The Obvious Kirkyan Inversion

Posted on Saturday 16 September 2006

With UbiComp 2006 starting Monday, I’m curious if there will be panels discussing transreal Things similar to kirkyans. I’m expecting there will be, but I’m wondering why no one (at least on the sites I visit) is openly discussing something that seems rather obvious; why it is I don’t read anything about an inversion of the kirkyan concept. There’s lots of talk about blogjects and spimes, but not much more it seems.

The closest thing I’ve read is in Sterling’s old keynote address at Siggraph 2004 (Link). He hints at it, but in the mess of converting Rashid’s superficial “blobject” into something with real meaning, he more or less winds up taking a shotgun approach primarily centered on physical objects and industrial manufacturing without providing much more than what some people had already been discussing. Read the following excerpt:

It is born digital. Its design specs accompany it throughout its life. It is inseparable from that original digital blueprint, which rules the material world. [ed - to keep this in context, the online discussions preceeding this centered around the use of RFID tags to make recycling easier; hence the need to have the "original digital blueprint" associated with the object at the end of its lifecycle so that it could be disassembled easily] This object is going to tell you – if you ask – everything that an expert would tell you about it. Because it WANTS you to become an expert. If you become an expert in wrangling that object, then, just like the gurus of SIGGRAPH, you will contribute to the advancement of the industry. The object will evolve faster, the industry will evolve faster. [ed - note the emphasis on "industry" which I'll get to in a moment.]

Yet even the pellets from a shotgun blast can hit the bullseye, and on some things he gets close. That’s why I imagine there must be someone who has either proposed something like an inverted kirkyan or written a story that includes one; I’m just unaware of it.

For those who’ve not come across that unnamed concept, let me start by saying that as previously mentioned, kirkyans came from thinking about niche products – not mass-produced stuff. Sterling’s approach or “process” begins with Rashid’s “blobject” designs (the word itself is attributed to Industrial Design professor Steven Skov Holt) and his apparently learning that from the CAD side of things, for many – though not all – manufactured goods, soft forms were more or less impossible during the 1980′s… something I’ve previously pointed out when talking about design trends (as in a comment on the Ishush blog – Link).

Sterling is coming at this from an old school, institutionalized Industrial Design point of view; the high-volume production side of things. Industry. However, in spite of his enthusiasm and recent experience, Industrial Design is dying and has been for years. The future many people imagine has much less mass production and significantly more custom product (as opposed to consumer-modified Gizmo’s to which he seems to be referring at times).

Here’s an image from a Design Trends powerpoint I gave back in the summer of 2000 discussing a return to “craft” (I’d previously mentioned it; note those are old Nike ID web images in the top part):

trends2000slide32

The above image is from the Transitional part of my presentation; the closing thoughts part of it discussed “Smart Products” and returned to the opening statement:

Manufacturing processes have limited shape, texture, and materials. However, as processes become more capable and complex, products will increasingly become more sculptural and craftlike… [where form is] driven by “Branding” [ed - i.e., whatever shape we want it to be.]

Forecasting a niche future more than six years ago is why when I coined the word “kirkyan”, there was only one physical “master” and one or more virtual doppelgangers. Sterling, while lecturing at ArtCenter, would have been immersed not in niche product development (although there is some of that, of course), but in high-volume product design which is what is still taught to ID students. He would likely have been confronted by and thinking about Industry and as a consequence looking at how to recycle all that stuff he was seeing being designed around him. Why do I believe that? Simple. He’s the monarch of the Veridian Movement which has as it’s focus the Greenhouse Effect (Link – see #2).

So what happens if we invert a kirkyan from being a singular physical to being mass-produced in multiples? Well, I imagine we’ll get something that will probably occur somewhere along the development pipeline between spimes and kirkyans (for lack of a word to describe these, I’ve taken to calling them “Stellayans”, inspired by – what else – an old Star Trek episode that has multiple androids originating from the same source file/mold – Link).

