Behind the music – Spark and the PDP-11

The DataDataData (Itsthedatatalking) blog is meant to focus on data today – not to rehash my history of computing. But sometimes it veers that way, and I will just be holding on to nothing but the wheel. But I digress.

Apple scruff at The Smithsonian

Spark is the latest new shiny object in data processing. That said, I don’t mean to belittle its potential. The folks that fashioned it in the vaunted AMPLabs at UC Berkeley are supersmart, and very aware of what the advent of multicore microprocessors meant to computing: that new means to big clusters of  parallelism were available, if only the complexity could be abstracted downwards in clever libraries and runtimes.

People selling Spark come in your door selling Hadoop. Which has had plenty of publicity and is borderline ready for primetime. Now once in there, they may mention  you can toss Hadoop, but only if they think you may cotton to that.  After writing about Hadoop for about two years I took some care in approaching Spark.  Finally some words from way back came back. Please, let me digress some more.

Long ago and far away I sat with my boss discussing the news. The news on that day in 1992 was the ouster of Digital Equipment Corp.'s co-founder Ken Olsen. His departure was an inflection point along a trail that saw DEC go from being a gutsy Maynard, Mass. mill town startup to being a serious threat to IBM's industry leadership to being a forlorn merger candidate.

Like those in other editorial offices, my boss and I wondered what went wrong. What went wrong was the company got confused about what business it was really in. Seems absurd, but it can happen.
DEC's Olsen did not like the PC or Unix, two very innovative industry trends that his subordinates learned to basically eschew.  Missing on the move to small personal computers was especially ironic, as DEC itself rose in the 1960s on the back of minicomputers that downsized capabilities of the larger, then-dominant mainframe computer. Anyway, on this particular day I was especially interested to see my editor's take on this. That was because his experience went beyond running a magazine called EDN.

You see, as a graduate student, Jon Titus's had been in the vanguard of what came to be known as microcomputers, or PCs.  A July 1974 Radio Electronics issue that featured Titus's 8088-based "Mark-8 Personal Minicomputer" kit predated Popular Electronics' Altair 8088 cover story by six months.

In Cambridge, Mass., Harvard college student Paul Allen picked up a copy of the latter magazine, brought it back to the dorm to share with Bill Gates, and a new era of computing was off and running. Note that Titus and the Radio Electronics editors called the Mark-8 a personal minicomputer. So, Titus had a unique perspective on Ken Olsen's quandary.

"DEC came to think they were selling minicomputers," Titus said. "But what they were selling was computing."

Anyway- I link below to the full story on this which ran on SearchDatamanagement.com. I'd like to add here what a great boss Jon Titus was for me. He stood by me, more than once, which I never will forget. My spousal unit and I got to Washington last week. We went to the Smithsonian museum (actually, just two days after this story went live) and were told that the computer exhibit was closed for repairs (a lot of people can relate to that, ay?!) so we did not see the Mark-8 on display. Instead there was the computer that has, and maybe rightfully so, gained the brunt of the fame.

A cruel old engineer.

That is the Apple II of Steve Wozniak and Steve Jobs.  A woman came by and asked the air: "Is that the first computer?" No, said I, trying to be courteous, "the first computers were as big as rooms - that is what many people consider to be the first personal computer." Sorry, that's it for now - I got to go digress. – Jack Vaughan

Read Apache Spark meets the PDP-11 -- in the end, it's all about the processing – SearchDataManagement.com, Mar. 31, 2015 http://bit.ly/1Im9n1l