Talk:DOS/360 and successors

Miscellany/trivia
I know that it is Wikipedia policy to squash trivia sections, as requested by a recent tag; but I think that sometimes they are appropriate, and I think this is an example. Both of the entries here are a) interesting enough to keep, b) out of the main discussion, and c) adjuncts to a small/stub article that IMO shouldn't be diluted with trivial asides in the main content. Perhaps they should go into footnotes, but then they would be nearly invisible. This is of course only my opinion, and it isn't the party line, but I don't think that a "miscellany" section is intrinsically something that needs to be removed/merged, particularly in a short article. I do recognize that they are a problem in mature articles; but isn't this a different situation? Trevor Hanson 07:00, 4 August 2007 (UTC)


 * I agree. The item about GECOS aims to refute an apparent urban myth; and the item about MS-DOS clears up an obvious source of confusion. Philcha 00:39, 21 October 2007 (UTC)

Restructure of articles about IBM mainframe operating systems
After a big edit of MVS I concluded that the whole set of articles about IBM mainframe operating systems from System/360 onwards needed to be re-structured to minimise overlap and to make clearer the evolutionary relationships between these operating systems (notably in memory management, which is historically a major distinguishing feature). There is already some support for this proposal. Please add comments at Talk: MVS. Philcha 23:56, 20 October 2007 (UTC)

Rewrite of Job Control Language in progress
As part of the proposed restructure of articles on IBM mainframe operating systems (above), I've rewritten Job Control Language to: cover IBM's DOS/360 and its descendants as well as OS/360 and its descendants; focus more on the facilities and flavour of the 2 JCLs rather than on details of some statement types and some of their options. Please comment in Talk: Job Control Language. I'd be particulary grateful for more info on DOS/360 and its descendants, especially after 1980 - I only used DOS JCL a handful of times, and only in the late 1970s.

The rewrite does not currently take account of Truthanado's point in Talk: Job Control Language about use of "JCL" by computer suppliers other than IBM, which may entail further restructuring of articles about JCLs. Philcha 00:41, 21 October 2007 (UTC)

Requested move and merge
I think this artice should be ranamed to DOS/360 and successors and be merged with DOS/VS, DOS/VSE, VSE/ESA. The Summary style will enable much better explanation and feature comparison between all the systems in line. It will result in a much more interesting article. For more discussion see Talk:History of IBM mainframe operating systems. To be clear, I don't think z/VSE should be merged. --Kubanczyk 14:39, 27 October 2007 (UTC)


 * Renamed and merged with DOS/VS. --Kubanczyk (talk) 19:54, 7 April 2008 (UTC)

Rewrite tag
On 7 April 2008 Kubanczyk tagged this article for a complete rewrite. Considering that (a) Kubanczyk gave no explanation as to why a rewrite might be needed (b) After nine months nobody else has responded in any way to the tag and (c) it is not evident that there is a need for a rewrite, I have removed the tag. JamesBWatson (talk) 23:20, 8 January 2009 (UTC)

Simple Note
"still widely used as of 2006"... ok, that's three years ago... we need an update... is it still widely used... and what does "widely" mean, for that matter... 3 places in the world? 2000?... Where? Some podunk business of in the sticks, or, like, a major military base? Let's get with it here, people... 69.179.183.180 (talk) 01:25, 29 November 2009 (UTC)
 * I updated it to 2011 using the best source I could find, such as it is. Peter Flass (talk) 20:13, 30 June 2015 (UTC)

Missing History
There are some topics that I believe belong in this article.


 * POWER
 * A Spooling system for DOS


 * DOS/VSE
 * A new release of DOS capable of exploiting ECPS:VSE


 * VSE/AF
 * A program product extending DOS


 * VS/SP
 * A program product incorporating DOS and VSE/AF.

Shmuel (Seymour J.) Metz Username:Chatul (talk) 19:26, 11 June 2010 (UTC)

DOS/VS Paging
I think the description of how DOS/VS organized the page table is in orger. Was it a single 16MB page table for all partitions? Could it be less than 16MB? Peter Flass (talk) 13:04, 19 May 2012 (UTC)


 * ITYM is in error.


