In all fairness, the lack of a single emperor of Linux (Linus has little impact on the desktop end of things) has its own problems, too. Standards take a long time to come in...
Granted, when they happen, they are good, but waiting for them and watching the bruises where they should be is not pretty. One place that should be standardizing soon is desktop search, and that will be very cool. May be aligned properly to beat Windows and MacOS to the punch! (Although MacOS does have an integrated, common desktop search via Spotlight, I find that there isn't enough imagination being put in to make it truly useful. It generally just appears in the file open dialog, and not very often elsewhere).
By the way, Ubuntu now writes to NTFS, and I realized that it's been able to do XFS (which Mac users had been quite excited about at one point) for quite a while.
Leopard is pretty cool, and definitely a good step with Time Machine, networking, and those new .Mac features.
I think Apple is putting way too much marketing focus on aesthetics and things that really don't matter, though, and they are promoting a lot of file manager use, which is something Apple has seemed to otherwise avoid...
First off, the coverflow view mode. It's a fricking flip-book style viewer with big preview icons. Please stop talking about coverflow like it's Jesus's successor! Cool effect, sure, but it isn't solving all the world's problems. Actually, it's causing one: View modes like coverflow should not stay on for all folders until toggled off. It is very annoying to have everything be appearing in that mode after turning on coverflow for a specific folder (such as Pictures). It's the sort of view mode that isn't going to be useful in the majority of situations, but can occasionally be really cool.
This isn't really a revolution for file management, and neither are Stacks. (Nothing against Stacks, though; it's a great idea! Really cleans up the desktop, which MacOS needed, the interface flows smoothly and it looks good. In the context of saving the universe, though, stacks aren't there yet). People still tend to hate the process of managing files and prefer when applications just figure it all out, themselves, really easily. This is because, particularly in the case of a photo management tool, a program has gone and arranged files in its own quirky way, probably placed them somewhere that the user shouldn't have to think about. Then, when he has to restore a backup of his photo library, he must figure out the program's abysmal file structure and restore manually via the file manager. One thing MacOS does really well is avoiding that in most cases, but many of these new features (tied directly to Finder) are returning us to that situation. Standards (like Drag&Drop) are good because they let programs handle file management without users having to figure them out as well. Drag & drop abstracts it such that any "thing" can be dragged between programs, and thus file manager usage is avoided, instead permitting a more pointed abstraction of the files to handle the matter of finding stuff. Granted, iLife isn't a part of the OS, so maybe they have added a backup button to everything. We'll have to wait and see...
Integrated (and I mean very integrated) desktop search, and programs that aren't afraid to use it, is the right direction to abstract file management. File management with a file manager like Finder, Explorer, Konqueror, Nautilus or even Dolphin is daunting because every file, whether it is likely to mean something to the user or not, is given the same weight. I will be pleased when a file manager determines the importance of files and folders in certain scenarios, giving particular routes more weight. This would feel like a more pleasant interface as we see in precision-tuned software, generated on the fly to represent (of course) files!
For example, lots of big photos in one place, the rest is databases and ini files. The user probably cares about the photos, and a technical user looking for the technical files will know where to look for them. Apply emblems to the folders to represent what they have, give extra weight to the folders with the user's photos (larger size, placed at top of list -- magazine-style layout). Folders each have information displayed on the side saying what kind of information they have, maybe quick previews of contents to make it more tactile. Differing ammounts of information displayed based on weight (where only a really prominent one would get the previews, for example). Context-sensitive helper text, such as "import the contents to iPhoto," acting as a quick route to the "open with" menu, though easy to find like a web page.
When that happens, I will permit Apple to chant as if they have invented a new universe.
Secondly, Spaces. I was really hoping for something fresh, but what it gives is really a bit weak compared to what I have even in GNOME or KDE. First off, there are no extra options for window management. I cannot set a window to be visible on all spaces, which could have been done quite intuitively with MacOS's proper division of windows and processes. There goes the only offering in virtual desktops that visibly gives an edge over just hiding and showing windows. Secondly, that big Spaces icon in the dock does not give a live preview, and is not interactive. Kind of a shame -- although they definitely did something nice with the full view...
"OOoooh, bigger drop shadows!"
If they can market something trivial like that as one of 300 new features, then a download I've just completed and dropped onto my desktop is the equivalent of three brand new "features" on my computer, invented by me, and I haven't even extracted the thing yet! Not only is there now a beautiful, glossy Archive icon on my desktop, but I can also drag it wherever I want and even roll the mouse over it to have it glow. My very own, customizable, desktop decoration. Wow, pretty.
