Not to be vicious, but I'm not quite understanding your position. You seem to imply that you have access to Panther builds, then you simultaneously solicit possible scoops from other people? Why would you need to?
I was referring to the newest builds, of which I have no knowledge of; I didn't take it as vicious.
Quote:
Since when does Mac OS X ever have extra RAM? Second, caches exist for situations where accessing memory isn't fast enough. Using memory, then, as a "cache" is counterintuitive, unless you're using it as a "cache" for data on your hard disk. Also, what's "fast RAM"?
Hmm, interesting. Sounds like something you'd use temporarily to work with many files and reduce desktop space. Ironic that folders, used to organize documents, would be more cluttering than a pile of documents. The computer world is a wacky one?
Actually, I still don't see any advantages to the "piles" prototypes some people have put forward over folders. Indeed, other than the "gee wheez" factor, piles seem to try to solve the same problem that folders are used for now, only in a less scaleable/useful manner.
Folders don't scale down all that well. The additional level of indirection is nice until you just have a handful of files, and then it's kind of annoying. One possible way to implement piles would be to have OS X model a "collection" of files, which would be presented as a pile if there were, say, fewer than five items, and which would be (visually) "reorganized" into a folder once you added a sixth. So the pile was just the OS automagically collapsing the (view of the) hierarchy to make access to small numbers of files more convenient.
Alternately, it can be a way of dealing with files that are only associated with each other as part of a current project, but which you might still want to be filed away in various folders (almost like a database view created as a join across tables).
I could think of a few other possibilities if I wanted to. The real issue is whether Apple can settle on an implementation that doesn't just add another layer of complexity and confusion, and that is actually, rather than just theoretically, useful.
Alternately, it can be a way of dealing with files that are only associated with each other as part of a current project, but which you might still want to be filed away in various folders (almost like a database view created as a join across tables).
This is how piles read to me when I scoured Apple's patent on the things. Piles are a GUI for database (metadata) functionality. They would allow both "procedural" and hierarchical (folder) organization along side their "smart" functions while keeping the two methods of organizing not only conceptually but also visually distinct. Otherwise, as moki says, they're all sizzle but no steak.
Actually, I still don't see any advantages to the "piles" prototypes some people have put forward over folders. Indeed, other than the "gee wheez" factor, piles seem to try to solve the same problem that folders are used for now, only in a less scaleable/useful manner.
Assume you're in the following situation: you're writing report about, say, the planned changes to your company network. You include an Excel spreadsheet of computers with associated IP addresses, features, DNS and WINS names, etc., you include an OmniOutliner file of steps involved, and an OmniGraffle graph of the future network structure. Oh, and a TextEdit file with further notes.
So there are four files on the Desktop, and you know they are related, but you're too lazy to create a folder, name it appropriately, and put the files in it (I know I often am :P ). Instead, you select the four files, and choose "turn selection into pile" (or something alike). The OS happens to have schemas which describe all of the four mentioned file formats, so it can analyze the files, and find common things. It'll find out that all of them are related to networking; maybe it'll find a topic heading in one or more of the files, and if it's really intelligent, it'll be able to name the pile - tada - "Upcoming company network changes" or something like that.
But a pile isn't bound to the FS hierarchy, but to the semantics. Meaning that you can generate a search query "Mail from Steve Jobs" and you'll get an empty pile, err, I mean, you'll get a pile containing all Mail - no matter what folder - sent from Steve Jobs. This pile, although accessible at its visual location from even the Terminal or from within other applications, physically consists of files spread over the hard drive.
So there are four files on the Desktop, and you know they are related, but you're too lazy to create a folder, name it appropriately, and put the files in it (I know I often am :P ). Instead, you select the four files, and choose "turn selection into pile" (or something alike).
I'm still not groking it -- how is choosing "turn selection into pile" easier than choosing "new folder"? Granted, you have to name it, but the OS can't reasonably determine the name for you anyway.
I'm all for meta-data, and some of the cool things the BeOS Tracker could do, but piles just seems a bit silly to me.
