I absolutely love how some apps implement their metadata editing systems. If Apple does implement a system like this it's very important that there are guidelines for which metadata tags are used and how.
Some methods for throttling metatag clutter seem necessary. Reminds me of scalability issues I read in an article about tag-based systems like del.icio.us a few months ago. There are analogous issues but I'm vague on details right now.
Also, Mel's comment:
Quote:
Companies are already putting metadata tags onto their file types for Spotlight to find.
Even if APIs are used there's something a bit ad-hoc about the tag-defining process that's bothering me. Maybe it's more an issue with current interface limitations than the tags themselves.
Quote:
Standard types such as image resolution should be shared between programs.
With shared redundancy instead of isolated instances of potentially sharable types, e.g. ITPC vs. iPhoto keywords. If that makes any sense.
Quote:
One interesting thing I've learned about HFS+ is that it has had support for arbitrary metadata via the resource fork since its release with Mac OS 8.1. This ability wasn't accessible until Tiger, however, and even now it isn't being used by the OS (though I could be wrong). Actually, can anyone confirm that?
Maybe you're thinking of extended attributes? There's a good explanation of them in John Siracusa's Tiger article on Ars. And UTI's, which may also be relevant to this discussion in the context of what technology Apple has already revealed to us even if not fully exploited.
Quote:
I hope the lack of quality in the OS X Finder is due to Apple concentrating on its resourced on the Finder's replacement. Hey, I can dream, can't I?
I optimistically proposed that particular "Finder rot" conspiracy a few times during my file/folder de-relevance rants.
Two-pronged approach... isn't that pretty much what I described in the long posts discussing folders, traditional and otherwise?
That's how I've understood your ideas although I didn't fully comprehend this one:
Quote:
Traditional folders would be view/controller objects (document objects? now there's a funky idea) into the metadata soup, while smart folders would be straight view objects.
Just because OS X is leading the pack doesn't mean they shouldn't fix what's broken.
I hope that includes fixing (*&# filename extension brokenness we "inherited" with OS X. With my traditional Unix background, I eventually got over the case-preserving filesystem but that extension inkiness has been persistently annoying (when superior alternatives have been proposed).
Last I looked, she had over 120 items on her Desktop. On her PowerBook. Because she's convinced that if she puts it down in a folder she'll never find it again.
Last I looked, she had over 120 items on her Desktop. On her PowerBook. Because she's convinced that if she puts it down in a folder she'll never find it again.
so, if your average person can't even organize in a hierarchy of folders, you think they'll be able to work with meta data.
Oh well, who cares about them.
In a program, it will appear to be the same as it is now. You don't have to know what's under the hood. Open/close menus will still have the files you need, even though they get there by a different route.
As for working with metadate directly, few people will be doing in a way that they know they will be. In other words, they will look for grandma's picture by typing that in. They don't know they are using metadata. If they want to label items, those labels will be used as metadata.
Over time, people will get used to filling in info on things they add to their system, the way we now name files. They will just be presented with more options. They can fill in those that they want to be bothered with. More sophisticated users can even make up their own metadata tags, and use those too.
I can almost picture Steve Jobs on stage, explaining, "Throw anything into OS X, and it will organize it for you. You don't need to remember where you put your photos or movies or documents, it's all in the OS database. Just throw in some information, and boom! - It's there."
Yeah, there's no way I'm showing her that. If she can't figure out static folders, how do you think she's going to manage that? Spotlight seems to hit the same sort of sweetspot of power vs. simplicity for her that google does. No real control over it, but the basic basic functionality does it right 99% of the time right off the bat.
For me, spotlight works great for its INTENDED use. It wasn't meant to replace the finder. It was meant to replace Finder's Find. If I'm searching for a particular file, Spotlight rarely ever fails to find it for me.
For example, I was chatting with my sister about her website that I'm designing for her. She told me some important information about what she wanted, so I spotlighted words I knew we had talked about in our conversation, and spotlight took me right to the chat transcript that I had done a few days earlier. Or when I was looking for a receipt for an item I purchased a year ago, Spotlight took me right to the mail message I had saved. Or when I was searching for a keyword to answer a question on some homework that I had, Spotlight took me straight to my appropriate notes on the subject, or a PDF I had downloaded that addressed the topic as well.
For me, spotlight works great for its INTENDED use. It wasn't meant to replace the finder. It was meant to replace Finder's Find.
In Tiger, certainly. The framework it has laid, however, hints fairly well at things to come. I don't believe that Spotlight itself will replace the Fnder. Instead, the Finder's replacement will be based on the same frameworks which were created as part of the Spotlight project.
To quote Dominic Giampaolo:
"Since Spotlight shipped, I'm back with my file system team hat on most of the time."
If you're familiar with who he is, that sums it up pretty well.
