Is there a better term than “architectures of control”?

Chapter 4 from Lawrence Lessig's 'Code and Other Laws of Cyberspace'

Welcome, readers from Metafilter and

One point raised in the Metafilter discussion is whether the term ‘architectures of control’ is a sensible one for this phenomenon, and whether ‘architectures of control in design’ is a good title for the blog. I understand the issue; it’s something (clearly) I considered at length when starting my research. It’s not an especially succinct title, and the use of the ‘architectures’ term is potentially a source of confusion (or irritation) if the link between the design of environments and the design of products and systems is not fully appreciated:

“Architecture is the design and construction of buildings… The noun is never pluralized, nor ever used as a verb, gag, except by the designers of computer software and hardware who needed to appropriate the term because they wanted to make their jobs sound more impressive. This kind of business school speak – always reaching for the most portentous word available when a simple one would do the job just fine – drives me nuts. That is one reason I have difficulty with the title of his blog.
The second is that it is a redundancy: Design of control in design.”

First time I’ve ever been accused of business school speak! But the term ‘architectures of control’ has been in reasonably wide use for a while before my research; from my own point of view, I originally borrowed it from chapter 4 of Lessig’s Code & Other Laws of Cyberspace, as the central thesis here is pretty much that Lessig’s ‘code is law’ principle – relating specifically to the way the internet is structured – applies equally to any product, system or environment with which a user interacts. Anything can be designed to enforce and restrict behaviour. Applying programming analogies to hardware, or architectural analogies to software, or other combinations, can be a useful way of allowing different disciplines to understand each other. Or so it would be nice to think!

But is there a better term than ‘architectures of control’? I’m completely open to suggestions.

Update (19th Sept): The term apparently has enough currency for eBay affiliates to buy Google Adwords using it, e.g.

Buy Architectures of Control on eBay!

…but then they’re not always noted for the most sensible key-phrase choices!


  1. R Brown

    The title is perfectly clear. The use of “architecture” that the complainant dislikes is based on an accepted definition, and, given that the blog frequently discusses control systems embedded in digital technologies, using a word that has special resonance in the field of computer science is justified. Though “architecture” may be used to the point of abuse elsewhere, it is not “business school speak”.

    You could omit the “in design” if only because “design” has so many specific senses that people might misconstrue it to mean a particular field of design at first glance. But it is not redundant if one understands “architectures” in the proper sense. And it reaches out to people who are interested in design in all its many forms. I would leave it.

  2. I thought the original title was fine, but dropping “in design” would work well also, IMO. Your URL will still jibe with the title if you keep the word “architectures” around.

  3. Flashman

    Sorry about all this – I had no idea I was opening such a can of worms, that pointing out the inconsistencies in the title – at least to an average reader of non-technical English, would prove so controversial and lead to me being not only despised, but villified. I’ve had a rough couple of days on ‘the Blue’ as the cool kids call it. First I personally offended Dave Gorman – how was I supposed to know he would come and read my description of his ouevre as ‘aw shucks schtick?! -(but that at least led to an interesting correspondence between us and some raprochement) and now this – the wrath of ‘systems architects’ being wrought upon me from all sided.
    I guess I should just be a bit more careful with my own cruel barbs.
    But still, doesn’t ‘Architecture of Control’ have a bit more zing? Does it really need that S?

  4. The title is perfectly fine, and I think the commenter is off on the statement that ‘architecture’ is never pluralized and only applies to buildings. I think the title describes what you talk about perfectly, so long as the reader can avoid political and ideological attachments to the words ‘architecture’ and ‘control’.

  5. I don’t really have a problem with the title, it is after all just a title. If you do want to get into the nitty gritty with the words, something like “constructs of control” or “structures of control” might be slightly more accurate. But really does it matter?

  6. robin

    Is this not a Forcing Function or Behavior-shaping Constraint, as described by Donald Norman?

  7. Dan

    Thanks for all the comments, everyone. I think the balance of them are in favour of keeping something along the lines of the original title but maybe making it clearer exactly what’s being talked about, especially with the vastly increased numbers of visitors in the last few days (part of the problem in this regard is that the boundaries of this ‘field’ aren’t necessarily as clear-cut as we might think – as a forthcoming post on ‘natural’ architectures of control will make clear).

    As such I’ll be making some modifications to the blog design, masthead, etc, in the next few days, probably also tidying up the links, making the categories clearer, and integrating the ‘pages’ more closely with the posts.

    @Anonymous: Which word in the title is a pronoun?

    @Robin: Forcing functions, interlocks, etc (and poka-yoke) as Don Norman describes are of course an important (and usually user-beneficial) subset of architectures of control – see ‘Everyday things and persuasive technology’ and ‘Simple control in products’ – but a large proportion of the examples on this site go beyond the usability/safety-enhancing intent that Norman implies in his examples.

  8. Sorry for the late interjection. I started this a few evenings ago and then had to pick it back up.

    I like the term ‘architectures of control’ for it’s alarming qualities which certainly have an immediate impact. It reminds me a bit of ‘information architecture’, another modern term, but with a far more Orwellian flavor.

    However, control is a strong word and things like “pig ears”, for example, don’t actually control skateboarders – but they certainly are effective inhibitors which challenge skaters to either play along or try harder in order seek ways to subvert the intented instruction that they are expected to interpret from the design. Skaters most likely adapt, and rather than slide such rails, do tail stalls or other tricks that leverage the space between the ears. In the end, they aren’t necessarily controlled, they just aren’t doing lengthy slide tricks.

    So I thought it would be fun to take your challenge to think of alternative terms. What seemed most familiar to me was the practice related to industrial design that focuses on creating a working product and is typically referred to as “functional engineering”. Therefore, I got to thinking of some terms and the following came to mind:

    – ‘restrictive engineering’
    – ‘regulatory engineering’
    – ‘supervisory engineering’

    These terms seemed to allow for some power still to be had by those interacting with end products being highlighted by this blog, as opposed to the powerlessness that any ultimate ‘control’ may seem to imply.

    Then I was reminded of the iconic “OBEY” ( effort and that make me think of my favorite possibility:

    – ‘obedience engineering’


  9. Pingback: Product Design Aimed at Limiting User Capability at Jim Lipsey

  10. Pingback: What’s happened to the website? at fulminate // Architectures of Control

Comments are closed.