But say you created a pile of the four items. Next you want to open one of the files. You click-and-hold on the pile, a preview of some sort pops up with some information about what kind of doc it is, when it was created, size and so on. You move the mouse to scroll trough the documents. When you have selected the document you wanted, the apropriate app opens the document, and the pile goes back to normal.
when you use a folder. You first have to open it, select the file you want, double click, file opens. Maybe it was the wrong file, you have to close it, back to the folder, select another file, open it ... you get the idea. And when your finished you still have an open folder to close.
And the pile could give you some sort of visulal feedback of the number of files in it, the kind of documents in it and so on.
Did this make any sense to you? This is at least how I thought piles could work. Maybe not a revolution, but kinda cool and kinda useful. Probably never gonna see it in a MacOS release though.
when you use a folder. You first have to open it, select the file you want, double click, file opens. Maybe it was the wrong file, you have to close it, back to the folder, select another file, open it ... you get the idea. And when your finished you still have an open folder to close.
Well, I generally use the browser view -- so I just click on the folder, and double-click on the file I want.
I can't think of a pile implementation that would work more efficiently than this, but hey, you never know...
Say you're a leecher (meaning you acquire lots of stuff from the internet, from friends' CDs, over the network, ...), and awful at organizing things. So you have a huge folder "downloads" or "stuff", but it's everything but useful, because it's UNTIDY.
Well I'm always in that situation, honestly
Anyways, the OS can now create a "from ambrosiasw.com" pile, a "from Bon Jovi CD" one and a "from Daddy's Mac" one from the metadata input it receives. Now ain't that convenient?
Actually, as a aort of pipe dream, it would be tre cool if a computer could (reliably) deduce relationships among files if you just pile them up on one another, then deduce what other files belong in that stack. Obviously, it can't be perfect, and you would probably want to see what you put in, what the computer added in, what you've OK'd, etc. But maybe this relies too much on guessing for the user, a la Microsoft. If their software is a proof of concept, it's a bad concept.
I guess as NetroMac has said, it's maybe more of a browsing tool than an organizational one.
Its interesting reading peoples examples of what a pile might consist of
Piles are a parallel mechanism for organising data. You still have a folder heirachy, but now you can relate files to each other in another manner. Here is an example.
In a folder heirarchy you might have a folder for a client, and in there a folder for letters, specifications, graphics, a web site etc. Say you write a letter to the client, it naturally gets filed in the letters folder. The letter happens to include images ( as external links ) that are in the graphics folder. There is no way for the file system to know that they are related. You could create a new folder, an copy, or put aliases in it, or you could turn them into a pile. Its doesnt affect how the folder heirachy works, it parallels it.
That example isnt smart at all. But having a mechanism to group data across a heirarchy really could be handy. Internally it would probably look just like a folder full of aliases, but the UI is different, to make it easier to setup and use. Im sure Apple have done a lot of work on alternative interfaces to folders, and found that there are several other paradigms that work really well in different situations. Piles could well be one. Spring loaded folders is another. BeOS let you access all the contents of a folder through the context menu, really nice, it became my primary folder browsing mechanism. I really miss it in windows and os x.
People here are talking about smart piles however. They are a bit different, but have so much potential. They dont really have anything to do with the previous example, except possibly the interface used to access them.
The best example of a smart pile ( but with a different UI ) is the 'Recently Used' items menu. That could easily be made more powerful. Say you could have a 'Recently Modified' pile, that was specific to a certain folder? This is related to the concept of metadata. Imagine you had a recently modified pile, and each of the results it brings up is itself a pile ( you might get 5 results ). Those resulting piles look up email, on the topic of the project, sent by the user who modified the file. Now you can click on a pile, see that File X has changed, and that bob, who changed file X, sent an email about the change yesterday.
In either case, piles are really just a UI convenience for something else.
I think that the UI paradigm for piles is that of small collections of disparate files. Big Piles probably turn into folders. Big folders, hmmm, no sure about that.
