ietf
[Top] [All Lists]

Re: from the horse's mouth

2005-11-02 18:54:56
At 18:08 02/11/2005, Dave Crocker wrote:
Very much so. In my opinion, a very high percentage of security problems are due to UI failure as well. (When I was an AD, I tried to get an evaluation of one (IMHO ghastly) document from a CHI expert. But he had no time, and said that the general area had never been studied due to lack of time.)

Mix variable system performance, with cognitive psychology, with bell-curve statistical distribution of behaviors, and the challenges for good usability design make QOS-based routing protocol design (and operation) look like a kindergarten exercise.

It is one of the very good reasons that the IETF has done well to keep almost completely away from that realm. As a community, we lack the expertise, and even the experts often lack the expertise.

The idea that random folk, like technical writers would have any particular clues about Interaction Design further highlights how far we are from wanting to deal with this topic.

how true you are. RFC 3935 preserves us against such errancies (principle of competence). Hence my surprise and my determination when the IETF wants to take the world's lead in the languages identification area, being used as a reference to influence the specialised ISO efforts. Also that no one at IESG rises question on the IETF competences and addresses its resulting responsibilities.

I would look a fool to suggest IETF ventures in music, education, en-commerce or maybe Olympic issues? Languages and cultures are much more complex issues. We have the experience of the IDNA failure. When we run into areas out of our domain of competence, the IETF runs into the increased risk of issuing political oriented documents without realising it. This is why I call for the restauration and the help of the ISTF/ISSG.

jfc


_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf

<Prev in Thread] Current Thread [Next in Thread>