interesting-people message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Subject: [IP] more on snobol & java

  • From: David Farber <>
  • To:
  • Date: Wed, 26 Jul 2006 07:40:26 -0400

Begin forwarded message:

From: Bob Frankston <>
Date: July 25, 2006 11:57:31 PM EDT
To: &quot;'John Shoch'&quot; <>
Cc: Dave Farber <>
Subject: RE: [IP] more on snobol & java

Memories memories ... I hesitate to send this because it&#x2019;s so incomplete and I fear generating thousands of responses from those of us young enough to remember those days.

I remember using Snobol a high school student using the IBM 7094 at NYU &#x2013; probably around 1965. We would place the binary deck for SNOBOL3 &#x2013; a few inches of cards &#x2013; in front of the few cards for our own programs. Copying the deck was a challenge &#x2013; the 519 wasn&#x2019;t up to the task so we&#x2019;d have to submit a job to get a copy of the deck.

Gio Wiederhold &#x2013; I remember visiting in 1971 to evaluate a start-up he was involved in and learning about his ACME terminals which had lights showing if you program was swapped in and running or waiting. Also visited Butler Lampson at Xerox PARC &#x2013; he had consulted for White-Weld where helped on the SDS-940 and designed the financial language for the online service. The implementation of which paid for much of my college education.

It wasn&#x2019;t that long ago was it?

It was sort of like Precambrian period in evolution &#x2013; there were many experiments in programming and languages. There was more than a enough to study in the history of programming languages class I took in 1966.

So many operating systems and machine architectures with bit lengths from 8 (on that new fangled IBM 360) to the common 36 bit machines to the 60 bits on the CDC 6x00 machines. The roots of timesharing with CTSS at MIT begetting Multics and then Unix. In that same building at 545 Tech Square CP/CMS was prototyped on the IBM 360/47 (one of a kind) and shipped on the 360/67 which we used at Interactive Data (spun out of White Weld where we started on the SDS-940, n&#xE9;e the Berkeley Project Genie machine whose creators went on to PARC). Dartmouth started with BASIC in the early 60&#x2019;s on GE machines creating its own generations of online services. There were a number of smaller online services on their own hardware.

More interesting than all the software or hardware is the flow of ideas between people and groups &#x2013; there were relatively few people in the field and I accidentally crossed a few communities. I really should write more of this down &#x2013; even better if someone could put together a map of the flow of ideas in both academia and commercial computing.

One example was Jerry Saltzer&#x2019;s 1966 Runoff which generated IBM Script (rewritten on CP/CMS) and merged with the IBM automated document effort yield GML and then SGML and eventually HTML. Stanford&#x2019;s Pub took advantage of the XGP (Xerox Graphic Printer). Impress and the Postscript came out of the West Coast efforts. I&#x2019;m omitting the names of the people at this point to avoid the need to make sure I get them all right and to avoid slighting those I fail to mention. And I&#x2019;m also leaving out the many side stories as ideas branched out and re-met. I&#x2019;d also want to do more learning &#x2013; for example, where did &#x201C;auto-secretary&#x201D; on the SDS-940 fit into this mix.

Well, enough nostalgia but I can&#x2019;t help but wonder what would have not happened had we today&#x2019;s concept of &#x201C;Intellectual Property&#x2019; as something to control rather than share.

-----Original Message-----
From: David Farber []
Sent: Tuesday, July 25, 2006 20:58
Subject: [IP] more on snobol & java

Begin forwarded message:

John Shoch <>

From:   John Shoch <>

Date:    July 25, 2006 7:53:48 PM EDT


Cc:       John Shoch <>

Subject:            RE: [IP] snobol & java


Now this is a trip down memory lane....

&#xB7; I first encountered Snobol in a course on &#x201C;Non-Numerical Methods&#x201D; taught at Stanford by Alan Kay and Gio Wiederhold, around 1971. [Is that really 35 years ago?!] It was running in batch mode on a 360/67.

&#xB7; I remember I was confused for a little while, until I figured out that (unlike other langauges) there was no EXPLICIT operator for concatenating two strings&#x2014;you just wrote them next to each other.

&#xB7; For the take-home exam at the end you could pick one of 3 problems, one of which was a Snobol project. As I recall, I struggled with it, finally figuring out that it could not be done&#x2014;to answer the particular question you needed to observe some internal state of the pattern-matching system, which was not available. With some trepidation I went to see Alan: &#x201C;I&#x2019;ve worked on this pretty hard, but I don&#x2019;t think it can be done....&#x201D; He looked up and said something like, &#x201C;Oh, I haven&#x2019;t even tried it....&#x201D; After about 5 minutes of explanation he agreed, told me I&#x2019;d done enough work, and said, &#x201C;What are you doing this summer? Xerox is opening a research center here in Palo Alto....&#x201D; I went for the summer, and stayed 14 years.


John Shoch

Alloy Ventures

-----Original Message-----

From:   David Farber []

Sent:    Tuesday, July 25, 2006 4:23 PM


Subject:            [IP] snobol & java

Begin forwarded message:

From:   Tim Bond <>

Date:    July 25, 2006 7:02:09 PM EDT


Subject:            snobol & java


Was going to send this to IP, and feel free to forward.

I remember &#x201C;experiencing&#x201D; a month of Snobol in the fall of 1984 as part of my CS Intro to programming languages class at RPI . . .


&#xB7;         Tim

You are subscribed as

To manage your subscription, go to

Archives at: people/

You are subscribed as
To manage your subscription, go to

Archives at:

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Search: Match: Sort by:
Words: | Help


Powered by eList eXpress LLC