Jump to content

Talk:RSTS/E

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

RSTS-11 and RSTS/E

[edit]

Should RSTS-11 be a separate article? I'm not sure. The RSTS/E article basically describes the final version of the OS, while the RSTS-11 article describes V4A. The path from one to the other is really just the usual software product evolution. There was a name change from V4 to V5, and a major added feature (extended memory support). But there were other releases with new features of similar magnitude.

I would suggest merging RSTS-11 back into this article by discussing the history of RSTS.

Some points on history:

  1. Sysgen using RT-11 appeared in V6B (from V4 to V6A it was done with DOS).
  2. Runtime systems existed around V5B but weren't really visible until V6B.
  3. DCL appeared in V8.0 if I remember right

and so on...

Paul Koning 14:33, 30 July 2007 (UTC)[reply]

Merge RSTS

[edit]

Should RSTS-11 be a separate article? I'm not sure either. Seems a bit redundant, and I did it.

Call me a bit nit picky but our college went from V4A to V8-06 and the "usual software product evolution" in RSTS/E was MILES ahead of what we had before. I guess I think having an article that "describes the final version of the OS", would be like having an article about US Presidents and just give information about George Bush (there were presidents before, but of no significance).

I agree with Paul, that RSTS-11 and RSTS/E should be one (1) article with a better history line...and maybe a nice page design like Megan's RT-11...and how about we put Spike up in the corner? Paul...can you think of any special RSTS contributor's that could be mentioned (YES, include yourself)? Sorry that I opened a can of worms, but I didn't know that there were STILL RSTS guru's out there.

Additional points of history that could be mentioned:

  1. Sysgen with LIGHTS (Ok, so I like the swirly display)
  2. Longer passwords not stored in RADIX-50 (after V9.x I believe)
  3. Creating [0,*] accounts (again after V9.x I believe)
  4. All the fancy set account/whatever stuff in V9 and later (no more just user/superuser)
  5. Reference to Y2k issues.
  6. Anything cool and fancy in the 10.x line that happened after most of us had moved on.

and so on...and so on...

Bbump 19:07, 30 July 2007 (UTC)[reply]

If you went from V4 to V8 in one step it's no surprise you'd find it a big one. There's a nice "History of RSTS" writeup, a humorous one that switches seamlessly from reality to science fiction. It's online, I'll have to look for it. (It was done as part of, I think, the 20th anniversary of RSTS or of the PDP-11.)
Re contributors: I don't think so -- it's a typical team effort and I don't think there was any single personality so different from the rest to be "notable" for separate mention.
So far I hear "yes". I'll do some digging for data. Paul Koning 15:43, 1 August 2007 (UTC)[reply]
Yes, it was a huge jump for us, and I didn't start back-tracking into previous versions until years later. And it could be that the major hurdle between RSTS-11 and RSTS/E really was nothing more than the memory management code (I really don't know). I think the page your refer to is this one http://www.silverware.co.uk/rsts_80th_birthday.htm I remember it, and have reviewed it often.
Your probably right about the contributors (too many to mention). Mike was always helpful whenever I talked to him, PKH as well. Then the flood of authors writing in Rsts-Pro. Ah well.
Yes, please dig up whatever you can. I'm glad to help as well (novice wikipedian that I am). I'm also sure Christine will let us use her version of Spike (unless anyone else has a better one?).
Thanks Paul Bbump 22:56, 1 August 2007 (UTC)[reply]

I think the merge is a good idea also. (though my name is anonymous here, you can find my initials in common.mac from v7.0 on -- SJK) Vantelimus 18:13, 8 August 2007 (UTC)[reply]

RSTS Merged, Items Needed

[edit]
  1. Commercial and Non-commercial information on sites that used RSTS. I was hoping that someone with more knowledge of the Citibank systems could help provide some insight in the HISTORY section. Banks, insurance, universities etc.
  2. Development history from 1990-1992. Anyone?
  3. Run-Time Section (too much?/not enough?), someone want to compose a nice TECO example?
  4. Trivia Section is currently unfinished (but you can see where I was heading with it).
  5. Current status of Mentec-inc (still selling PDP-11 software)? Their website is still down, phone disconnected and no replies to emails.

