Sunday, May 31, 2009

Weekend Links

Were the Challenger catastrophe and the fuzzy Hubble going to kill NASA?

Wayne Hale's qualifications and experience are impeccable, and the quality of his perspective is rare (and all are infinitely superior to my own), and he raises good points. However, though it pains me to say so, I'm not yet persuaded by his arguments, as expressed in his blog entry, "Great Expectations", referenced above. I'm not convinced that NASA was on the brink of disbandment after flying Hubble with a botched mirror, and destroying Challenger and killing her crew. It was undoubtedly in deservedly terrible shape politically, and therefore some sort of penance was inevitable. Maybe that even made the first Hubble servicing mission a necessity, in order to do some repair to the reputations of Hubble, Shuttle and NASA. However, NASA has its centers and it contractors and subcontractors spread all over the nation, and killing it outright would have meant job losses in almost every state, and a great many congressional districts. That puts a lot of Senators and representatives in Congress in a very awkward situation when talking of making big changes to NASA, let alone killing the agency. Some say NASA has made a practice of spreading the work around to protect itself. In other cases, it seems that politicians have made it happen for their own benefit. In practice, it means that NASA's work, especially its human spaceflight programs, are ensured to go on in some form, provided that a minimum of congressional districts are left out in the cold by any change. (I've often wondered if this was the main reason the original Constellation architecture relied so completely on Shuttle and Shuttle-derived hardware, even where that raised obvious problems.)

I must confess that, as a child of Apollo, Shuttle has always looked to me like a dead-end and a disaster, despite the incredibly hard work that was, and is, put into it. It was over budget and behind schedule – not uncommon in such things, but as a kid those schedule slips felt like they took forever. More siginificantly, it promised two-week turnarounds, and a dramatic reduction in the cost of reaching low earth orbit, neither of which it delivered (not even close, in the former case, and quite the opposite in the latter). Of the five shuttles flown, two have failed catastrophically and a third came within inches of doing so, so its safety record is terrible, but roughly, I gather, what NASA originally predicted (98%, if memory serves). If the out-of-focus Hubble and the loss of Challenger had killed the Shuttle program, I wouldn't have missed it. If that had eventually translated into more robotic exploration, I'd've been well pleased. But I don't see how the politics of government spending would have left us without a human spaceflight program of some sort for too long (the gap between the end of Shuttle and the beginning of Ares I flights is causing enough trouble). Maybe the replacement would even have been one that could, as Constellation promises to do, take us somewhere beyond low earth orbit. Which, though I prefer the bang-for-buck that robotic exploration delivers, I must admit still excites the child of Apollo in me.

Sunday, May 24, 2009

Weekend Links

Sunday, May 17, 2009

Weekend Links

Tuesday, May 5, 2009

J. David Bamberger Honored by Texas House of Representatives

Yesterday, my friend, J. David Bamberger, creator of Selah, the Bamberger Ranch Preserve, was honored in the Texas House of Representatives for his forty years of unique conservation work.

Congratulations, David.

Thursday, April 9, 2009

Holy Relic - Core Memory

I've recently been on a quest to find and begin digitally archiving the surviving history of the University of Texas at Austin's Computation Center, which was responsible for the academic side of computing on campus from 1958 until it was merged with our data processing and telecom shops into a new department called Information Technology Services (ITS) a handful of years ago. The result of that merger and the passing of time is that much of ITS seems never to have heard of the Computation Center, which did a lot of good work that shouldn't be forgotten.

While searching for copies of the Computation Center newsletters (none of which I had had the foresight to keep), the same colleague who seems to have the only existing set that is even potentially complete, revealed to me one of our forgotten holy relics: A core memory card from our Control Data Corporation Cyber 6600, or 6400 computers, or the extended core memory that they shared. Anyway, core memory.

The card, which measures 4.125 inches on each side, looks like this:

In extreme close-up you can see the ferrite rings (the "cores") that magnetically stored the data, supported by the matrix of fine wires (slightly thinner than a human hair) that run through the middle of the rings, allowing data to be written-to, and read-from, them. Each one stored a single bit of data. If I've counted correctly, this board stored 4,356 bits of data.

I remember our Cybers, but never did any work with them. However, years before joining the Computation Center, I did do much of my early programming on a Digital Equipment Corporation PDP-11/34 (running RSTS/E) that also used core memory.

Saturday, April 4, 2009

Weekend Links