Do you believe they had him working with the Spotlight team in some small way? Do you believe that a company would develop an advanced filesystem and keep the same shell on top of it?
He did but I'm pretty sure it had nothing to do with his Humane UI.
On a separate note:
I like folders. I dislike poor implementations of them. Right now, OS X has a poor implementation of them. Everything's too hierarchical. The "Desktop" isn't the top of the tree, yet appears to be in some views and not in others. The Finder goes out of its way to annoy anyone who wants to browse spacially. My home directory is in /Users/userid, and there's a metric Windows load of poorly defined folders within. The desktop is the default destination for downloaded files in most browsers (why not have a ~/Incoming directory?)
It's better than most. Most GNU/Linux implementations aren't sure if they're supposed to be GUI based or command line based, so they maintain a kind of hybrid file system that doesn't work for either. OS X has the same problem, only slightly less so (at least the config files are in ~/Library, and the bundle format keeps application support files organized without users being bothered by unnecessary complexities); Windows... gah. I never can figure out where an application has automatically dropped a file into.
Someone commented above on the HFS being the first version of Mac OS with folders. Not so, the original MFS also had folders. However, they had limitations which prevented the system from being considered truly hierarchical. The major one was that two files couldn't have the same name on the same disk. In the sense of "Folders as meta data", the original MFS implemented that as close to literally as has ever been.
To a certain extent, folders are suffering because people who write operating systems aren't sure how to deal with them. Spacial browsing is where it's at, but because of a small number of problems, usually exaccebated by a poor folder policy to begin with, everyone has attempted to replace it with ugly and poorly implemented alternatives that, for the most part, are less usable than what they replaced, and seek to turn folders from intuitively understandable collections to mysterious locations on a hard drive that don't quite work the way you expect them.
Comments
Originally posted by Ringo
I absolutely love how some apps implement their metadata editing systems. If Apple does implement a system like this it's very important that there are guidelines for which metadata tags are used and how.
Some methods for throttling metatag clutter seem necessary. Reminds me of scalability issues I read in an article about tag-based systems like del.icio.us a few months ago. There are analogous issues but I'm vague on details right now.
Also, Mel's comment:
Companies are already putting metadata tags onto their file types for Spotlight to find.
Even if APIs are used there's something a bit ad-hoc about the tag-defining process that's bothering me. Maybe it's more an issue with current interface limitations than the tags themselves.
Quote:
Standard types such as image resolution should be shared between programs.
With shared redundancy instead of isolated instances of potentially sharable types, e.g. ITPC vs. iPhoto keywords. If that makes any sense.
One interesting thing I've learned about HFS+ is that it has had support for arbitrary metadata via the resource fork since its release with Mac OS 8.1. This ability wasn't accessible until Tiger, however, and even now it isn't being used by the OS (though I could be wrong). Actually, can anyone confirm that?
Maybe you're thinking of extended attributes? There's a good explanation of them in John Siracusa's Tiger article on Ars. And UTI's, which may also be relevant to this discussion in the context of what technology Apple has already revealed to us even if not fully exploited.
Quote:
I hope the lack of quality in the OS X Finder is due to Apple concentrating on its resourced on the Finder's replacement. Hey, I can dream, can't I?
I optimistically proposed that particular "Finder rot" conspiracy a few times during my file/folder de-relevance rants.
Originally posted by Kickaha
Two-pronged approach... isn't that pretty much what I described in the long posts discussing folders, traditional and otherwise?
That's how I've understood your ideas although I didn't fully comprehend this one:
Traditional folders would be view/controller objects (document objects? now there's a funky idea) into the metadata soup, while smart folders would be straight view objects.
Originally posted by Ringo
Just because OS X is leading the pack doesn't mean they shouldn't fix what's broken.
I hope that includes fixing (*&# filename extension brokenness we "inherited" with OS X. With my traditional Unix background, I eventually got over the case-preserving filesystem but that extension inkiness has been persistently annoying (when superior alternatives have been proposed).
Originally posted by sjk
I hope that includes fixing (*&# filename extension brokenness we "inherited" with OS X.
Funny you should mention it...
They already made something better. Once again, we're just waiting for them to use it.
Originally posted by sjk
Yeah, I referred to UTI's a couple posts ago.
Whoops. I missed that.
Originally posted by sjk
If near-real-time searching of arbitrary amounts of (meta)data were already possible we'd be having a different discussion.
And, eventually, we will.
Last I looked, she had over 120 items on her Desktop. On her PowerBook. Because she's convinced that if she puts it down in a folder she'll never find it again.
Originally posted by Kickaha
Why not, my Mom does.
Last I looked, she had over 120 items on her Desktop. On her PowerBook. Because she's convinced that if she puts it down in a folder she'll never find it again.
She's probably right!
Oh well, who cares about them.
