Name games

I have lots of files in my computers. I have lots of folders in my computers. Folders are like files in that they both need names. URLs are like files in that they are names too. I need to name the names. Otherwise, I cannot navigate, I cannot surf. I cannot organize.

Without a name, I cannot put my content in the computer. That is bad. With a name, I can find the content afterwards. That is good. As long as I can remember the name. Names identify things. Identification is the root of all retrieval. Or is it? Hold that thought.

How to name the names? Words. Words are names. I will use words - with some limitations. I need to keep my words short. Otherwise my computers will be unhappy with me. Best to avoid certain characters too. The world of the computer is littered with special characters : periods, colons, ampersands, spaces...

Speaking of limitations, I need to arrange the names so that I do not have too many files in any one folder. If I don't, my computers will be unhappy with me.

Let us pick some words to use. Let us start with those office memos. I could use memo1, memo2 and so on. I could use memoYYYYMMDD_1, memoYYYYMMDD_2 and so on. I could use memo_re_fiscal_rectitute_1, memo_re_fiscal_rectitute_2 and so on. I could use memo_mcgrath_1 or memo_mcgrath_YYMMDDD_1 or...

How to choose? The more descriptive the name, the harder it is to find a way of filing them sensibly. The more numerical (numbers, dates etc) the name, the easier it is to dream up filing strategies but the harder it is to grok what a name means just by looking at them.

Wordy names are hard. Good wordy names are really hard. Good, short, sort-friendly wordy names are fiendishly hard. Wordy names are always trade-offs. Ease of filing, easy of interpretation, ease of typing... Numbers are a different sort of name. I could use numbers. Maybe memo1, memo2 and so on, has a lot to recommend it after all?

Hmmm. Maybe it is just too darned opaque to go with meaningless numbers? I could add meaning but still use numbers? I could allocate number ranges. Memos 1-100 for Mr X memos. Alternatively, memos 1-100 for financial subjects...What happens when a range fills up? memo100a,b and so on. That gets ugly. Better to start with large ranges. But, then file names will be long from day 1, like memo0010001. Why not just ditch the padding on the names? memo1, memo100 etc? Darn, now the natural alphabetic sorting of files and folders is all messed up.

I can code anything with short sets of letters and numbers. Subjects, author names, dates...All I need is a coding scheme. Dewey decimal? Custom taxonomy? ISO 11179? Cutter numbers? Colon classification? Hmmm. That is a lot of work and I still end up not being able to grok what a name means just by looking at it as I need to look up the classification tables...

Maybe the best name is not one name? Maybe I need multiple names? Take http://www.tinyurl.com for example. I can feed it a long name and it will give me back a short name that “means” the same thing. But then everything is opaque which is best avoided isn't it?.

I sense a circle and I'm going around and around it I think. Time for an outrageous thought. Maybe the trick is to avoid naming the name at all? Can this be done? What would it mean?

Imagine an operating system with no file names, no folder names – just a machine generated, uninteresting, opaque number allocation scheme. You throw content in, you get an opaque, unwieldy number back out. How do you then find your stuff?

You search it. Is a file name anything other than a query in disguise? “find the file X where X.name = something”? Bertrand Russell thought so. Would he approve of the concept of a search engine? I suspect so.

Introspecting for a moment...The last time I was on amazon.com, or rte.ie, or techcrunch.com How did I get there? I launched Firefox and typed, for example, “amazon” into...no not the address bar, but the Google search bar beside it. I did not use a unique identifier – I used a search query. I don't navigate the Web via names. Do you?

Are we all becoming more and more used to the idea that naming things is too hard/ineffective/impossible? Are we slowly morphing into a race of finders rather than namers? Is this forward or backward movement. Sloth or smarts?

What odds on the next killer OS being based on a search engine abstraction rather than named files/folders?

Pretty good I think.

I would hate to have to give it a name though.

For all sorts of reasons.

Related:
ITWorld DealPost: The best in tech deals and discounts.
  
Shop Tech Products at Amazon