Bbump (talk) 04:42, 4 January 2008 (UTC)[reply]

Wow, you sure did a lot of work on this. It looks very good. Opportunities for cleanup, of course, but this really moves things ahead. Paul Koning (talk) 17:03, 4 January 2008 (UTC)[reply]
Thanks Paul. I'm glad you cleaned up the RSX memory section. I debated about that one but tried to leave as much content as possible (not wanting to step on anyones toes). Bbump (talk) 14:45, 12 January 2008 (UTC)[reply]

The Great KWord Debate

[edit]

I am a little confused by the information in Peter's article concerning the 1975 release of RSTS/E Version 6A-02. It's clear the 11/70 will address 4MB of memory (or more appropriately 2MW). My confusion stems from his comment that "maximum memory on the 11/70 was now 2M bytes." This would lead me to believe that there was an "as yet" unresolved problem with the memory management code (ability to address 21 of 22bits), that Peter actually meant 2M words instead of 2M bytes, or that somewhere along the road (editing), communication was lost. Without having access to any RSTS/E version prior to V7.0 (that I can get working), I can not test this information to resolve this adequately (thus, it remains a debate). I am unsure if the 6A-02 Version section should be posted as 21 bit instead of 22 bit (sorry, there never was a 21 bit PDP-11). Left as 22 bit, obviously someone will come along and correct our 2MB statement. If Peter's article is wrong and 6A-02 (on an 11/70) will actually run 4MB of memory, then the 1978 section becomes redundant (please edit accordingly). Bbump (talk) 14:45, 12 January 2008 (UTC)[reply]

I don't have sources that would help, either. I believe someone around here does... Anyway, it would be quite unusual for an article about RSTS, or about any PDP-11 software for that matter, to quote memory sizes in (mega)bytes. The universal convention for PDP-11 software was to refer to (16 bit) words.
It's certainly possible that older versions had some limitation around 22-bit addressing, perhaps due to abuse of signed arithmetic -- but RSTS in general did a good job on getting that issue right.
Paul Koning (talk) 14:58, 14 January 2008 (UTC)[reply]

TECO

[edit]

It might be worth mentioning that when installing or upgrading RSTS (i.e. during SYSGEN), during the configuration process it was necessary to edit several configuration files. At that point the half-installed OS was unable to load any normal program; TECO was the only editor that could be run. At the same time, TECO was always officially "unsupported" by DEC, although at various points in the voluminous system documentation one would find little footprints -- a file open mode, say, or a bit in the keyboard status word -- marked "Reserved for TECO". This was also true of RSX. So DEC's premier operating systems in the 1970s could not be installed on DEC hardware without an "unsupported" chunk of code... -- Craig Goodrich 206.39.12.241 (talk) 17:31, 24 April 2009 (UTC)[reply]

Hobbyist license status unknown

[edit]

Because it might be misleading, the text talking about Mentec issuing a no cost hobbyist license should probably be removed because it is still being actively pursued in an official fashion. It was actually DEC who granted this to Bob Supnik and only for use on emulation products, not on a general scale like the article would suggest. —Preceding unsigned comment added by 64.107.133.124 (talk) 19:33, 17 March 2008 (UTC)[reply]

This is incorrect. There is a license text Mentec released, which basically said that anyone running any of DEC's software that was sold to Mentec, if they were using it for personal, non-commercial use, were allowed to do so. It was not specifically a license to Bob, nor did it even mention the emulator. You were on your own as far as getting copies of any of the software. I know the license was for anyone because I saw it and read it once when looking at some of the PDP-11 stuff. Paul Robinson (Rfc1394) (talk) 14:06, 30 May 2010 (UTC)[reply]

Security issues

[edit]

I recall some serious security issues with this; programmers from user accounts in BASIC were able to get superuser privileges. I'd love to see a section here about these early hacks, but don't have any authoritative references. —Preceding unsigned comment added by 98.234.189.161 (talk) 20:52, 17 April 2009 (UTC)[reply]

