After reading over "Filenames as a Strategy to Managing Your Assets", my head was spinning. The authors system seemed a little.....overcomplicated? What the author ends up with is a UID for his images and that is whole point, I get it.
I am not knocking what works for him, because clearly he put a lot of thought and possibly, some trial and error into this system. I just think he made it a little harder than it had to be.
I have been raised (by my minimalist father) on the K.I.S.S. system. Keep it simple, stupid. I think this system can be applied to file naming, no matter what type of file it is. In the case of image collections, sure, we have to think about what type of systems we are working with, who is using this information, if the UID is in fact unique to each "asset", but this seems a little excessive...
What do you think? Can you still achieve UID's for image collections by following the K.I.S.S. system?
OR, if I am completely missing the point (which may be the case the more I read over the original blog post), I would love someone to break this down to me.
I am not knocking what works for him, because clearly he put a lot of thought and possibly, some trial and error into this system. I just think he made it a little harder than it had to be.
I have been raised (by my minimalist father) on the K.I.S.S. system. Keep it simple, stupid. I think this system can be applied to file naming, no matter what type of file it is. In the case of image collections, sure, we have to think about what type of systems we are working with, who is using this information, if the UID is in fact unique to each "asset", but this seems a little excessive...
What do you think? Can you still achieve UID's for image collections by following the K.I.S.S. system?
OR, if I am completely missing the point (which may be the case the more I read over the original blog post), I would love someone to break this down to me.
I like K.I.S.S., and yes, a UID could be made with a much simpler system, but it depends on what problems you want that ID to help you solve, and whether the effort in solving the problem is more than just having the problem! Librarians have that issue, too, with our classification numbers. European libraries have closed stacks, and don't need to be able to browse related books on the same shelf, so their book numbers are in the order in which they receive them; book "1145" is the 1145th book gotten by the library. It's a lot quicker than Dewey or LCC, and it still allows them to find books. But if patrons are allowed to use the stacks, then we need a way to put "related" books together.
ReplyDeleteLook at all of the problems the author wants to solve with his naming system:
- keeping track of three kinds of originals (slide, physical negative, born digital photos)
- keep some subject classification for physical originals
- tie digital files to physical item location
- interoperability with OLD Windows systems and any possible photo app
- tie multiple versions of the same photo together
- complete UID in filename, with no dependence on folder names
That's... a lot of problems to try to solve with a filename, especially when the author just goes and sends it off to a DB anyways. And the system definitely shows the signs of evolving by trial-and-error. But, obviously, the author feels that more benefit was gotten out than effort put in. You or me? We might say differently. Well, I would say differently; I'm definitely a fan of "right tool for the right problem," and filenames aren't THAT versatile.
This article left my head spinning as well and I ended up with more questions than answers. I thought his method was a bit over the top and I can't imagine having to do all that legwork - develop schema and suffix code, create database, add descriptive information about image assets into the database - whew! (head spinning again). . . I think I'll stick to the K.I.S.S. method - it works for me (for the time being).
ReplyDeleteIt seems we're all trying to figure out this digital thang! What's the correct way to go foward? By trial and error, of course!!
ReplyDeleteI think KISS works for some files but not for all. I only say this because when I worked in publishing I would have to send image files out to various companies that each wanted the files a different way so if the designer didn't do it then I would have to keep a copy of the original, a copy of a large version, a rbg version set to certain specs, a bw version set to certain specs...this list goes on and on. I had to do this each season (Spring/Summer and Fall/Winter) for over 75-100 images. And don't get me started on the images that had to go out with their ISBNS (screw you Ebooks).
ReplyDeleteI feel like I've went off on a tangent...sorry.