For me, a stellayan is very similar to – though not quite the same as – manufacturing CAD inside the 3D virtual space of the development application. This is a singular virtual object but it spawns multiple physical objects. I do that all the time as do many designers/CAD technicians. The significant difference is that the virtual Thing is “smart”, and the physical versions act as inputs to the virtual master so that it can evolve and thus produce improved physical versions. Hopefully it uses previous product through a reclamation system, but as there is no one-to-one correspondence between generations of product, the recycling condition imposed on kirkyans – and spimes for that matter – doesn’t apply (i.e. a first generation stellayan Thing might fab 10,000 units, but after receiving inputs and modifying its design, it may experience increased demand and fab 100,000 units; and perhaps only 5,000 first generation units were returned for recycling – some users may not want the upgrade).

If someone isn’t doing it already, we could do this now.

For example, imagine that there was one virtual Blackberry that was connected to all the physical versions so that as people were using their devices, their interaction with the user interface was tracked: how many times a button was used, how much pressure was applied to the buttons, aso (this should sound similar but inverted to the kirkyan weapon example I posted recently – reLink). By mapping user interaction from numerous sources, a more efficient interface might be developed which could then be used to automatically modify the CAD model. That updated geometry serves as the 3D template for a new version; an evolving product that is mass-produced.

The interesting thing is that a similar kind of evolutionary product is inside the Second Life virtual world. For anyone who read “The Kirkyan Weapon” entry (reLink), one of the guns in the composite image is a virtual weapon that receives updates from its creator (the auto-update is mentioned in a long list of features; “Just rez or attach a Seburo, and it will check for updates” – Link). That virtual product does not, as far as I know, evolve using inputs from other virtual products purchased by consumers, but it could. After all, the master file doesn’t care from where the input data originates – it receives information in electronic form. So whether that comes from the designer directly, from a wirelessly-connected physical device or another virtual product sending out emails full of data to be parsed is irrelevant. And by the way, I also suspect that it doesn’t change its shape during an upgrade, but that too would be easy to do since production in Second Life is the code.

If we don’t have such a Thing yet, it’s difficult for me to believe that we will not have it very shortly. It may turn out that the Blackberry’s firmware might be upgraded based on the inputs from users in the field. And when you think about this from that perspective – the thought that physicality and information are the same – it’s hard to believe it’s not already being done somewhere; maybe within a specialized field. That’s why it’s so easy to give this idea a really silly name – there must already be a word to describe this.

And to close this long entry, I’ll finish with one last image link in recognition of the new and improved original Star Trek shows coming this Fall… a loving reminder of Harcourt Fenton Mudd…!!

  1.  
    9/25/2006 | 3:17 pm
     

    Sven,
    I always imagined kirkyans to be what you’ve here described as stellayans. Kirkyans never seemed to be limited to any particular number of physical instantiations — never seemed necessarily singular…
    W

  2.  
    9/25/2006 | 3:58 pm
     

    … except as defined. The kirkyan idea was deliberately approached from a niche perspective since I was being specific in the same way I saw others getting specific – first “spime” and then “blogject”.

    When I read the definition of a “spime”, I considered it unnecessarily limited and suspected Sterling was both focused on a particular use (as mentioned above) and, at that time, still not entirely versed in CAD and RP fabrication technologies. The stellayan is more obvious and is where I thought “spime” should have gone. A shame it didn’t.

    Thing is, I’ve not come across any discussion of it either, which seems extraordinarily odd especially since this general idea can be traced at least back to Gibson’s “Idoru” trilogy (I think I mentioned previously how before finishing the second book in the trilogy I was certain how the third would finish). Gibson doesn’t give us anything more than an AI that replicates en masse, but there is not, to my recollection, any suggestion of a feedback loop. But someone must have written something that took that step. I’d like to find it. I suspect it’s in some work of fiction, but perhaps not.

    And by the way, there is, of course the next obvious version: a combination kirkyan-stellayan. Right now that just sounds like one of Bill Joy’s nightmares.

Sorry, the comment form is closed at this time.

Close
Powered by ShareThis