As an early and dedicated RSTS hacker (in the purest sense) ISTR that the main problem was with interrupting the operation of SYSTAT which left the user with elevated privileges.
There was one spectacularly sophisticated (for its time) hack that I was involved with at Southampton University which allowed a user in the know to gain privileged access at any time without any obvious indication of how it was done - a small hint here - a privileged program was marked on disk by a single bit in its protection code in the MFD. —Preceding comment added by Nickds1 (talkcontribs) 16:31, 10 July 2009 (UTC)[reply]
More on this - Just remembered that the issue was that SYSTAT, like all RSTS system utilities, was itself written in BASIC Plus. If you CTRL/C'd SYSTAT, you dropped back to the interactive prompt but were left with elevated privileges - you could then execute SYS(CHR$...) (pronounced sys-chris) commands to further enhance your entry into the system. Subsequent versions of RSTS/E removed this loophole... —Preceding unsigned comment added by Nickds1 (talkcontribs) 09:29, 11 July 2009 (UTC)[reply]
That would have had to be very old, because I remember how the RSTS manuals (probably after version 7) said that when you were a non-priveleged user running a privileged program, and if it exits for any reason: normal exit, you ^C out of it, or it crashed without an ON ERROR GOTO (or did a RESUME 0), the process loses its temporary priveleges, the current program is destroyed (erased from memory), and the user is placed at the Ready prompt with the NONAME program and nothing in the user's job. If you were a non-logged in user, when the program exits it is destroyed and you get a Bye prompt. Paul Robinson (Rfc1394) (talk) 14:18, 30 May 2010 (UTC)[reply]
As for changing the MFD, I didn't think you could do that unless you had privileges, for obvious reasons, e.g. if you can change the MFD the whole point of having logins and accounting is meaningless - as it would be on any system - if you can just go out and edit the disc directly, you simply can change anything that isn't what you want to something else that you do. It's like the issue of database systems, if someone can backdoor the databases through a database manager, all the controls, auditing and privilege checks being done by the application program are meaningless for them. Paul Robinson (Rfc1394) (talk) 14:22, 30 May 2010 (UTC)[reply]
I think you are missing the point here - obviously you couldn't do this on a normal system - I'm dredging my memories a bit (it was 35 years ago), but what we did was write a short program that opened the disk in raw mode and changed the data mapping pointers for an extant non-privileged file in a neutral user's directory to overlay the MFD entry for another program that asked for a program name an attempted to set its privileged bit - obviously, when run normally, this latter program would fail with a privilege error. The hack centred round tricking an operator to run the mapping program just once (which we managed to do by hiding it in other code - a traditional, and extremely early, trojan). Once the mapping pointers were changed, you could manipulate the MDF directly from user mode and set the privileged bit in the application mentioned above which in turn allowed you to elevate any application you wished. This was extremely dangerous in that if you accidentally deleted the re-mapped file, the file system would put the mapped chunk of the MFD on the free list which could then be trashed by any new file or file extension, resulting is a manifestly corrupt disk... Worked a treat though; For months we had privileged access whenever we wanted - could turn it on or off at will from user mode - and no-one knew... Nickds1 (talk) 15:11, 5 June 2013 (UTC)[reply]

Author of BASIC Plus...

[edit]

If my memory serves me correctly, the original author of BASIC Plus was Mark Bramhall, who, allegedly, wrote the first version during a transatlantic flight... I still have a whole bunch of tapes of this stuff, including, I believe, some source code. I do recall using the DECOMP package that Dave Garrod & I wrote to decompile most of the RSTS utilities. As you might have guessed, the DECOMP package would not decompile itself ;-) We sold a separate utility that would make BAC files undecompilable... —Preceding unsigned comment added by Nickds1 (talkcontribs) 09:36, 11 July 2009 (UTC)[reply]

More hacks

[edit]

I remember someone figured a way to patch a BAC file to change the running name of a program, a guy was using (what we would now call) a hex editor on a Basic-Plus program (since the PDP-11 was octal, it was probably either an octal editor or decimal byte editor) so that if you did a control-T for the terminal running it, or a SYSTAT elsewhere, instead of the program being "MYPROG" it would show up as "..OHNO". Paul Robinson (Rfc1394) (talk) 14:28, 30 May 2010 (UTC)[reply]

