I have more than 54 million files on my computer. I am swimming in information – UP TO MY NECK!
Apple’s MacOSX Spotlight and Devon’s EasyFind are terrific tools for using metadata and anything else you know about a file to aid in your search for that needle in the haystack. There are even better tools, I am sure, but for the search function, these are fine for me – and essentially FREE.
My problem #1- to have something in the file that is valuable and useful to search. Its title, original metadata and contents for starters – to augment the metadata if necessary – this is the easy part…
My problem #2 – to collect useful and meaningful metadata from the source of the file when I receive it – this may require an IETF RFC or an ISO standard – a standard schema – this is almost outside my opportunity for influence, but not entirely…
My problem #3 – to edit the metadata easily and nondestructively – even automatically to make it even more useful – more useful to my search tools and to me, too – to refine the metadata if necessary – this is much more difficult…
Text-based files are pretty straight forward as they stand – you can search the content and any annotations you or others have made.
Images and other media have content that is difficult to search without highly specialize comparison tools. Apple’s iPhoto for example allows nearly the full scope of context to be captured – who (with the new facial recognition feature), what (with the sense of an event), when (with the time stamp from the camera media file) and where (with GPS tagging) – you can add manually to the metadata in iPhoto – maybe you want to characterize the subject of the photo with a keyword, genre, “film rating” and emoticon). Apple has done a good job to accumulate and display metadata in this application. I am unsure how much of this metadata is available outside of the iPhoto environment, though – I’ll have to find out…
Metadata gets “massaged” every time it is passed between platforms – this is an area that I am particularly troubled by. ASCII encoded metadata is frequently converted improperly to text when passed via a browser or between Windows computers: 40205b315d becomes %40%20%5b%31%5d through a browser or 252525402520255b2531255d from some Windows applications – or even worse, %2525%2540%2520%255b%2531%255d – or even worse… What I want to see is simply: @ [1]
When I grab a pdf or jpg file off the Internet, I want to add the URL to the file’s metadata – automatically.
When I work with a jpg file and change its compression or dimensions, I may want to add those file characteristics to the file’s metadata.
When I find a file I may not remember much about, MacOSX has a nifty feature I use called QuickLook to give me a quick “peek” at the file without launching an application – I would like QuickLook to let me “peek” at the metadata, too – maybe present the metadata in a generic XML editor like the Property List Editor… That would be even more useful!
But, I don’t want to get so wrapped up in maintaining and managing my metadata that I get distracted from meaningful and productive uses of my time. Metadata – it’s not a big thing – it is a little thing. And that is the challenge – to keep metadata a little thing.
I’m working on this in some spare time. Any volunteers to help me???
More later – please stay tuned!