Guys, I'm going to send this forum to Steve Jobs, in hopes that we can save piles. In my perspective I think it's a great thing that should happen and be immediately implemented. Just think, it limits the use of folders and therefore, hopefully, less time navigating through them.
Even though this idea is in it's infancy, it could get better with more work.
I've compiled some of the ideas below.
Quote:
Originally posted by NETROMac
But say you created a pile of the four items. Next you want to open one of the files. You click-and-hold on the pile, a preview of some sort pops up with some information about what kind of doc it is, when it was created, size and so on. You move the mouse to scroll trough the documents. When you have selected the document you wanted, the apropriate app opens the document, and the pile goes back to normal.
when you use a folder. You first have to open it, select the file you want, double click, file opens. Maybe it was the wrong file, you have to close it, back to the folder, select another file, open it ... you get the idea. And when your finished you still have an open folder to close.
And the pile could give you some sort of visulal feedback of the number of files in it, the kind of documents in it and so on.
Did this make any sense to you? This is at least how I thought piles could work. Maybe not a revolution, but kinda cool and kinda useful. Probably never gonna see it in a MacOS release though.
Quote:
mmmpie
BeOS let you access all the contents of a folder through the context menu, really nice, it became my primary folder browsing mechanism. I really miss it in windows and os x.
Quote:
Chucker
Say you're a leecher (meaning you acquire lots of stuff from the internet, from friends' CDs, over the network, ...), and awful at organizing things. So you have a huge folder "downloads" or "stuff", but it's everything but useful, because it's UNTIDY.
Well I'm always in that situation, honestly
Anyways, the OS can now create a "from ambrosiasw.com" pile, a "from Bon Jovi CD" one and a "from Daddy's Mac" one from the metadata input it receives. Now ain't that convenient?
Guys, I'm going to send this forum to Steve Jobs, in hopes that we can save piles.
If you had been around in the pre-10.2-GM days, you'd know that this is a fruitless effort. Recall the brief introduction of minimize-in-place at the time and how people thought sending feedback about it would help. If anything, this might just annoy some people at Apple that you guys are complaining about the removal of a feature that was 1. never meant to be seen by the public and 2. very incomplete in its temporary implementation.
Quote:
I've compiled some of the ideas below.
Frankly, I can't say I agree with NETROMac's idea. I wasn't going to bother responding to it earlier, but if you think you're actually going to send it off, I'd like to make some comments first.
When I picture what NETROMac describes, I don't see anything that is really any better than what we already have. Giant tooltips or popup windows with information about files when you mouse over them? No thanks. Click-hold-wait-read-move-read-release? How is this any easier than opening a folder in list view? At least in list view you can see everything at once and you don't have the added inconveniece of having to keep the mouse button held down as you browse its contents.
The comment "Maybe it was the wrong file, you have to close it..." also bothers me. How would piles improve this situation in any way over what we have in folders? To provide any additional information about the file than we already have in normal windows would require some kind of new metadata implementation. If some friendly metadata is indeed introducted, don't you think it would carry over into the regular browse views as well?
I just don't see the point other than the "gee whiz" factor like moki said. What Amorph said makes a bit more sense to me. Piles, as he describes them, would actually be the same as folders, just be represented differently to the user depending on how many items are contained within them. Even in this case it doesn't really add anything too useful to the user experience.
I can't imagine the system automagically piling up my files much better than it does today, especially since most files downloaded off the internet would have no metadata at all to use for reference. Something like the "downloads pile" that Chucker wants is already available today. It's called a folder. I have my downloads set to go to a specific folder appropriately called "Downloads" so they don't clutter up other places.
Something like the "downloads pile" that Chucker wants is already available today. It's called a folder. I have my downloads set to go to a specific folder appropriately called "Downloads" so they don't clutter up other places.
And you've got a folder action for the various stuff you put in it, so it automatically sorts everything for you? Didn't think so.
Frankly, I can't say I agree with NETROMac's idea. I wasn't going to bother responding to it earlier, but if you think you're actually going to send it off, I'd like to make some comments first.