 * Yes, it was a single page table, but the same was true for OS/VS1 and SVS, derived from MFT, MVT  respectively. Shmuel (Seymour J.) Metz Username:Chatul (talk) 22:22, 22 May 2012 (UTC)

Improper link for DOS/VSE
DOS/360 and successors has, but VSE (operating system) is actually about z/VSE; there is no article on DOS/VSE. Shmuel (Seymour J.) Metz Username:Chatul (talk) 19:50, 13 August 2013 (UTC)

Incorrect heading
DOS/360 versions== ==

None of TOS/360, DOS/VS, DOS/VSE, SSX/VSE, VSE/SP, VSE/ESA or z/VSE is a DOS/360 version; the original header, ==DOS/360 versions==, was correct. Also, DOS/360 should not have been deleted from the list. Shmuel (Seymour J.) Metz Username:Chatul (talk) 18:12, 5 March 2014 (UTC)
 * I may have written the heading originally; maybe "versions" isn't the right word, but they're all in the DOS/360 "family."  "DOS/360 and successors" isn't right either. BOS should probably be mentioned also. Peter Flass (talk) 20:24, 5 March 2014 (UTC)


 * The problem isn't the word versions, but changing DOS to DOS/360 in the section heading and deleting the DOS/360 subsection. DOS/360 is just one of the versions in the family. Shmuel (Seymour J.) Metz Username:Chatul (talk) 21:36, 7 March 2014 (UTC)

No technical similarities ...
... other than being Disk Operating Systems. This typifies what it means to be a typical oblivious technical moron. Lycurgus (talk) 10:01, 10 November 2014 (UTC)


 * adding that of course I'm familiar with both of these products having worked with them in their day. Lycurgus (talk) 18:31, 28 June 2015 (UTC)


 * Yes, believing that being a disk operating system is a technical similarity shows you to be a typical oblivious technical moron. DOS/360 and TOS/360 are technically similar; DOS/360 and PC-DOS are not. Shmuel (Seymour J.) Metz Username:Chatul (talk) 20:31, 28 June 2015 (UTC)


 * So what semantics do you associate with the English noun phrase "Disk Operating System"? What do you ascribe as far as intent of IBM and MS in choosing this name? Or perhaps more the point, what is the essential thing that differentiates "technical similarity" from other instances of the concept as applied to members in a set (in this case operating systems). I'm not really interested in your reply but others may be interested in your wisdom on this. Lycurgus (talk) 03:23, 29 June 2015 (UTC)


 * The semantics are that the system residence device is a disk; only this and something more;


 * AS for your other question, if you're not really interested then it would be pointless to answer unless and until someone more reasonable asks. Shmuel (Seymour J.) Metz Username:Chatul (talk) 14:16, 30 June 2015 (UTC)


 * Other members of the set may be ROM-based OSes and tape-based systems. Peter Flass (talk) 14:41, 30 June 2015 (UTC)


 * A particular way in which DOS/360 and PC-DOS are wildly different: PC-DOS depended on considerable functionality that was implemented in the PC's "BIOS" (firmware), invoked via software interrupt "calls". See for example []. For example, PC-DOS did not include the code to access the floppy or hard disk; the "drivers" for these devices were part of the PC's firmware. This was not the case for DOS.


 * Another fundamental difference: PC-DOS implemented a file system with named files and directories and, eventually, subdirectories; a file could occupy as little as one allocation cluster on a disk. DOS/360 did not; it had no concept of a named file at all. "Data sets" were identified within "jobs" by volume, starting track number, and number of tracks, via parameters on EXTENT cards in the JCL deck. These "extents" had to be manually allocated and kept track of by the installation's management. Once a "job" was over all information about where the "data sets" it had worked on was gone from the OS's memory.


 * And yet another: The PC-DOS command language is implemented in COMMAND.COM and is token- and syntax-driven. As far as I know DOS/360 did not have a separately loaded program to interpret JCL. It was just part of the OS in memory. Since that necessitated it be very small, DOS JCL used a lot of positional parameters so that e.g. keywords did not have to be parsed. See ([].)


 * Let's see. They're both operating systems, they're both loaded from and allow loading programs from disk, and they both implement primitive command languages (very different from each other, though). These rate as superficial (i.e. on the surface) similarities. Beyond that I wouldn't even say they are superficially similar, let alone that they have technical similarities. Jeh (talk) 02:29, 11 September 2015 (UTC)

I think DOS-360 JCL was implemented via B-transients. Peter Flass (talk) 10:28, 11 September 2015 (UTC)


 * DOS/360 Job Control ran in the partition to initiate the user program and then clean up and initiate the next job. The minimum 16K system requirement was based on a minimum 6K Supervisor and the 10K required to run Job Control. Certainly, Job Control could have invoked services requiring B-Transients (not my area of expertise). WarrenFW (talk) 06:25, 15 August 2020 (UTC)


