macnews.net: 10.4/10.5 info? Database Finder?
"First Mac OS X 10.4/10.5 info?
We must admit, we're doubting this one. We've received word that 10.4 is planned for 'end of 04'. So far, so good. The info goes on to inform us that Mac OS X 10.4 will go "further than anticipated", introducing not only a database-driven new Finder (although the file system itself will still be HFS+) but also a wide support for file metadata. According to the source, this will rather replace than extend current implementations of Creator/Type, Labels etc. According to the source, current (very early) builds of Mac OS X 10.4 are best described as 'radical'. Our source ends the (short) information with the addition that because of the 'radical' new features, Apple might call this one Mac OS X 10.5 instead of 10.4.
Our other sources near Apple weren't able to confirm or deny any of this - they haven't heard anything about the next OS X version yet. ADC Premier and Select members inform us that nothing about it has yet been said about it to the (ADC) public and that first information is still expected at WWDC, which will be even later this year than in 2003. While we doubt the source of this piece of information, we thought you might still be interested. We've asked for more (and more precise) information, and we'll post it as we get it."
It sounds quite fascinating, to say the least. I haven't seen anything kicking around the Mac web about 10.4 or 10.5, whereas I remembered a constant stream of info about 10.3 and 10.2 a while before they were officially announced.
What makes me curious is the "database-driven Finder". What exactly would be the ramifications of doing that? And how would it be different from the Finder that we know in OS X today?
[Edit: The place where I saw this is macnews.net.tc. I've never heard of or seen this rumour site before, so I guess you'd have to take this with the usual grain of salt.]
We must admit, we're doubting this one. We've received word that 10.4 is planned for 'end of 04'. So far, so good. The info goes on to inform us that Mac OS X 10.4 will go "further than anticipated", introducing not only a database-driven new Finder (although the file system itself will still be HFS+) but also a wide support for file metadata. According to the source, this will rather replace than extend current implementations of Creator/Type, Labels etc. According to the source, current (very early) builds of Mac OS X 10.4 are best described as 'radical'. Our source ends the (short) information with the addition that because of the 'radical' new features, Apple might call this one Mac OS X 10.5 instead of 10.4.
Our other sources near Apple weren't able to confirm or deny any of this - they haven't heard anything about the next OS X version yet. ADC Premier and Select members inform us that nothing about it has yet been said about it to the (ADC) public and that first information is still expected at WWDC, which will be even later this year than in 2003. While we doubt the source of this piece of information, we thought you might still be interested. We've asked for more (and more precise) information, and we'll post it as we get it."
It sounds quite fascinating, to say the least. I haven't seen anything kicking around the Mac web about 10.4 or 10.5, whereas I remembered a constant stream of info about 10.3 and 10.2 a while before they were officially announced.
What makes me curious is the "database-driven Finder". What exactly would be the ramifications of doing that? And how would it be different from the Finder that we know in OS X today?
[Edit: The place where I saw this is macnews.net.tc. I've never heard of or seen this rumour site before, so I guess you'd have to take this with the usual grain of salt.]
Comments
People seem to be sick of paying $129 yearly (hell, many of us are shelling out $100/year for .mac before paying for any OS updates). I certainly wouldn't have shelled out the money for 10.3 if i hadn't gotten it for $20 because i had bought my PowerBook in september.
The database finder seems like a very powerful feature, but it also seems like the kind of thing that will take a lot time and energy to finish. I really hope apple is willing to spend the time optimizing the new finder as well as making it easy to use for people who are used to the current finder.
Apple should spend some extra time on making 10.3 a little more stable and less prone on unexpectedly quitting apps. 10.4 should not be less stable than its predecessors and i would much rather have apple spend time on this area of development than release an OS that requires minor updates to gain stability.
Either way i'm quite happy with panther in its current state and i'm hoping that 10.3.3 will stop the unexpectedly quitting apps.
I doubt we'll see 10.4 until mid-late ?05.
Sounds like good news. I don't believe the end of '04 part though. I hope apple starts spreading out OS updates a little more. Not the way MS does, but i think one polished updated every 2 years would be nicer than the current schedule of releasing an update almost annually.
Well that depends. I think we will see cycles of roughly 18 months between upgrades now that OSX is more mature but 24 months may be a bit long. I don't really worry about the price because I have never felt like I had to "keep up with the jones'" regarding Operating Systems.
The database finder seems like a very powerful feature, but it also seems like the kind of thing that will take a lot time and energy to finish. I really hope apple is willing to spend the time optimizing the new finder as well as making it easy to use for people who are used to the current finder.
Time + Energy= Progress.
When users have Terabyte drives and we're stuck with a filesystem that is inefficient in finding the smallest files on the huge drive that won't sit well with people. It won't take as much time as you think. Since day one Apple has had the option of plugging in a new FS into OSX when they're ready. We have yet to really see how extensible this OS is. It's basically a gui wrapper on a bunch of frameworks. Look at the alacrity Apple showed in developing a browser and finishing Quartz. They weren't lying when they said OSX and Cocoa can shorten development cycles.
pple should spend some extra time on making 10.3 a little more stable and less prone on unexpectedly quitting apps. 10.4 should not be less stable than its predecessors and i would much rather have apple spend time on this area of development than release an OS that requires minor updates to gain stability.
I'd be willing to bet money that the Development Teams for 10.3 and 10.4 are two very distinct groups. If you look at panther it is basically "diet" as for end user features and heavy on tweaking internal code as opposed to Jaguar which had plenty of end user features. We all want a polished Panther but Apple surely has the 10.4 Team focused on getting the OS ready.
