brainstorms

Tuesday February 15, 2005

open source beginnings...

through a series of loosely related events and text, i was flooded with the thoughts of free and open source origins of my computing life at york university. no, it was not linux, it was not something from MIT or Berkeley, and not a part of the GNU project. they all came later. this is a remarkable bit of hardware-specific open-source engineering that all the hack "historians" tend to overlook:

David G. Conroy's DECUS C compiler.

this was the first C compiler i ever used for hello, world. i did not have a copy of K&R at the time [my first K&R is dated 82], but a printed copy of the C reference manual and Kernighan's C tutorial. i think this compiler came to us with the 1980 RSX11 sig tape or the Torlug tape. it ran fine under our 11/780's compatibility mode and remained in use until DEC's native C compiler showed up. [conroy compiler kept going for many years after that under pdp-11 thanks to hard work by late martin minow. it was the preferred compiler for pdp-11 hackers, so far as i can tell. its latest incarnation can be found in johnny billquist's pdp-11 archive.]

all this came to mind after reading a surprising fragment from a foreword by larry lessig to Open Source Licensing: Software Freedom and Intellectual Property Law [recommended]

Why code must be propriatery is a question whose answers have changed over the past ten years. At first the reasons were technical: no free or open source project, it was said, could develop the highly complex and robust code necessary for modern software applications. But when the GNU/Linux project began to produce an operating system that rivaled Microsoft's in robustness and efficiency, this technical argument began to fade.

emphasis mine.

(2005-02-15 08:52:46.0) Permalink Comments [0]