I am not a data viz designer, but I've read VDoQI and I enjoy following the field as a lay person. I build a lot of charts but I don't design them. A lot of my

I am not a data viz designer, but I've read VDoQI and I enjoy following the field as a lay person. I build a lot of charts but I don't design them. A lot of my clients have philistine data viz taste--and many of their graphic designers too, even the ones whose graphic design taste is good. By Tufte's criterion of information density, their ideas are a total failure: streetlights, speedometer-like gauges, etc. But often they just don't have that many numbers to show, or the numbers have few dimensions. Maybe in that case Tufte would recommend a plain old table, but that isn't always helpful for "at a glance" reading, especially if there is a lot of non-numeric data too. I'm starting to think that Tufte is really excellent inspiration when you have a "hard" data viz problem, but he might lead you astray when you have an easy one. His crusade against chart junk is probably relevant any time though. :-) Anyway, I'm usually just the implementer, so I'm curious what real professionals thing about all that.