I'm thinking 10.4 comes January of 2005 with a preview at WWDC. Honestly I'm expecting big things from it. I believe Xcode 2.0 will be announced. I believe that Quicktime 7 will a huge upgrade that will come along with 10.4 if it's not announced at NAB 2004. I see 10.4 as the culmination of 2 years of work by Apple started right after Jaguar and planned. My Theory is that we will see Apple stagger the OS updatee. One year will be feature laden while the next update will be performane oriented. Therefore I don't expect 10.5 to have as many new features as 10.4 but it will undergo further "internal" tweaks in prep for 10.6. I believe 10.6 will be the OS that battles Longhorn.
10.4 will be cool as will 10.5 and 10.6 and 10.X....
I doubt we'll see it ready so soon, though we might see a peek of it at WWDC. Still, the thought is quite fun.
sooner or later, someone will discover a new way to handle all this, and my money's on the horse with the apple logo on it.
Originally posted by Cosmo
People seem to be sick of paying $129 yearly (hell, many of us are shelling out $100/year for .mac before paying for any OS updates). I certainly wouldn't have shelled out the money for 10.3 if i hadn't gotten it for $20 because i had bought my PowerBook in september.
Dude, no one is forcing you to buy those products or services.
Apple should spend some extra time on making 10.3 a little more stable and less prone on unexpectedly quitting apps. 10.4 should not be less stable than its predecessors and i would much rather have apple spend time on this area of development than release an OS that requires minor updates to gain stability.
Either way i'm quite happy with panther in its current state and i'm hoping that 10.3.3 will stop the unexpectedly quitting apps.
I doubt we'll see 10.4 until mid-late ?05.
The OS cannot be blamed for the Apps that unexpectedly quit, the people who develop the apps that come with Mac OS X are not working on the actual OS. 10.3 is very stable for people.
When Apple announces "Smart Folders" it is going to be big stuff. The thing is, you cannot create the same things as an iTunes Library (all your files) and then create normal and smart folders, because it would be too hard. It seems that the old filesystem would need to be remained, sorting stuff into folders. But all the files in your home folder are in a database that can be used for files being put into smart folders.
Originally posted by Harald
It doesn't.
Huh?
Instead of searching through the entire HD during first run, iTunes could query for all .mp3 and .aac files. And having them all neatly sorted inside an iTunes Music folder would be nothing more than for organizational value since file location wouldn't matter anymore.
It doesn't mean Apple has to throw away the hierarchical/spatial organization "front end". I'm sure most consumers prefer this since they don't have that many files to store. But the database aspect will allow apps to break free from this structure of organization that is painfully slow to search through to find the files it needs.
Oh, children, it'll be glorious!
Originally posted by talksense101
I am not too thrilled about a database finder. I am curious about the next update with support for powertune for the 970s. Gimme a 3GHz laptop with battery life > anything in the market. Thanks. We can worry about faster finder implementations later.
Well if you're that curious, there's a few other threads that talk about 'the next update' with powertune support. k thx bye.
Originally posted by BuonRotto
If Apple introduced a database-driven Finder and filesystem this year, Microsoft would soil its collective pantaloons. Holy cow, heads would roll.
Possibly, but the report explicitely stated Apple was sticking to HFS+ and integrating the DB functionality only into the Finder. MS, by contrast, is looking to replace traditional file systems with their database driven WinFS. This is a much more ambitious project.
Originally posted by Smircle
MS, by contrast, is looking to replace traditional file systems with their database driven WinFS. This is a much more ambitious project.
And how successful will this be? As I understand it Longhorn will depart from the norm. It's Microsoft's FIRST attempt at creating something on their own. It will flop because Microsoft can't do anything without a model to copy from. Unless... that model is OS X.
Originally posted by Smircle
MS, by contrast, is looking to replace traditional file systems with their database driven WinFS.
Nope.
WinFS is *not* a file system and does *not* mean "Windows File System". It means "Windows Future Storage" and is a perfectly optional extension to NTFS.
Originally posted by Smircle
Possibly, but the report explicitely stated Apple was sticking to HFS+ and integrating the DB functionality only into the Finder. MS, by contrast, is looking to replace traditional file systems with their database driven WinFS. This is a much more ambitious project.
Ah, good point. I missed that. How would WinFS be different, in more specific terms, if it's running on top of NTFS? As you can tell, I know nothing of the subject, so maybe it's not possible to answer without more info about WinFS available to us.
You would think that a central database would be better for things like the iApps too though, wouldn't you? (Perhaps the iApps would simply use their own DBs like they basically do now.) Or would that be the same mistake like the Windows Registry to put all your eggs in one basket?
I should add that I expect the MS solution to be more ambitious in scope, at least in terms of introducing it in one big chunk. Apple and MS clearly have different MO's with regard to their system software development where Apple takes smaller steps more frequently and MS takes things one big step at a time. After all, that's why OS x has had 4 revisions in 3 years, and Longhorn is going to be five years after their last major OS revision. It's not perfectly apple-to-apples (pardon the pun) like that, but it's a general trend.
What I foresee isn't that the Finder will be the only app metadata-capable, but that we'll see the Finder being the metadata handler, just as it is now, with metadata layered on top of HFS+. Other applications will call a simple API, just as they do now, to work with files and their metadata.
If Apple were to replace the filesystem with something that was utterly different, all the Unix tools would go bye-bye. That's what known technically as a Bad Thing(tm).
So instead they'll layer the metadata over the top, with an API for applications to use, with the Finder being the main access point.
Think of it as Address Book being the main access point to your PIM info, but *any* app can access it through a unified API, such as Mail, iCal, etc.