Originally posted by DeaPeaJay
so, if your average person can't even organize in a hierarchy of folders, you think they'll be able to work with meta data.
Oh well, who cares about them.
In a program, it will appear to be the same as it is now. You don't have to know what's under the hood. Open/close menus will still have the files you need, even though they get there by a different route.
As for working with metadate directly, few people will be doing in a way that they know they will be. In other words, they will look for grandma's picture by typing that in. They don't know they are using metadata. If they want to label items, those labels will be used as metadata.
Over time, people will get used to filling in info on things they add to their system, the way we now name files. They will just be presented with more options. They can fill in those that they want to be bothered with. More sophisticated users can even make up their own metadata tags, and use those too.
Originally posted by DeaPeaJay
so, if your average person can't even organize in a hierarchy of folders, you think they'll be able to work with meta data.
Oh well, who cares about them.
Ha! You know what the solution was for her?
Spotlight.
I showed her to throw everything in her Documents folder, and she uses Spotlight to find just about everything now. She loves it.
Pblblblblblblblblt.
Neener neener.
Originally posted by Kickaha
I showed her to throw everything in her Documents folder, and she uses Spotlight to find just about everything now. She loves it.
I suggest you have a look at MoRu then:
MoRu - the right interface to Spotlight
Originally posted by livatlantis
I can almost picture Steve Jobs on stage, explaining, "Throw anything into OS X, and it will organize it for you. You don't need to remember where you put your photos or movies or documents, it's all in the OS database. Just throw in some information, and boom! - It's there."
...at least you got the "boom" part right.
Originally posted by macanoid?
I suggest you have a look at MoRu then:
MoRu - the right interface to Spotlight
MoRU does a good job with creating smart folders. I would use it except for two things:
- It is slow for some reason
- No integration with the Finder
I wish it could just allow you to create the smart folders and then export them to be used in the Finder.
At least for her.
For example, I was chatting with my sister about her website that I'm designing for her. She told me some important information about what she wanted, so I spotlighted words I knew we had talked about in our conversation, and spotlight took me right to the chat transcript that I had done a few days earlier. Or when I was looking for a receipt for an item I purchased a year ago, Spotlight took me right to the mail message I had saved. Or when I was searching for a keyword to answer a question on some homework that I had, Spotlight took me straight to my appropriate notes on the subject, or a PDF I had downloaded that addressed the topic as well.
succintly: Spotlight Rocks
Originally posted by DeaPeaJay
For me, spotlight works great for its INTENDED use. It wasn't meant to replace the finder. It was meant to replace Finder's Find.
In Tiger, certainly. The framework it has laid, however, hints fairly well at things to come. I don't believe that Spotlight itself will replace the Fnder. Instead, the Finder's replacement will be based on the same frameworks which were created as part of the Spotlight project.
To quote Dominic Giampaolo:
"Since Spotlight shipped, I'm back with my file system team hat on most of the time."
If you're familiar with who he is, that sums it up pretty well.
Do you believe they had him working with the Spotlight team in some small way? Do you believe that a company would develop an advanced filesystem and keep the same shell on top of it?
Er, wait... shoot. Other than Microsoft, I mean.
Originally posted by DeaPeaJay
didn't he die?
He did but I'm pretty sure it had nothing to do with his Humane UI.
On a separate note:
I like folders. I dislike poor implementations of them. Right now, OS X has a poor implementation of them. Everything's too hierarchical. The "Desktop" isn't the top of the tree, yet appears to be in some views and not in others. The Finder goes out of its way to annoy anyone who wants to browse spacially. My home directory is in /Users/userid, and there's a metric Windows load of poorly defined folders within. The desktop is the default destination for downloaded files in most browsers (why not have a ~/Incoming directory?)
It's better than most. Most GNU/Linux implementations aren't sure if they're supposed to be GUI based or command line based, so they maintain a kind of hybrid file system that doesn't work for either. OS X has the same problem, only slightly less so (at least the config files are in ~/Library, and the bundle format keeps application support files organized without users being bothered by unnecessary complexities); Windows... gah. I never can figure out where an application has automatically dropped a file into.
Someone commented above on the HFS being the first version of Mac OS with folders. Not so, the original MFS also had folders. However, they had limitations which prevented the system from being considered truly hierarchical. The major one was that two files couldn't have the same name on the same disk. In the sense of "Folders as meta data", the original MFS implemented that as close to literally as has ever been.
To a certain extent, folders are suffering because people who write operating systems aren't sure how to deal with them. Spacial browsing is where it's at, but because of a small number of problems, usually exaccebated by a poor folder policy to begin with, everyone has attempted to replace it with ugly and poorly implemented alternatives that, for the most part, are less usable than what they replaced, and seek to turn folders from intuitively understandable collections to mysterious locations on a hard drive that don't quite work the way you expect them.
We risk losing something great.