archive.Joe Lamantia.com

Over at 37Signals, just before the new year started, David made the prediction that by the end of 2006, “Enterprise will follow legacy to become a common insult among software creators and users.”

I think this is already the case, unless the people you’re talking to earn their bread and butter by doing something related to enterprise software – but there’s interesting ground here that I’d like to explore for a bit. On the 37signals site there were some good comments to Dave’s posting – from developers, entrepreneurs, and quite a few other perspectives – but no one made the connection to Conway’s Law, from Melvin Conway’s “How Do Committees Invent?”, which I’ll quote here:
“…organizations which design systems… are constrained to produce designs which are copies of the communication structures of these organizations.”

A good example of Conway’s Law in action is PowerPoint. As Edward Tufte says in Metaphors For Presentations: Conway’s Law Meets PowerPoint,”The metaphor of PowerPoint is the software corporation itself.” [Aside: As a hard-working consultant who spends *waaaayyy* too much time creating presentations to use as discussion vehicles when instead a direct conversation between relevant parties is by far the best use of everyone’s time and money, I can’t say enough good things about Tufte’s campaign to remind the business world how to communicate clearly, by avoiding PowerPoint unless it’s appropriate…]

It’s no surprise then that ‘enterprise software’ as it is installed and configured in many large corporations is generally massive, anonymous, byzantine in structure and workings, indifferent or hostile to individual needs, offensively neutered in all aspects of it’s user experience, and often changed arbitrarily to align with a power calculus determined by a select few who operate at great remove from the majority of the people who use the environment on a daily basis. After all, that is the nature of communication in many large (and quite a few small and medium sized) corporations.

That enterprise software is bad – excruciatingly bad, if you’ve tried to enter expenses using a generic installation of PeopleSoft or Siebel – is hardly news. But it is interesting that David from 37Signals, Peter Merholz of Adaptive Path, Jared Spool of UIE, and many others who are less visible but still important in directing the evolution of the Internet, would all say in one form or another that they see enterprise software as on the outs.

It’s interesting because I think it highlights a shift in the realm in which the Web community sees itself as relevant. If there were ever a potential enterprise platform, it is the Web – the new Web, Web 2.0, whatever you want to call the emerging information environment that is global, ubiquitous, semantically integrated, socially informed and / or collaborative, architected to provide readily consumable services, etc. But aside from occasional bouts of megalomania, and potential success stories like Salesforce.com, the enterprise realm has been pro-forma outside the boundaries of the possible – until now…

Will enterprise software die? Not right away, and not totally. Remember, there’s A LOT of big iron happily humming away like WOPR in data centers all over the world that will run the enterprise apps we all know and detest for many years to come. More important, let’s keep in mind that enterprise software is really just one part (the installable and configurable software part) of what is easiest to describe as a way of doing things. It’s a reflection of a command and control, hierarchical viewpoint on how to achieve business goals through standardization. That way of doing things comes from a way of thinking. Which comes from a type of organization that will (of necessity?) be with us for a long time.

But the new stuff, the things that new school CIOs and CTOs will commit to, will likely be very different in origin, manner of working, user experience, fundamental assumptions, and capability. It will come from different kinds of organizations; leaner and more agile multi-disciplinary systems and environment design consortiums or aggregates, perhaps. This matches well with some of Jared Spool’s observations on the nature of organizations that create good designs, from his keynote address at UI 10 last fall.

Closing the circle, Conway confirms what these creators will look like; “Primarily, we have found a criterion for the structuring of design organizations: a design effort should be organized according to the need for communication.”