Distributed Systems and the End of the API (meta)

I just published a written distillation of my talk at PhillyETE 2013, Distributed Systems and the End of the API, over at the "writings" blog for the Quilt Project.

Take a look, especially if you are interested in distributed systems, CRDTs, the general suckage of how APIs work, or if you're curious about what this Quilt thing is all about. (Spoiler / hint: the talk+post isn't strictly about Quilt, but is very strongly related.) My inspiration for writing up the content of the talk comes largely from Michael Bernstein's writeup of his RICON West 2013 talk, Distributed Systems Archaeology. Giving talks can be a compelling way to spread information and evangelize different (and hopefully better!) ways of thinking about problems, but depending on slide decks and video dumps is a poor way for people to discover and access that information. Giving it a proper home that is easily searchable and accessible to all (despite visual-, aural-, attention-, or time-related disadvantages) seems to be a total win, especially if one is hoping to have a lasting impact.

My apologies for the piece being so long (~7,250 words). I really didn't want to disturb the single narrative, and so resisted splitting it up onto parts. Hopefully the length does not detract from the message. It's very clear that my biggest failing as a writer remains my verbosity, something I'll be watching closer than usual for the rest of the year. It would not hurt to work on more tightly-scoped, concise pieces, if only to be able to put some bounds on the time spent on the writing itself.