Thanks, you're flattering me
Quote:
When I picture what NETROMac describes, I don't see anything that is really any better than what we already have. Giant tooltips or popup windows with information about files when you mouse over them? No thanks. Click-hold-wait-read-move-read-release? How is this any easier than opening a folder in list view? At least in list view you can see everything at once and you don't have the added inconveniece of having to keep the mouse button held down as you browse its contents.
I have been thinking a little about this, and I have to agree with you Brad. It won't be any easier than using a regular folder. One problem is that you will probably not see details about more than one file at the time, and that is a major problem with "my" piles idea.
Quote:
The comment "Maybe it was the wrong file, you have to close it..." also bothers me. How would piles improve this situation in any way over what we have in folders? To provide any additional information about the file than we already have in normal windows would require some kind of new metadata implementation. If some friendly metadata is indeed introducted, don't you think it would carry over into the regular browse views as well?
I guess it would.
Quote:
I just don't see the point other than the "gee whiz" factor like moki said.
Both you and Moki have a point here.
Quote:
I can't imagine the system automagically piling up my files much better than it does today, especially since most files downloaded off the internet would have no metadata at all to use for reference. Something like the "downloads pile" that Chucker wants is already available today. It's called a folder. I have my downloads set to go to a specific folder appropriately called "Downloads" so they don't clutter up other places.
Agree. And some sort of "smart" (regular) folder would seem more appropriate here as you can easily see whats actually inside it.
But how about this then:
If you have some files on the desktop making it look cluttered, you select them and choose a "make pile" from the menu. They are then grouped together in a pile with some sort of visual feedback on how many files it contains. When you then click on the pile, the documents and files go back to the position they had before adding them to the pile. If you double click it - the file opens and the pile goes back to its "closed" state. If you want to remove a document from the pile you click on the pile, the files "flotes" back to their positions, and you just drag the file a little and the document/file is removed from the pile, and the pile closes again.
In a folder it could work in a similar way to, but say only in icon view. In column view the files contained in the pile should display as normal.
If piles are going to work as I describes in my previous post, they would add nothing to the computing experience other than some cool effects maybe. Creating a new folder on the desktop would be an equally simple way to group the documents.
Comments
Originally posted by frawgz
Not to be vicious, but I'm not quite understanding your position. You seem to imply that you have access to Panther builds, then you simultaneously solicit possible scoops from other people? Why would you need to?
I was referring to the newest builds, of which I have no knowledge of; I didn't take it as vicious.
Since when does Mac OS X ever have extra RAM? Second, caches exist for situations where accessing memory isn't fast enough. Using memory, then, as a "cache" is counterintuitive, unless you're using it as a "cache" for data on your hard disk. Also, what's "fast RAM"?
"fast RAM" - maybe 'faster-then-hard-disk RAM'
Sadly my two forum threads are running into each other, for a greater discussion on all of this stuff, checkout the "future hardware > Panther - SPEED 'DAEMON' (future/older hardware) ."
Originally posted by willywalloo
I was referring to the newest builds, of which I have no knowledge of; I didn't take it as vicious.
So which builds do you have???
Originally posted by Dog Almighty
Hmm, interesting. Sounds like something you'd use temporarily to work with many files and reduce desktop space. Ironic that folders, used to organize documents, would be more cluttering than a pile of documents. The computer world is a wacky one?
Actually, I still don't see any advantages to the "piles" prototypes some people have put forward over folders. Indeed, other than the "gee wheez" factor, piles seem to try to solve the same problem that folders are used for now, only in a less scaleable/useful manner.
Alternately, it can be a way of dealing with files that are only associated with each other as part of a current project, but which you might still want to be filed away in various folders (almost like a database view created as a join across tables).
I could think of a few other possibilities if I wanted to. The real issue is whether Apple can settle on an implementation that doesn't just add another layer of complexity and confusion, and that is actually, rather than just theoretically, useful.
Originally posted by Amorph
Alternately, it can be a way of dealing with files that are only associated with each other as part of a current project, but which you might still want to be filed away in various folders (almost like a database view created as a join across tables).
