[ih] PDP-11 high level language [Re: how big was the host file]

[John Day]

It was called PDP-11 Espol, or PEESPOL designed by David Grothe of our group. Grothe wrote the complier, Gary Grossman designed and wrote much of the OS.  We filled in.  This was about 1970.  As I said, we were building Illiac IV and had a B5500 for software development. The 6700 was in the process of being built. IlliacIV was to be a peripheral processor to the 6700.  As the name implies, PEESPOL was ESPOL for the PDP-11.

Grothe latter added a very sophisticated macro processor that essentially made it an extensible language.  The 2nd OS we did was a process per function, message passing OS. The device drivers were state machines and ‘instructions’ were passed among the drivers. PEESPOL allowed us to define instructions, declare them, treat them as a data type, access their fields, etc. We had macros for procedure variables, for queues, etc. We basically could expand the language in terms of the objects in the system we were building and then program in those terms.  We were so much into structure that sometimes statically scoped languages like Algol force you to make things more global than you want. We introduced an “UN” operator, so we could un-declare things, so we weren’t tempted to use things we weren’t suppose to be able to see.

We did a 3rd OS called the Hub, that was so simple, the whole thing fit 4K of code, 6 with buffers and later version was proven secure.

Then in the Summer of 75, we put the first Unix system on the Net. Then stripped it down to fit on an LSI-11 with a plasma screen and touch to create an intelligent terminal for a land-use management system for the 6 counties around Chicago. Everything could be done by touch. We could maps down to the square mile, graphs, etc. The system ran on various systems on the ARPANET.

John

> On Feb 6, 2020, at 14:31, Brian E Carpenter <brian.e.carpenter at gmail.com> wrote:
> 
> John,
> 
> We can take this off-list if you think it's too off-topic, but:
> 
>> Our compiler for the PDP-11 high level language ran on 6700
> 
> What was that? Which language and which compiler?
> 
> I ask because:
> 
> B.E. Carpenter, A PL-11 Package on the Burroughs B6700, Massey University Computer Unit, Report No. 15, December 1974.
> 
> PL-11 was Bob Russell's language for the PDP-11, developed at CERN in 1971, which I ported to to the 6700. As you may know, all 5 NZ universities bought 6700s and it was one of the reasons I went to Massey. After CERN, Bob Russell went to UNH.
> 
>> The 5500/6700 was probably the finest system design I have ever encountered.
> 
> Absolutely. A lovely machine.
> 
> Regards
>   Brian Carpenter
> 
> On 07-Feb-20 04:57, John Day via Internet-history wrote:
>> Politics.
>> 
>> It probably had to do with IlliacIV not coming to Illinois. (That is a very long story.) Ken Bowles at UCSD develop some interesting software on it. We sent one of our best system guys with it. Our group knew both machines inside out. He had added IPC to the 5500 and we had what must have been pre-beta versions of the MCP.  Our compiler for the PDP-11 high level language ran on 6700 and we developed two OSs on it.  Compiling at UCSD and downloading to the 11 at Illinois.  NASA got their own 6700 but got rid of it for a Tenex.
>> 
>> It was a very low serial # machine (like first 5). That machine had been heat stressed.  The machine was in the basement of the (original) Coordinated Science Lab at Illinois.* The idiot in charge of the installation thought it was too much trouble to run A/C duct work to the raised floor and had an A/C vent pouring cold air out from the ceiling . . . directly over the CPU cabinet!  Needless to say it didn’t work well. There was rain in the machine room.
>> 
>> The 5500/6700 was probably the finest system design I have ever encountered. Even Organick said, ’it appears they got everything right.’ Seeing how elegant designs could be was a life changing experience. Organick’s book describes what they did, but I would like to know how they did it. How a group in 1964 (when the 5500 was done) could do something so radically different than anything else being done and get so much right.
>> 
>> John
>> 
>> *The new CSL is where my house was! ;-)
>> 
>>> On Feb 6, 2020, at 10:26, Steve Crocker <steve at shinkuro.com> wrote:
>>> 
>>> What caused the B6700 to be moved to UCSD?
>>> 
>>> On Thu, Feb 6, 2020 at 10:22 AM John Day via Internet-history <internet-history at elists.isoc.org <mailto:internet-history at elists.isoc.org>> wrote:
>>> I am pretty sure Illinois never had two hosts connected at the same time. I am guessing this was more reserving names.
>>> 
>>> il-67 was probably a placeholder for the Burroughs 6700 but it was not put on the Net while it was at Illinois. Later it was connected when the machine was moved to UCSD.
>>> There was a PDP-11 connected. Later we might have had both a PDP-11/20 and a PDP-11/45 connected but I doubt it. 
>>> For a while, there was a VDH to Purdue, but they eventually got their own IMP.
>>> 
>>> I will have to do some checking.
>>> 
>>> John
>>> 
>>>> On Feb 6, 2020, at 09:48, Lars Brinkhoff <lars at nocrew.org <mailto:lars at nocrew.org>> wrote:
>>>> 
>>>> Craig Partridge wrote:
>>>>> ISI had three IMPs. SRI had three IMPs. SAC had two IMPs. ARPA had
>>>>> two. Those are additions to your list of Stanford, MIT, UCLA and BBN.
>>>> 
>>>> Not quite.  They had multiple hosts on one IMP.
>>>> 
>>>> Here are some more.  Excluding those with just an additional TIP.
>>>> 
>>>> cmu-10alt  116
>>>> cmu-cc     016
>>>> 
>>>> ll-67      012
>>>> ll-tsp     212
>>>> ll-tx2     112
>>>> 
>>>> parc-maxc  040
>>>> parc-vts   140
>>>> 
>>>> rand-csg   107
>>>> rand-rcc   007
>>> 
>>> -- 
>>> Internet-history mailing list
>>> Internet-history at elists.isoc.org <mailto:Internet-history at elists.isoc.org>
>>> https://elists.isoc.org/mailman/listinfo/internet-history <https://elists.isoc.org/mailman/listinfo/internet-history>
>> 
>