For me it really is more about the intellectual "value" in the design than about any experiential component. To be honest, I don't particularly like using computer to do things. I like making computers do things. As a result, I don't run any commercial software and very little of the other usual stuff most people run. So what does make a machine interesting to me? There are the architectures that I like, that make me feel an intellectual connection to the creators. These are ones that are clean or elegant or orthogonal, or whatever characteristic is particularly nice about it. These are the machines like the various models of the PDP-11. On the other hand there are those architectures that are far enough out of the ordinary that they're interesting mostly in just how different they are. The Intel 432 and the older AS/400s discussed recently fall into that catagory. In some cases it's the details of the design, like the aync PDP-10s and pretty much anything designed by Cray. Maybe it's interesting because of its historical influence like with the GE645 or the PDP-7. So the experience of using it is one of the least significant factors in it for me. Not to say it's negatively correlated. Some of my fondest memories are of low level stuff on PDP-11s, but most of the machines I'd love to add to my collection are ones I've never had any experience with at all. For me, the "modern hardware" in the sense of PCs are uninteresting because they fail all the tests for interestingness. The architecture is something only its mother could love. But it's not so far out in left field that it's interesting in its own right. While the Intel and AMD guys have done a bang-up job squeezing performance out of a turnip and I'm sure some of the chip level details are interesting, the design of the overall machines is singularly boring. Now some people might think there is some significant historical value in the PC per se, but I've got to say I see it just as another step on a road toward smaller and smaller devices that will eventually approach the ubiquitous computing model Xerox talked about a while back. I can't honestly say that I see much intellectually interesting about the fact that it put computers in the hands of the masses. Sure it lets more people do more with them and I can now keep in touch using e-mail with people I couldn't 20 years ago, but I don't find that an interesting subject of study. It's kind of like the web. When things first started happening, one could recognize it as basically as networked, graphical 3270. The only thing that really seemed different and original was the fact that now the recipient was responsible for the format of the information instead of the producer (by virtue of HTML being a markup language not a layout language). But then people came along and made everything a table whether it was tabular data or not. "I don't care how wide your browser window is; these have to be next to each other." Then came style sheets and we lost the one thing that was really interesting, the chance to see what would happen with the one major change to the printing press model. So now the web is just another mass medium driven by advertising. It became less interesting as it became more "useful" and "friendly." If you've been skimming to see what the punch line is, it's this. My question is always, "Is it interesting as an object of study?" not "Is it interesting as a tool to use?" As always, YMMV BLS