This is how piles read to me when I scoured Apple's patent on the things. Piles are a GUI for database (metadata) functionality. They would allow both "procedural" and hierarchical (folder) organization along side their "smart" functions while keeping the two methods of organizing not only conceptually but also visually distinct. Otherwise, as moki says, they're all sizzle but no steak.
Originally posted by moki
Actually, I still don't see any advantages to the "piles" prototypes some people have put forward over folders. Indeed, other than the "gee wheez" factor, piles seem to try to solve the same problem that folders are used for now, only in a less scaleable/useful manner.
Assume you're in the following situation: you're writing report about, say, the planned changes to your company network. You include an Excel spreadsheet of computers with associated IP addresses, features, DNS and WINS names, etc., you include an OmniOutliner file of steps involved, and an OmniGraffle graph of the future network structure. Oh, and a TextEdit file with further notes.
So there are four files on the Desktop, and you know they are related, but you're too lazy to create a folder, name it appropriately, and put the files in it (I know I often am :P ). Instead, you select the four files, and choose "turn selection into pile" (or something alike). The OS happens to have schemas which describe all of the four mentioned file formats, so it can analyze the files, and find common things. It'll find out that all of them are related to networking; maybe it'll find a topic heading in one or more of the files, and if it's really intelligent, it'll be able to name the pile - tada - "Upcoming company network changes" or something like that.
But a pile isn't bound to the FS hierarchy, but to the semantics. Meaning that you can generate a search query "Mail from Steve Jobs" and you'll get an empty pile, err, I mean, you'll get a pile containing all Mail - no matter what folder - sent from Steve Jobs. This pile, although accessible at its visual location from even the Terminal or from within other applications, physically consists of files spread over the hard drive.
Originally posted by Dog Almighty
We can still hint and hope at features we'd like to see in Panther. Who knows? maybe Apple is reading this right now? ¬.¬
Ooh, they could be among us?
Probably not, but 35 days is a long wait.
actually, they have used ideas off these boards verbatim in the past, so it is highly likely that, well, they'll take a good idea if they hear it.
Originally posted by Chucker
So there are four files on the Desktop, and you know they are related, but you're too lazy to create a folder, name it appropriately, and put the files in it (I know I often am :P ). Instead, you select the four files, and choose "turn selection into pile" (or something alike).
I'm still not groking it -- how is choosing "turn selection into pile" easier than choosing "new folder"? Granted, you have to name it, but the OS can't reasonably determine the name for you anyway.
I'm all for meta-data, and some of the cool things the BeOS Tracker could do, but piles just seems a bit silly to me.
when you use a folder. You first have to open it, select the file you want, double click, file opens. Maybe it was the wrong file, you have to close it, back to the folder, select another file, open it ... you get the idea. And when your finished you still have an open folder to close.
And the pile could give you some sort of visulal feedback of the number of files in it, the kind of documents in it and so on.
Did this make any sense to you? This is at least how I thought piles could work. Maybe not a revolution, but kinda cool and kinda useful. Probably never gonna see it in a MacOS release though.
Originally posted by NETROMac
when you use a folder. You first have to open it, select the file you want, double click, file opens. Maybe it was the wrong file, you have to close it, back to the folder, select another file, open it ... you get the idea. And when your finished you still have an open folder to close.
Well, I generally use the browser view -- so I just click on the folder, and double-click on the file I want.
I can't think of a pile implementation that would work more efficiently than this, but hey, you never know...
Say you're a leecher (meaning you acquire lots of stuff from the internet, from friends' CDs, over the network, ...), and awful at organizing things. So you have a huge folder "downloads" or "stuff", but it's everything but useful, because it's UNTIDY.
Well I'm always in that situation, honestly
Anyways, the OS can now create a "from ambrosiasw.com" pile, a "from Bon Jovi CD" one and a "from Daddy's Mac" one from the metadata input it receives. Now ain't that convenient?
I guess as NetroMac has said, it's maybe more of a browsing tool than an organizational one.