The new background removal thing in iChat / Photo Booth is quite a joke in many setups. The problem, from what I have gathered, is to do with lighting and how the camera adjusts for it. If someone is wearing white, and the background is dark (even if that background does not move), it will be completely messed up. When the picture of the background is taken, the camera gets it nice and light. Then, when that big, bright person fills the frame, it gets confused and the whole effect breaks away because the camera starts adjusting its exposure (thus a changing background). Beyond me how they can't control this in software. I rarely see the thing working well (though it has done so in scenes where the camera does less adjusting, and I agree it's amusing). As a fun little gimick, I guess it is fine, but it doesn't work most of the time, giving a pretty bad impression. Overlays controlled by hot points in the captured image would have been an equally impressive, though much less likely to screw up, addition. (And potentially useful, too, unlike most of the other additions there).
From what I understand, iChat's desktop sharing only has a single mouse pointer? I haven't had the chance to try it myself, so please correct if wrong. Nothing horribly wrong with that, but may I remind of a few months ago when Mr. Jobs made the claim that the iPhone was running OSX, and that the OS had multitouch capabilities? If that is the case, then multiple mouse pointers should be possible, and indeed compatible with legacy apps, unless they have done something quite strange, or had an entirely different library handle all of the multitouch stuff (which would be silly for an operating system).
It is nicely done, mind. Very cool.
Still no writing to NTFS, which is an unfortunate thing when we have Bootcamp here. Windows can't write to MacOS's stuff, either, so it would be good to have it working in one direction! ...And for Apple's anti-Windows thing, I'm sure they would appreciate the power to say "we can write to Windows, but Windows is too dumb to write to us!" as Linux users now enjoy on a daily basis.
If open source software can do it out of the box, then surely MacOS can.
Okay, there is my (unfortunately) unconstructive ramble for today. Fire away!
Meh, one other thing, this time constructive.
Winni, I have seen small developers (not necessarily shareware, or entirely Linux) making business with Linux sorts of solutions. For example, Iron Realms, who specialize in a lot of MUDs, have that free-to-play scheme with some interesting details for paying users. Their games are very often played via Linux clients. They seem to do quite well with it.
There is definitely a different pattern of thought with Linux software. People there will pay, or tolerate a lot of advertising, if they can see a valuable service being delivered to them consistently.
I can't quite put my finger on why I think it, but I also think that open source games are rather different beasts from open source tools. Getting a really good one (and not just a clone of a commercial game) off the ground would be rather tricky. Amazing if it happened, but I don't think it really has yet. Commericial games, with closed development (maybe open to artwork and level contributions, as any nice community-focused game could do), still has a lot of room even amongst the communist software geeks :P
A question: Have you played much with the web clips feature in Safari? It looks interesting, but I can't decide how it works. Is it doing it a good way, or a chunk of the rendered output taken from set coordinates that never change? (How much is that feature affected by arbitrary changes in page layout? Do you have to redo the clipping, or is it awesomely clever?)
Granted, when they happen, they are good, but waiting for them and watching the bruises where they should be is not pretty. One place that should be standardizing soon is desktop search, and that will be very cool. May be aligned properly to beat Windows and MacOS to the punch! (Although MacOS does have an integrated, common desktop search via Spotlight, I find that there isn't enough imagination being put in to make it truly useful. It generally just appears in the file open dialog, and not very often elsewhere).
By the way, Ubuntu now writes to NTFS, and I realized that it's been able to do XFS (which Mac users had been quite excited about at one point) for quite a while.
Leopard is pretty cool, and definitely a good step with Time Machine, networking, and those new .Mac features.
I think Apple is putting way too much marketing focus on aesthetics and things that really don't matter, though, and they are promoting a lot of file manager use, which is something Apple has seemed to otherwise avoid...
First off, the coverflow view mode. It's a fricking flip-book style viewer with big preview icons. Please stop talking about coverflow like it's Jesus's successor! Cool effect, sure, but it isn't solving all the world's problems. Actually, it's causing one: View modes like coverflow should not stay on for all folders until toggled off. It is very annoying to have everything be appearing in that mode after turning on coverflow for a specific folder (such as Pictures). It's the sort of view mode that isn't going to be useful in the majority of situations, but can occasionally be really cool.
This isn't really a revolution for file management, and neither are Stacks. (Nothing against Stacks, though; it's a great idea! Really cleans up the desktop, which MacOS needed, the interface flows smoothly and it looks good. In the context of saving the universe, though, stacks aren't there yet). People still tend to hate the process of managing files and prefer when applications just figure it all out, themselves, really easily. This is because, particularly in the case of a photo management tool, a program has gone and arranged files in its own quirky way, probably placed them somewhere that the user shouldn't have to think about. Then, when he has to restore a backup of his photo library, he must figure out the program's abysmal file structure and restore manually via the file manager. One thing MacOS does really well is avoiding that in most cases, but many of these new features (tied directly to Finder) are returning us to that situation. Standards (like Drag&Drop) are good because they let programs handle file management without users having to figure them out as well. Drag & drop abstracts it such that any "thing" can be dragged between programs, and thus file manager usage is avoided, instead permitting a more pointed abstraction of the files to handle the matter of finding stuff. Granted, iLife isn't a part of the OS, so maybe they have added a backup button to everything. We'll have to wait and see...
