- From: Karen R. Sollins <sollins@lcs.mit.edu>
- Date: Fri, 11 Aug 1995 14:15:01 -0400
- To: weibel@oclc.org
- Cc: uri@bunyip.com
Stu, I believe that an alternative position one could take is that humans should never see those things. In order to engineer something properly (like my car) there are a great many components that I as a mere mortal will never see and would not be able to do anything with if I could. That doesn't mean they shouldn't exist in the form their in or that they should be friendly and accessible to me. It means the engineers did something right if it works efficiently and addresses my problems. We as systems builders also create a number of things that normal humans should never see, and URIs should fall into that category. You can be sure that I don't want to have to explain to my 88 yr old father all the components of telnet, SNMP, HTTP, in order for him to be able to log in somewhere, send me mail, or browse the Web. He doesn't have to know about IP numbers to do those things. If he has to know about the locations of objects in order to use the web of the future we will have made a giant mistake. It sounds to me like RandTriever had some basic systems flaws. I'm sorry that the library built their system around it, but that doesn't mean that we should continue pay the penalty into the future. No person should ever transcribe any of those numbers or identifiers; humans are notoriously bad at doing that. I am very much in favor of human friendly naming schemes, in fact lots of them. Human brains are twisted on in many different ways and humans do lots of different things. So, there is probably a need for a wide variety of people friendly naming schemes. In fact, I've spent many years thinking about and working on such problems. But, we are talking about something below that here, at that lower level, please lets distinguish who OUR clients are. They should be programs, not people. Karen
Received on Friday, 11 August 1995 14:15:01 UTC