Piles are a parallel mechanism for organising data. You still have a folder heirachy, but now you can relate files to each other in another manner. Here is an example.
In a folder heirarchy you might have a folder for a client, and in there a folder for letters, specifications, graphics, a web site etc. Say you write a letter to the client, it naturally gets filed in the letters folder. The letter happens to include images ( as external links ) that are in the graphics folder. There is no way for the file system to know that they are related. You could create a new folder, an copy, or put aliases in it, or you could turn them into a pile. Its doesnt affect how the folder heirachy works, it parallels it.
That example isnt smart at all. But having a mechanism to group data across a heirarchy really could be handy. Internally it would probably look just like a folder full of aliases, but the UI is different, to make it easier to setup and use. Im sure Apple have done a lot of work on alternative interfaces to folders, and found that there are several other paradigms that work really well in different situations. Piles could well be one. Spring loaded folders is another. BeOS let you access all the contents of a folder through the context menu, really nice, it became my primary folder browsing mechanism. I really miss it in windows and os x.
People here are talking about smart piles however. They are a bit different, but have so much potential. They dont really have anything to do with the previous example, except possibly the interface used to access them.
The best example of a smart pile ( but with a different UI ) is the 'Recently Used' items menu. That could easily be made more powerful. Say you could have a 'Recently Modified' pile, that was specific to a certain folder? This is related to the concept of metadata. Imagine you had a recently modified pile, and each of the results it brings up is itself a pile ( you might get 5 results ). Those resulting piles look up email, on the topic of the project, sent by the user who modified the file. Now you can click on a pile, see that File X has changed, and that bob, who changed file X, sent an email about the change yesterday.
In either case, piles are really just a UI convenience for something else.
I think that the UI paradigm for piles is that of small collections of disparate files. Big Piles probably turn into folders. Big folders, hmmm, no sure about that.
Even though this idea is in it's infancy, it could get better with more work.
I've compiled some of the ideas below.
Originally posted by NETROMac
But say you created a pile of the four items. Next you want to open one of the files. You click-and-hold on the pile, a preview of some sort pops up with some information about what kind of doc it is, when it was created, size and so on. You move the mouse to scroll trough the documents. When you have selected the document you wanted, the apropriate app opens the document, and the pile goes back to normal.
when you use a folder. You first have to open it, select the file you want, double click, file opens. Maybe it was the wrong file, you have to close it, back to the folder, select another file, open it ... you get the idea. And when your finished you still have an open folder to close.
And the pile could give you some sort of visulal feedback of the number of files in it, the kind of documents in it and so on.
Did this make any sense to you? This is at least how I thought piles could work. Maybe not a revolution, but kinda cool and kinda useful. Probably never gonna see it in a MacOS release though.
mmmpie
BeOS let you access all the contents of a folder through the context menu, really nice, it became my primary folder browsing mechanism. I really miss it in windows and os x.
Chucker
Say you're a leecher (meaning you acquire lots of stuff from the internet, from friends' CDs, over the network, ...), and awful at organizing things. So you have a huge folder "downloads" or "stuff", but it's everything but useful, because it's UNTIDY.
Well I'm always in that situation, honestly
Anyways, the OS can now create a "from ambrosiasw.com" pile, a "from Bon Jovi CD" one and a "from Daddy's Mac" one from the metadata input it receives. Now ain't that convenient?
Originally posted by NETROMac
So which builds do you have???
>nice.
Originally posted by willywalloo
>nice.
Sorry 'bout that one
Originally posted by willywalloo
Guys, I'm going to send this forum to Steve Jobs, in hopes that we can save piles.
If you had been around in the pre-10.2-GM days, you'd know that this is a fruitless effort. Recall the brief introduction of minimize-in-place at the time and how people thought sending feedback about it would help. If anything, this might just annoy some people at Apple that you guys are complaining about the removal of a feature that was 1. never meant to be seen by the public and 2. very incomplete in its temporary implementation.
I've compiled some of the ideas below.
