*gulp* 😳
Conversation
This is roughly equivalent to “dogfood your software and talk/listen to your users.”
It’s good advice. Is it all that surprising?
1
1
I draw two distinctions from the adage:
1. Often early adopters do not themselves have any serious context of use—they just like new tools.
2. Many technologists trying to invent tools which augment science, math, art, etc are too far outside the domain to dogfood effectively.
1
2
16
(and of course, the excerpted note is in part in reaction to academic HCI, which usually fails to properly internalize even the simple version of the adage you mention)
1
5
do you think apple-levels of secrecy/team fragmentation (which I’m by no means a fan of) defy this? I guess iPhone was so personal it didn’t need domain expert involvement (people who use cell phones) but, say, FCPX?
1
1
Actually, I think Apple illustrates this really clearly. They tend to succeed iff the team is itself serious users. iPhone is designed for the iPhone team. FCPX's key abstractions were designed by an obsessive amateur filmmaker. But then: Ping; Sherlock; Preview's PDF viewer; etc
Another angle is that this is the software equivalent of “write what you know.”
You can write about stuff you don’t know, but it’s more difficult, requires much more research, and is prone to simple errors that will quickly out you to an expert.
2