One problem that is likely impeding the development of critical approaches in HCI is equivocation. Equivocation occurs when different meanings or uses of the same word are used interchangeably. “Criticism” appears to be just such a word, as well as the origin of this post was to offer some fundamental distinctions among different uses of “criticism” in the hope of helping prevent this sort of confusion.
Some quick examples shall manufacture the point clear. The following are examples of “criticism”:

A book or film review in a newspaper
A “close reading” of a work of Shakespeare, to explicate its greatness (etc.)
A “close reading” of a magazine ad on behalf of spaghetti, to develop a theory of semiotics (e.g., Barthes in Mythologies)
When a peer offers a critique to a design mid-process (e.g., studio critique)
The utilize of a case study as a deep, representative example
Theorization surrounding an experiential quality accompanying a cultural artifact
Comparison of a given example with a rubric, heuristic, or other evaluative framework
The act of constructing such an evaluative framework out of numerous examples
The development of an overarching explanation of a large group at a given time or place (e.g., postmodernism, Victorianism)
The development of a comprehensive philosophical system (e.g., Kantianism)

One might quibble that I have included an item or two I shouldn’t have, or failed to include an item I should have. Be that as it may, I hope the central argument holds: we utilize the word “criticism” in drastically different ways. And therefore, terms such as “critical HCI,” “interaction criticism,” as well as the like are subject to widely different understandings, except when given proper context. And except when we create that context, we are likely to be mired in confusion.

To help think through this problem, I moved up a level of abstraction to consider how different meanings of “criticism” might be situated in a field of possibilities in interaction design. I represent that field as follows:
The range of audiences (X axis) versus levels of abstraction (Y axis) on behalf of criticism
The X axis represents the primary audience on behalf of criticism, that is, on behalf of whom it is offered. It does not represent the position of critics themselves (who lamentably are not really represented on this graph). It ranges from laypeople to academics, as well as in between are professional practitioners. The Y axis ranges from particulars (i.e., individual artifacts, such as that film, or this version of the iPod, or that novel) to abstractions (such as genre, qualities, categories, as well as types).
It is important to note that the picture I shall develop in this post represents only my take on criticism in interaction design. If one were to explore criticism in film or literature, it likely would like quite different. Much of this has to do with relationships among key categories, such as the academic, the layperson, as well as the professional. In HCI, the professional seems to occupy a privileged role, in that any HCI theory that fails to offer explicit take-home points on behalf of practitioners is generally considered deficient (I am merely stating, not endorsing, this point of view). In literary criticism, by contrast, the distinction between professional practitioners as well as academics, as well as on behalf of that matter, between lay connoisseurs as well as the latter two categories, is much less clear.
So, in the context of interaction design, criticism at the level of the layperson can be charted as follows:
Interaction criticism at the level of the layperson
A review of a piece of software, e.g., in MacWorld or Electronic Gaming Monthly, is written about a particular piece of software on behalf of a lay reader. Thus, it winds up in the lower-left corner. I put Taste in the upper left corner, because of the fact that it gets at how laypeople feel collectively about all of the relevant particulars in their time. The work of Genevieve Bell (or on behalf of that matter, Shaowen Bardzell) exploring domestic spaces in non-Western homes as a prelude to domestic and/or intimate IT design would drop into this category. Maybe “taste” isn’t the best word (disclosure, on behalf of what it’s worth: I was thinking about Bourdieu when I put that there).
With that out of the way, let us now turn to professional interaction design practice. Here’s how I see criticism playing out.
Interaction criticism as it manifests on behalf of professional interaction designers.
These are all presented as ranges, because of the fact that I see lots of flexibility in their application. Across the bottom is studio critique. This is where peers offer one another expert design critiques, as well as where designers offer their posses design rationales. Toward the layperson side of the range are student applications of critique, while workplace critique is professional. (Arguably, this line could be extended further rightward; I can imagine studio critique of avant-garde design–e.g., someone like Bill Gaver or a visionary Ph.D. thesis–taking place in relatively academic settings.)
Informing studio critique are abstract evaluative frameworks, heuristics, etc. These may be more or less abstract (hence the range). Nielsen’s Web Usability Heuristics are fairly general, as well as one can easily imagine heuristics on behalf of Web e-commerce shopping cart usability, on behalf of example, which would extend further downward toward the particular than Nielsen’s heuristics.
I also include case studies in this group. Case studies are interesting. On their face, they would appear to be particulars, so it might be a little mysterious why I positioned them higher on the abstraction axis than studio crit. The reason is that an estimated all of the case studies I have ever seen are accompanied by an at least implicit claim that this case represents something broader (i.e., more abstract). So it is on the basis of this additional claim that I consider case studies to be at least partially abstract. I have seen them used in professional as well as academic contexts, hence the horizonal range.
OK, next! In the top-right corner is abstract/academic, as well as no one shall be shocked to see what lives up there:
The abstract-academic aspect of interaction criticism features theory.
In this I am trying to represent a distinction that I consider to be very important on behalf of HCI. In the extreme top-right is philosophy as well as critical theory that itself has origins outside of HCI, but which is took into HCI via HCI theory. An obvious example of this is phenomenology, a philosophical system developed in continental philosophy by the likes of Heidegger as well as Merleau-Ponty as well as appropriated into HCI by Winograd & Flores, Dourish, Sengers as well as Gaver, Fry, Willis, as well as others. Likewise, Bakhtinian literary theory has been introduced into HCI by McCarthy & Wright, Blythe, as well as others. Each of these HCI writers is engaging in a double-move: On the one hand, there's the contribution of new interaction design theory that has been informed by philosophy as well as critical theory (this is the Interaction Design Theory contribution); at the same time, there's some representation in HCI literature of what this philosophy is. That representation includes Winograd & Flores’ summary of Heideggerian concepts, McCarthy & Wright’s summary of Bakhtin, as well as so on. I desire to stress that the original philosophy as well as critical theory are not represented in my picture, only their summaries as well as representations in HCI discourse. (The extent to which this philosophy and/or critical theory is robustly represented–and even understood–in HCI is also a matter with which we should be concerned.)
I see the Interaction Design Theory as forming a bridge between Philosophy/Critical Theory (which is academic) as well as Interaction Frameworks as well as Heuristics (which are professional practitioner-based). In a way, they are meta-frameworks, in that they offer the criteria out of which frameworks as well as heuristics are made. As XML is to MathML, so is design theory to design frameworks/heuristics.
My posses perspective suggests that one potentially significant area of innovation in HCI research is the downward as well as leftward movement from philosophy as well as critical theory through interaction design theory as well as into professional practice through frameworks as well as heuristics. I do not mean to suggest that this is the only or even preeminent mechanism on behalf of this innovation to happen; but it does represent an alternative to technology driven innovation, as well as insofar as it does that, it creates a human-centered counterbalance to a technologically driven HCI agenda. In that dialogue–between technological as well as philosophical innovation–HCI is surely positioned to do its best work.
I desire to return to the beginning of this post. The question was what are the species of criticism, insofar as criticism is a part of interaction design. As I have tried to show, it is incredibly diverse. To think of “criticism” as “magazine software reviews” or even “studio crit” is overly limiting. Criticism informs the full range of our work, as academics, as professionals, as everyday users. The following is the complete picture. I must reduce it on this screen to manufacture it fit, but if you click it, you should be able to see the full-resolution version.
Criticism as well as interaction design: The field of possibilities. Click to enlarge.