Via Stefan Tilkov, some observations on closed vs. open source software from Jaron Lanier...
A closed-software team is a human construction that can tie down enough variables so that software becomes just a little more like a hardware chip—and note that chips, the most encapsulated objects made by humans, get better and better following an exponential pattern of improvement known as Moore’s law.Yeah, and we're seeing so many innovations occurring on closed software today, compared to open software, it is frightening. (insert "sarcastic" emoticon here)
Even the iPhone is based on an open source Unix. I don't think Jaron is being fair to compare the iPhone to Linux. Maybe compare the iPhone to Android-based systems in a few years. That would seem fair. Even then I would expect some of them to have various "closed" aspects.
And let's not take that hardware analogy too far. I think it was Edsger Dijkstra who observed about 30 years ago or so that CPUs iterate over the same design but different process technologies. That's more or less true still, but where it is less true today what also is true is that the design complexity and design approaches are more like software than they are like the hardware of yesterday.
Penultimately, let me note from personal experience that with either closed or open source, the challenges of getting "better and better following an exponential pattern of improvement" in software is damn hard. I've not seen many systems that do that well, but the best candidates at this point seem to be open in various ways.
Finally Jaron's entire debate does not seem worth having. Questioning whether closed or open source is "the right" model is a false dichotomy, a source of info-pollution. The answer is clearly, from observation, somewhere in the middle. Sometimes a little this way, other times a little that way.
1 comment:
Nice. For the better part of a decade I've been mystified by why Jaron Lanier gets any play at all. He's the sci/tech equivalent of political pundits: a Very Serious Person talking about Very Important Things. All of this launched from some VR work for which I'd love to see any real code or output. If we're looking for public intellectuals in computing, we can do much, much better. Someone should give Alan Kay a regular column, for example.
Post a Comment