Actually, this is a very easy process to do from within a BASIC program, compiled under BP2. I outlined this procedure in the October 1988 DECUS Newsletter. Once compiled, changing your current running program can be done very easily at any point by loading SYSCOMMON with the name you want, and then making a system EMT call such as:

10 FOO$=SYS(CHR$(6%)+CHR$(-10%)+"FOOBAR")
20 CALL NAME

The NAME call will convert your 6 ASCII character program name into the correct RAD50 format which is loaded into the File Request Block (FIRQB). I'm hoping Paul will correct me if I am wrong, but I seem to recall that CORECOMMON and SYSCOMMON were the same reference, just newer terms (my own memory also being now more dynamic than CORE). Bbump (talk) 15:27, 11 September 2010 (UTC)[reply]

[edit]

In the article Symbolic Link it is mentioned that symbolic links were already present in 1978 minicomputer OSs from DEC. Since I knew RT-11 quite well I'm shure that RT-11 did not have smbolic links. I guess that it must have referred to RSTS. But also there I found only a kind of command aliases, which are far from symbolic links.

Is there anybody around who used RSTS and can confirm that symbolic links did not exits on a file system level in RSTS? Treutwein (talk contribs) 08:20, 24 January 2011 (UTC)[reply]

"RSTS/E" pronunciation

[edit]

Hi, I'm reading the autobiography of Kevin Mitnick who, in his teenage years, became an expert of RSTS/E. In the book he states "RSTS/E (spoken as 'RIS-tisEE')". This pronunciation conflicts with the two we have in the lead sentence of the article. I am not familiar enough with the subject to change it. Could someone who knows a bit more about the subject confirm this and change it? --Oldak Quill 18:26, 17 November 2011 (UTC)[reply]

When I worked for DEC (in the UK & Europe) we pronounced it more as 'RIS-tusEE', but its literally a semantic point. The above is close enough. Nickds1 (talk) 09:00, 26 March 2012 (UTC)[reply]

File:MicroPDP-11 53wRSTSDocs.jpg Nominated for speedy Deletion

[edit]

An image used in this article, File:MicroPDP-11 53wRSTSDocs.jpg, has been nominated for speedy deletion for the following reason: Wikipedia files with no non-free use rationale as of 3 December 2011

What should I do?

Don't panic; you should have time to contest the deletion (although please review deletion guidelines before doing so). The best way to contest this form of deletion is by posting on the image talk page.

  • If the image is non-free then you may need to provide a fair use rationale
  • If the image isn't freely licensed and there is no fair use rationale, then it cannot be uploaded or used.
  • If the image has already been deleted you may want to try Deletion Review

This notification is provided by a Bot --CommonsNotificationBot (talk) 10:05, 3 December 2011 (UTC)[reply]

Sorry for not getting to this sooner. I would not have even noticed it today, had I not received an email from Wikipedia saying someone was trying to change my password (not me). It would appear that the robots are about to take down the photos I created (File:MicroPDP-11 53wRSTSDocs.jpg and probably Example of a RSTS/E Document shelf). If anyone can come up with an excuse within the next 4 days as to why these should remain on the page (as non-free content), I would encourage them to go ahead and edit them. The photos were taken by me (me, myself and I, nobody else...metadata of the camera included in the files). At the time, I really did not know how to post them (is this free, non-free, pizza-fax?). DEC no longer exists. Three (3) of the four (4) books were created by DEC. The RSTS Professional was done by Professional Press (who no longer exists). Professional Press was sold about 3 different times, and when I contacted the REAL owners, they didn't know what it was. The tile floor and paint on the wall was purchased at Menard's and installed by my brother-in-law (is that a copyright issue?). Sorry if I sound sarcastic, but as I say, I really do not know how these should be categorized (I took photos of my bookshelf, books and a PDP-11/53...does it appear we are advertising soda pop?). As always, I encourage anyone to post something better if they have it. As for me, I have the originals of everything I wrote and files that I posted, so if the bots rule the world and take them down, at least I have my original copies. 00:48, 27 June 2012 (UTC) — Preceding unsigned comment added by Bbump (talkcontribs)