Frankly, I can't say I agree with NETROMac's idea. I wasn't going to bother responding to it earlier, but if you think you're actually going to send it off, I'd like to make some comments first.
When I picture what NETROMac describes, I don't see anything that is really any better than what we already have. Giant tooltips or popup windows with information about files when you mouse over them? No thanks. Click-hold-wait-read-move-read-release? How is this any easier than opening a folder in list view? At least in list view you can see everything at once and you don't have the added inconveniece of having to keep the mouse button held down as you browse its contents.
The comment "Maybe it was the wrong file, you have to close it..." also bothers me. How would piles improve this situation in any way over what we have in folders? To provide any additional information about the file than we already have in normal windows would require some kind of new metadata implementation. If some friendly metadata is indeed introducted, don't you think it would carry over into the regular browse views as well?
I just don't see the point other than the "gee whiz" factor like moki said. What Amorph said makes a bit more sense to me. Piles, as he describes them, would actually be the same as folders, just be represented differently to the user depending on how many items are contained within them. Even in this case it doesn't really add anything too useful to the user experience.
I can't imagine the system automagically piling up my files much better than it does today, especially since most files downloaded off the internet would have no metadata at all to use for reference. Something like the "downloads pile" that Chucker wants is already available today. It's called a folder. I have my downloads set to go to a specific folder appropriately called "Downloads" so they don't clutter up other places.
Originally posted by Brad
Something like the "downloads pile" that Chucker wants is already available today. It's called a folder. I have my downloads set to go to a specific folder appropriately called "Downloads" so they don't clutter up other places.
And you've got a folder action for the various stuff you put in it, so it automatically sorts everything for you? Didn't think so.
Originally posted by Brad
Frankly, I can't say I agree with NETROMac's idea. I wasn't going to bother responding to it earlier, but if you think you're actually going to send it off, I'd like to make some comments first.
Thanks, you're flattering me
When I picture what NETROMac describes, I don't see anything that is really any better than what we already have. Giant tooltips or popup windows with information about files when you mouse over them? No thanks. Click-hold-wait-read-move-read-release? How is this any easier than opening a folder in list view? At least in list view you can see everything at once and you don't have the added inconveniece of having to keep the mouse button held down as you browse its contents.
I have been thinking a little about this, and I have to agree with you Brad. It won't be any easier than using a regular folder. One problem is that you will probably not see details about more than one file at the time, and that is a major problem with "my" piles idea.
The comment "Maybe it was the wrong file, you have to close it..." also bothers me. How would piles improve this situation in any way over what we have in folders? To provide any additional information about the file than we already have in normal windows would require some kind of new metadata implementation. If some friendly metadata is indeed introducted, don't you think it would carry over into the regular browse views as well?
I guess it would.
I just don't see the point other than the "gee whiz" factor like moki said.
Both you and Moki have a point here.
I can't imagine the system automagically piling up my files much better than it does today, especially since most files downloaded off the internet would have no metadata at all to use for reference. Something like the "downloads pile" that Chucker wants is already available today. It's called a folder. I have my downloads set to go to a specific folder appropriately called "Downloads" so they don't clutter up other places.
Agree. And some sort of "smart" (regular) folder would seem more appropriate here as you can easily see whats actually inside it.
But how about this then:
If you have some files on the desktop making it look cluttered, you select them and choose a "make pile" from the menu. They are then grouped together in a pile with some sort of visual feedback on how many files it contains. When you then click on the pile, the documents and files go back to the position they had before adding them to the pile. If you double click it - the file opens and the pile goes back to its "closed" state. If you want to remove a document from the pile you click on the pile, the files "flotes" back to their positions, and you just drag the file a little and the document/file is removed from the pile, and the pile closes again.
In a folder it could work in a similar way to, but say only in icon view. In column view the files contained in the pile should display as normal.
Rant away.
netro
If piles are going to work as I describes in my previous post, they would add nothing to the computing experience other than some cool effects maybe. Creating a new folder on the desktop would be an equally simple way to group the documents.