Link This | Email this | Blog This | Comments (1)

Howtheheckareyougoingtosearchforthis?
July 18, 2008

I just received a nice new book on digital work flows. It covers all the bases: how to plan, how to take photos, how to transport the photos, how to download the photos, and how to archive them.

In the middle of the chapter on how to archive the photos, the author tells us: “Avoid the dreaded word space!”

He argues that all file names should be concatenated into onewordjustlikethis, or he suggests – if you must – use_the_underscore_character.

That’s just plain stupid. Allowmetomakemycase.

Where is it written that we must continue to hold on to the anachronistic requirements of a long-dead operating system? The original brain-dead filename which later became known as 8.3 (eight characters followed by a period and a three-letter suffix) was Gary Kildall’s CPM. Kildall’s predecessor to this was a simple, but effective operating system called PLM. He wrote CPM to expand on the capacity of PLM, adding control for the early floppy disk devices. Kildall was a brilliant mathematician and computer scientist who brought full-fledged operating systems to microprocessors.

CPM was a reasonable operating system back when we were actually counting the number of characters in a file in order to avoid running out of RAM in a file buffer. I tried CPM, but never liked it or the computers on which it ran; and even in the late 1970s, when CPM was in-vogue, the 8.3 file name concept was lame.

Kildall likely adopted his 8.3 file naming structure because of memory limitations on the processors he was using. He would have used more (see UNIX, below), but there wasn’t enough room in some buffer to hold more characters, so he shortened it to 8.3

It seemed stupid back then, and it’s still stupid. Unfortunately, in the famous race to get the attention of International Business Machines, Kildall (read the story on Wikipedia – it’s amazing!) lost his bid to become “the” operating system, and Bill Gates won. Even more unfortunately, Bill Gates’ operating system, called DOS for Disk Operating System, used the same stupid 8.3 file name protocol.

Apple’s original operating system used something at least as bad (I don’t remember, as I have purged that part of my memory) and UNIX had an only slightly better format, that being 16.4.

All of these now-dumb naming schemes were of necessity. The disk operating systems of the time (operative word: disk) “required” them (because some engineer decided so), and we consumers of computers and operating systems had to live with them.

Decades later, the World Wide Web was developed along the lines of UNIX (the underpinnings of the Internet are UNIX-based) and so we continue to be hamstrung by file naming for the Web that requires the UNIX 16.4 file structure, which is a structural requirement, but it’s dumb.

So, where does that take us? People think that file names MUST be named with concatenatednamesthathavenowordspacesbecause... somebody in 1973 couldn’t get a larger file name into a memory buffer? That’s just brain-dead.

Concatenating file names for any purpose other than putting them into HTML structures for a web site is unnecessary. Search engines have to work to find strings, and search engines are the key to our success in today’s media-rich world. But, to a computer, finding a needle in a haystack is no problem, so finding one word embedded in the middleofastringofotherwords is really no big deal. In other words, a word space, though it comes in handy for human readers, is of no consequence to a computer search engine. Looking for a word in a string, or a word separated by word spaces is the same task, to a computer more or less.

And there is the crux of my thoughts today. Though it may be simple for a computer to find that needle in a haystack, why force the human operators of computers to do the same? We are human readers, and we really need word spaces! I search for photos every single day of the year, and my archive of photos and files is now at 716,460 files, folders and documents (I checked).

Searching for, and finding any file, photo or folder requires two things: file naming discipline, and the use of file names that are searchable and find-able. If you are looking for the term Boy Scout, it’s much easier to search for it with either Boy or Scout. If you concatenate the words, you’ll probably have to search for it longer, and you have to try three times until you get it. And, successful search is not a lark, it’s a requirement for business success. Time wasted looking for concatenated words is money lost. Make your life simpler, and just use words! They are easier to read when they have spaces between them. Computers don’t care, but we should.

I’ll continue this rant this weekend.

Posted by Brian Lawler on July 18, 2008 | Comments (1)