habit for using.
We now have coverflow in the finder, which I did use one time to flip through a photo directory quickly.
That's really all I can think of. I know there were many under the hood changes with Grand Central and OpenCL, but it hasn't made any difference to my day to day life.
I will admit that I've become accustomed to my workflow so it's possible that some great new features are waiting there for me but I'm too lazy to learn.
Services makes a huge difference for me. You should take a look at what can be automated and changed in how you work with OS X. It saves me a whole lot of time each day.
Computers are fundamentally different from things like the iPad. The iPad is a console, not a computer. It is heavily restricted in what you can do with it.
Yes, computers are progressing a little more slowly in a sense. This isn't a bad thing. You shouldn't have to buy a new one every two years. But they're still different.
That's because operating systems are fundamentally broken in terms of how they handle application installing, application uninstalling, etc.
OSX's .app drag-and-drop is the closest to the right answer. Linux's package management is a working but hugely complex solution to a problem that should be entirely sidestepped. Windows is totally broken.
Each app should live in its own world. Management of libraries and support infrastructure should also be neatly modular. An app should not have an "installer" and should not be allowed to mess with the OS.
Which is easy on any modern OS (including Windows) ... provided the app doesn't need to interact with others (by giving them a way to handle its file format, for example). At which point it you have the same problems which need to be dealt with one way or another.
Until you want to do something that "Just Can't Be Done", which is where the strengths of having a general-purpose computer that acts like a general-purpose computer come into play.
True, but I think it's a matter of the paths of least resistance.
Let's say there's a 80/20 split, between what is really easy or quick or fun --some subjective measure of "better"-- to do on the console vs what you need a general purpose computer for. At some point, the console is going to win out and those 20 percent items, and tasks that you'd need a general purpose computer for are going to somehow change. Either you'll find another way to accomplish the same general task or tools for the console will arrive that are "good enough". So the question would be, does the console do enough of what people generally do to change the things they rely on a general purpose computer for?
Granted, the successful consoles I can think of are all TV related, and the TV has key advantages over general purpose computers for those consoles (bigger screen, setup to be watched instead of worked on, better sound, etc).
Are these iPad stories ever gonna die? Honestly, why is this news? pg eats breakfast, nothing. Jobs eats breakfast and we spend 6 weeks talking about how awesome the cornflakes are. We get it. Cornflakes are nice. Can we move on?
I disagree. This article isn't really about the IPad, more about how desktop computers have been pretty much the same for a while. The IPad only comes up as a counter-example, though he doesn't spend much time talking about it.
The iPad is a a new product. An entirely new device with (maybe) a new market. Why would incremental updates to existing products supplant a new product?
The iPad is unable to do many of the things I normally do on a computer. That's not a bad thing, it's simply out of the scope of the device. We still need "desk" computers.
I'm not excited by Apple spending less resources on OS X, surely they can do both?
I'm not convinced that this rumor about OSX is true. Has Apple officially said such a thing?
BTW, part of the problem with computers not progressing faster is that the software world is dropping the ball. We're not taking advantage of things like GPGPU, multicore, etc. because we're too busy building me-too whiz-bang webapps.
... begging the question of whether me-too whiz-bang webapps are using things like GPGPU, multicore, etc.
And the desktop software world has been consistently dropping the ball since the beginning. Today's wasteful code that ignores processing resources has solid company among yesterday's wasteful code that was pissing away ram and CPU cycles with locked-in middleware, sloppy code and the rest of our sins. It's nothing new.
And it's not the fault of a shiny new horizon. Desk computers seem to have simply hit a local maxima.
(What is virtualization but an industry-wide condemnation of the state of modern operating systems? We can't get a single box to run a dozen services reliably; instead we run a dozen isolated OS images on that same hardware and things are better.)
The obvious ways to notably improve desk computing require massive change. But we can't quite sell the idea of massive change when things are working reasonably well and most of the planet is accustomed to the current, weird state of desk computing.[1]
So the small incremental changes are the only things left.
[1] I think mobiles and the web are so popular with developers right now precisely because they present the chance to work with and on platforms designed from the ground-up with modern techniques, for modern hardware and services and without all the cruft in the corners.
Since I've started working recently, I've started reading a lot more about GPGPU. Now I kinda want to thrown down the money to build myself a new desktop so I can try to teach myself programming for CUDA...
Still, Im very impressed with the sort of technology we can deal with in our own homes. It's insane to realize just how beefy gpu's are today.
Yeah, it'll be a while before desktops are on the out and out. At that point, the few of us using them will be grateful linux can be installed on anything.
I don't think there is any cause for worry. Lets look at some dates...
(1) Snow Leopard was released 8 months (Aug. '09) ago. (2) Leopard was out about 20 months (Oct '07) before that.
Given that, Apple could announce 10.7 at the 2011 WWDC, release that Fall, and be within their normal timeframes. I think for Apple this is the "Year of the iPad (and OS 4)." We just have to get used to that.
"surely they can do both?" Are you sure? Top quality engineers don't grow on trees and relegating Mac OS X development to second tier engineers because you don't have enough top engineers may be the road to ruin.
Just having billions in the bank doesn't mean you get top engineers when you are hiring. I rather have a somewhat slower developing OS, if the result is good, than a faster developing crappy OS. I was going to compare to Windows, but that is slow in developing and crappy. Oh well.
And I wonder if Apple's insular corporate culture discourages certain kinds of top quality software engineers from joining? I've been reading about Robert Merton's idea that a key motivator for practicing scientists is recognition, so I'm thinking in particular of software engineers who are accustomed to the norms of open source projects.
This might tie back to John Siracusa's recent observation that Google is "a huge gravitational sink for engineering talent," and Apple is not. There's a small discussion here: http://news.ycombinator.com/item?id=1257201
This article lost me at the price comparison, honestly. How many people have a $3000 computer that do not actually need it, or at least something in the same ballpark? If you compared the $700-1500 range to the iPad, maybe. $1500+? You're a pretty serious power user, or you have enough disposable income that this argument is outright irrelevant.
The message is clear: Macs aren’t Apple’s focus right now.
The author is surprised? They just launched a new product category. They're probably pretty sure that most people, by now, already know that they sell laptops.
The author isn't surprised. One main idea of the article is that general-purpose computers are necessarily fading from the limelight as special-purpose computing devices such as the iPad come into their own. The other, more significant, main idea of the article is that this is no cause for alarm.
"If your CPU power dropped by 75% for an hour every day, how long would it take you to notice?"
As long as the hour was one when I was using it, I would immediately notice. But then, I usually only work on dual and quad-core boxes and our optimizing compiler isn't. Maybe I wouldn't notice if I were on a 16 core box and it was just one of the cores.
Quantitative increases in hardware capabilities are boring. (Qualitatively) new software capabilities enabled by those hardware improvements are not. If those aren't forthcoming, it's because software creators (or the creators of their tools) have dropped the ball.
I'm very happy with 10.6. I was very happy with 10.5
I am quite happy with 10.4 actually. I have no reason to spend $140 for 10.6. I'll continue as long as there is a workable Omnigraffle version for 10.4 :)
