Digesting Thingamy 08:06 on Monday
Last Thursday Sigurd Rinde gave me a virtual demonstration of Thingamy, his workflow application. I mean business modeling application. Err, database. No, reporting tool. Maybe an “enterprise app”? A web service back-end? No, not that either. And at the same time, all of these.
San Fransisco in Lego (photo by Tektsu)
Scalable from a system you would build in 30 minutes (or less, really) to make something silly, such as manage your favorite restaurants, to being a potential SAP killer. And I’m not so much talking about technical scalability as conceptual scalability. As Sig prefers to put it, it’s like Lego (or what Lego used to be before they lost the plot). From the few different building blocks you can build anything from a toy car to a city.
Thingamy talks to humans using familiar concepts. Things of any particular area of (work-) life are created in the system with required detail and then arranged into workflows that represent what actually happens. Not geeky database relations that have a vague connection to the world. The familiarity of the interface doesn’t exactly equal talking in plain English, but the visual Thingamy “language” is learnable in a couple of hours, instead of months. For clarity, there is no programming language to learn. The geek in me could say the idea is related to domain specific languages, though.
The workflows interweave the operation of the system with actions taken by real people on real, but not necessarily physical things. Workflows move in time, can be paused, resumed, edited while the system is running… they can stretch and skip parts… just the way work flows with people: unpredictably (as much as we might like to think the contrary).
All data is stored raw and live. No snapshots of data are needed for reports or anything else. Things in the system can have a memory of where they’ve been, who’s touched them and why. (Now the blogger in me can’t refrain from linking to ThingLinks.)
Thingamy doesn’t fight complexity by arranging things in complex hierarchies but instead makes every web 2.0 head happy with full tag support. Select multiple tags to filter and find anything. I think it’s time to expand the conversation on enterprise tagging out of the bookmark cloud.
And finally, Thingamy talks to computers too. It does this using a language easy for computers to understand and handle, namely XML-RPC.
Sig, did I get it right? :)