 * I viewed IBM pub SY33-8555-1 DOS/VS (Release 29) IPL and Job Control Logic online. It had been scanned and it appeared to have been printed on an impact printer, probably an IBM 3211.  Obviously the text I grabbed is less than perfect, but you can see that it states Job Control is loaded into the partition:
 * The job cont~ol progran prcvides jeb-te-job transition for all programs. It also prepares program job steps fer executien. (One or more programs can be executed within a single job. Each such executicn is called a job step.) The job control program is automatically leaded into the background ~artition after IFL. Te obtain the job control progran in a fcreground partition, a BATCH or START conn and must be issued for the desired partiticn. The jeb control program is then loaded intc the desired partition if the following minimum requirements are met:
 * It wasn't worth my time to clean it up and I didn't include the requirements because by R29 it required a 64K partition! WarrenFW (talk) 03:47, 25 September 2020 (UTC)

Pronounciation
When I worked with IBM 360, DOS was an initialism, rather than an acronym. In other words, it was pronounced "dee-oh-ess" rather than "dos". Does anyone else think this should be mentioned somewhere? SlowJog (talk) 00:45, 17 November 2016 (UTC)
 * My experience also. Peter Flass (talk) 07:20, 17 November 2016 (UTC)
 * Remember the Wikipedia mantra: We'll need a reference. ;) Jeh (talk) 09:55, 17 November 2016 (UTC)
 * Good point. Because the people I associated with used the initialism doesn't mean it was universal, or even majority. SlowJog (talk) 00:55, 12 December 2016 (UTC)


 * I've heard ASP, BOS, DOS, OS and TOS both pronounced and spelled out. But is the pronunciation in the field notable? FWIW I've only heard HASP, IbJob and IbSys pronounced and ATS, BPS, CMS, CP, TSS and VM spelled out. Shmuel (Seymour J.) Metz Username:Chatul (talk) 21:46, 18 November 2016 (UTC)


 * Wow! Now I'm confused. I think we may have said both DOS and D-O-S, but I know we never said BOS, TOS, or OS. We always said B-O-S, T-O-S, and O-S. Once DOS/VS arrived, I never heard anyone say D-O-S/V-S; it was DOS/V-S. WarrenFW (talk) 06:52, 15 August 2020 (UTC)


 * To clarify, I've heard both DOS and D-O-S for DOS/360 "Big Oz" and O-S for OS/360; I've always heard OS/VS spelled out. Unfortunately, pronunciation can vary from shop to shop. Shmuel (Seymour J.) Metz Username:Chatul (talk) 06:20, 16 August 2020 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on DOS/360 and successors. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20120503124243/http://www-05.ibm.com:80/de/entwicklung/history/1970.html to http://www-05.ibm.com/de/entwicklung/history/1970.html

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 20:47, 4 December 2016 (UTC)


 * I found where the page moved, and updated the link. Guy Harris (talk) 21:58, 4 December 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on DOS/360 and successors. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20071220191855/http://www.os390-mvs.freesurf.fr/mvs360.htm to http://www.os390-mvs.freesurf.fr/mvs360.htm

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 01:44, 3 September 2017 (UTC)

Transients
I propose replacing the Transients section with the following, which puts all the A-Transient info together, followed by all the B-Transient info (first sentence is unchanged):

To further reduce memory usage the supervisor employed overlays called transients that were read into one of two reserved transient areas as required:


 * Physical transients were loaded into the 556 byte A-Transient area to handle hardware errors (ERPs), record error-specific data (OBR/MDR) on SYSREC, and issue error messages. All A-Transient module names began with $$A.
 * Logical transients were loaded into the 1200 byte B-Transient area to provided common program services like OPEN and CLOSE for LIOCS. All B-Transient module names began with $$B.

The use of $$A and $$B prefixes ensured rapid loading of transients because their names were stored first in the directory.

DOS/VS added Channel Check Handlers, which were another set of transients all starting with $$RAST and executing in their own transient area. Before this addition, channel errors caused termination of the program accessing the device at the time of the error.

I don’t intend to include this, but they should have reversed the $$A and $$B prefixes so the logical transient names came first. OPEN and CLOSE were regular occurrences, while ERP/OBR/MDR were initiated only after a hardware error, time was already lost because of the error, and some of those errors were not recoverable: bus out check, data check, equipment check, overrun, etc.

I put the CCH transient info at the end because it was added as part of the reliability, availability, and serviceability (RAS) enhancements with System/370 and I can’t remember the size of that transient area. WarrenFW (talk) 08:15, 15 August 2020 (UTC)