Integrated (and I mean very integrated) desktop search, and programs that aren't afraid to use it, is the right direction to abstract file management. File management with a file manager like Finder, Explorer, Konqueror, Nautilus or even Dolphin is daunting because every file, whether it is likely to mean something to the user or not, is given the same weight. I will be pleased when a file manager determines the importance of files and folders in certain scenarios, giving particular routes more weight. This would feel like a more pleasant interface as we see in precision-tuned software, generated on the fly to represent (of course) files!
For example, lots of big photos in one place, the rest is databases and ini files. The user probably cares about the photos, and a technical user looking for the technical files will know where to look for them. Apply emblems to the folders to represent what they have, give extra weight to the folders with the user's photos (larger size, placed at top of list -- magazine-style layout). Folders each have information displayed on the side saying what kind of information they have, maybe quick previews of contents to make it more tactile. Differing ammounts of information displayed based on weight (where only a really prominent one would get the previews, for example). Context-sensitive helper text, such as "import the contents to iPhoto," acting as a quick route to the "open with" menu, though easy to find like a web page.
When that happens, I will permit Apple to chant as if they have invented a new universe.
Secondly, Spaces. I was really hoping for something fresh, but what it gives is really a bit weak compared to what I have even in GNOME or KDE. First off, there are no extra options for window management. I cannot set a window to be visible on all spaces, which could have been done quite intuitively with MacOS's proper division of windows and processes. There goes the only offering in virtual desktops that visibly gives an edge over just hiding and showing windows. Secondly, that big Spaces icon in the dock does not give a live preview, and is not interactive. Kind of a shame -- although they definitely did something nice with the full view...
"OOoooh, bigger drop shadows!"
If they can market something trivial like that as one of 300 new features, then a download I've just completed and dropped onto my desktop is the equivalent of three brand new "features" on my computer, invented by me, and I haven't even extracted the thing yet! Not only is there now a beautiful, glossy Archive icon on my desktop, but I can also drag it wherever I want and even roll the mouse over it to have it glow. My very own, customizable, desktop decoration. Wow, pretty.
The new background removal thing in iChat / Photo Booth is quite a joke in many setups. The problem, from what I have gathered, is to do with lighting and how the camera adjusts for it. If someone is wearing white, and the background is dark (even if that background does not move), it will be completely messed up. When the picture of the background is taken, the camera gets it nice and light. Then, when that big, bright person fills the frame, it gets confused and the whole effect breaks away because the camera starts adjusting its exposure (thus a changing background). Beyond me how they can't control this in software. I rarely see the thing working well (though it has done so in scenes where the camera does less adjusting, and I agree it's amusing). As a fun little gimick, I guess it is fine, but it doesn't work most of the time, giving a pretty bad impression. Overlays controlled by hot points in the captured image would have been an equally impressive, though much less likely to screw up, addition. (And potentially useful, too, unlike most of the other additions there).
From what I understand, iChat's desktop sharing only has a single mouse pointer? I haven't had the chance to try it myself, so please correct if wrong. Nothing horribly wrong with that, but may I remind of a few months ago when Mr. Jobs made the claim that the iPhone was running OSX, and that the OS had multitouch capabilities? If that is the case, then multiple mouse pointers should be possible, and indeed compatible with legacy apps, unless they have done something quite strange, or had an entirely different library handle all of the multitouch stuff (which would be silly for an operating system).
It is nicely done, mind. Very cool.
Still no writing to NTFS, which is an unfortunate thing when we have Bootcamp here. Windows can't write to MacOS's stuff, either, so it would be good to have it working in one direction! ...And for Apple's anti-Windows thing, I'm sure they would appreciate the power to say "we can write to Windows, but Windows is too dumb to write to us!" as Linux users now enjoy on a daily basis.
If open source software can do it out of the box, then surely MacOS can.
Okay, there is my (unfortunately) unconstructive ramble for today. Fire away!
Meh, one other thing, this time constructive.
Winni, I have seen small developers (not necessarily shareware, or entirely Linux) making business with Linux sorts of solutions. For example, Iron Realms, who specialize in a lot of MUDs, have that free-to-play scheme with some interesting details for paying users. Their games are very often played via Linux clients. They seem to do quite well with it.
There is definitely a different pattern of thought with Linux software. People there will pay, or tolerate a lot of advertising, if they can see a valuable service being delivered to them consistently.
I can't quite put my finger on why I think it, but I also think that open source games are rather different beasts from open source tools. Getting a really good one (and not just a clone of a commercial game) off the ground would be rather tricky. Amazing if it happened, but I don't think it really has yet. Commericial games, with closed development (maybe open to artwork and level contributions, as any nice community-focused game could do), still has a lot of room even amongst the communist software geeks :P
A question: Have you played much with the web clips feature in Safari? It looks interesting, but I can't decide how it works. Is it doing it a good way, or a chunk of the rendered output taken from set coordinates that never change? (How much is that feature affected by arbitrary changes in page layout? Do you have to redo the clipping, or is it awesomely clever?)