Replying to @Ethan Gruber's post:
I have to say, I disagree with most, if not all, of this. There is plenty of coding work in DH that is not web-related. Text-mining and distant reading are examples, as is work encoding texts in TEI. I think I remember some very awesome visualizations of ancient architecture, too, that I would call non-web DH coding? :)
The thing that I disagree with most is the idea that we (counting myself as a digital humanist) need to rethink our notion of the Digital Humanities. Like the #transformDH hashtag, it is predicated on a set idea of DH that I do not think exists. It seems like less than a year ago that there was still very active discussion about the role of code in DH, what skills could/should be part of it, and how it all fits together. No one, I think, ever came up with a notion of DH that had widespread consensus as being a good definition. Now, it seems, we are chasing after responses to a non-existent notion.
That said, I do see the existence of two branches of DH, with different histories. In broad strokes, one is the "Humanities Computing" branch, on is indeed the more web-oriented branch (which shares more in common with ed tech). This has been recounted many times before, sometimes in this very site.
I'd ask that we all recognize the youth of DH, and this it is -- whatever it is! -- still developing through informal channels like THATCamps and blogs and twitter conversation, and through the institutional decisions made by administrators and through cluster hires. In short, I think it is far too early to say that DH is far too focused on textual disciplines. It might be at present, but the ultimate constitution of DH is still very much to be determined. Awesome visualization and interpretation of data -- textual, architectual, visual, geographical, temporal, archaeological, and more -- through digital/code-based means can and should be part of DH. I consider it to be implicit that it _is_, but maybe hasn't had as much attention as other aspects. Seems parallel to the Theory and DH questions that way?