Sig comments on the "technology vs culture" thingumy:
Software in general represents/is a model of the real world.This would explain why Sig's company is also called "Thingumy". Anyway, so I'm having a good ol' time trying to figure out what I'm doing in the "Middleware" market, besides looking rather confused.And enterprise software is a bit particular in that it models some other model: Management theories, marketing, hierarchies and a few others. Accounting software is conceptually built for how it's done in the finance department, CRM is built to support the current (and old) ways of the marketing department and so on.
What then if the model the model is based upon is wrong? Would then not the software-cementing-culture be highly unproductive, even inhuman?
Time to revisit the underlying models I say. Let the technology solutions follow.
[NOTE TO SELF:] Stick to cartooning. This is so out of your league.
Posted by hugh macleod at April 28, 2005 4:20 PM | TrackBackC'mon Hugh, you're most definitely in the right league: Cartooning is conceptualizing reality, so tries software - only diff is one is fun and better at it, the other is..eh.. seen as boring :-)
Posted by: sig at April 28, 2005 4:50 PMBecause technology is an amplifier, whenever you automate a screwed up model, all you do is create a faster, more efficient screwed up model, no? Revisiting the underlying models, however, is so conceptually and emotionally taxing that almost no one ever really does it.
Posted by: Tom Guarriello at April 28, 2005 4:59 PM"Middleware" is highly relevant to the whole long-tail discussion, so it is definitely worth thinking about.
Look at it this way: “Middleware” is conceptually a continuum that lies between 100% technology and 100% labour (e.g. what is the RIAA if not some ossified labour-centric “middleware”?). Technology obviously offers much higher leverage than labour.
The long-tail connection is how the labour/technology ratio of “middleware” is plummeting, making it economical to service heretofore unserviceable markets. Look at any prospective long-tail market and if you see a high labour/technology ratio you should turn and run away, screaming for your mommy.
As far as Sig’s point that most software is simply a model of a model and the underlying models often suck, well, welcome to the real world of computing. Literally since the instant computers were discovered, people realized that any application would suffer from the garbage in, garbage out problem.
The issue that Sig is putting his finger on is that by the time large, complex, monolithic chunks of domain expertise are codified into large, complex, monolithic chunks software and then widely distributed, it is often too late to undo the damage. The switching costs are simply too high.
Couldn't agree more witrh Sig. Enterprise software is "electronic concrete" poured over business processes that increasinglky will have to flex and change more and more, as the electronic grains of sand represented by microcontent and links keep on eroding the concrete.
Big integrated systems often (usually ?) still follow the blueprints of the organization's hierarchical top-down org charts more than is responsive or friendly to customers, and by extrapolation, the markets made up of those customers.
Posted by: Jon Husband at April 28, 2005 8:49 PMBTW .. don't agree you're out of your league. I think you're more like the artists who use the "art naif" genre, often cutting through to the essence in light-hearted ways. keep it up .. it used to be about technology and the technological infrastructure .. more and more, it's about the sociology, psychology and employing truth instead of smokescreens.
Yo understand what's going on .. it's the big shakedown at the OK Corral between accepted management/organizational science and theory and the dynamics of interlinked real-time sociology. There's a reason why blogging has grown so rapidly, and continues to grow .. it's real and it works.
The trouble with playing in the "right league" is that you kinda have to fit in or fuck off, reciting and regurgitating the conventional wisdom for mega-bucks (most of which go to the partners who created the brand that proffers the advice which gets bought). This is both the mainstream advertising and the mainstream consulting game, no?
Typically, although not always, those partners made it to such elevated staus on the back of ideas and concepts that as often as not are either obsolete, fading in significance or just plain farked but got bought a lot by previous clients.
Posted by: Jon Husband at April 29, 2005 3:04 AM"Art naif" or court jester? - only the fool can tell the king he's naked.
I wholeheartedly agree with the thought that monolithic software can only, by definition, represent an out-dated model ("Any program, once working, is obsolete"). Where the middleware (and I don't agree that it is boring BTW) fits in is making it possible to fit changing modular pieces together to support (create?) new business models which disrupt the integrated, monolithic models currently extant - check out Clay Christensen's thoughts about this: http://www.claytonchristensen.com/.
Open source software helps here because it is increasingly possible to get "free" (as in speech, and often as in beer) middleware, making it possible for the 'long tail' programmer (like Alan Gutierrez http://engrm.com/blogometer/2005/04/11/living-the-long-tail.html
to fashion systems from the building blocks scattered around the internet.
Good points, Ric. Looks like the future to me ... plus,. accordinmg to IBM advertisements, it's invisible ;-)
More wirearchy raw material, surrounding and penetrating us (figuratively) and what we do (literally)
Posted by: Jon Husband at April 29, 2005 4:42 PM