User talk:Vt320/Archives/2021/July

ArbCom 2020 Elections voter message

Hello! Voting in the 2020 Arbitration Committee elections is now open until 23:59 (UTC) on Monday, 7 December 2020. All eligible users are allowed to vote. Users with alternate accounts may only vote once. The Arbitration Committee is the panel of editors responsible for conducting the Wikipedia arbitration process. It has the authority to impose binding solutions to disputes between editors, primarily for serious conduct disputes the community has been unable to resolve. This includes the authority to impose site bans, topic bans, editing restrictions, and other measures needed to maintain our editing environment. The arbitration policy describes the Committee's roles and responsibilities in greater detail. If you wish to participate in the 2020 election, please review the candidates and submit your choices on the voting page. If you no longer wish to receive these messages, you may add {{NoACEMM}} to your user talk page. MediaWiki message delivery (talk) 02:57, 24 November 2020 (UTC)

Copying within Wikipedia

Thanks for identifying the source of the material in your edit.

This type of edit does get picked up by Copy Patrol and a good edit summary helps to make sure we don't accidentally revert it. However, for future use, would you note the best practices wording as outlined at Wikipedia:Copying_within_Wikipedia? In particular, adding the phrase "see that page's history for attribution" and including a link to the source article helps ensure that proper attribution is preserved.

I've noticed that this guideline is not very well known, even among editors with tens of thousands of edits, so it isn't surprising that I point this out to some veteran editors, but there are some t's that you need to be crossed.S Philbrick(Talk) 13:28, 2 January 2021 (UTC)

Noted, will keep this in mind in future when moving text between articles. Thanks Vt320 (talk) 22:33, 4 January 2021 (UTC)

Mica not a microkernel - DEC Prism currently says it was

I've added a "citation needed" note to the claim in DEC Prism that it was; nothing in the Mica documents on bitsavers.org seem to indicate that it was. You might want to take a look at that page. Guy Harris (talk) 22:10, 4 January 2021 (UTC)

Thanks, I am currently working on that page to add a write up of Mica, along with references to the appropriate Bitsavers documents. As you've pointed out, nothing in the documentation on Bitsavers refers to it as a microkernel. IMHO, Mica, is more a hybrid kernel since its Executive layer ran in kernel mode. However, my preference is to avoid categorizing the kernel type, since that is mostly a matter of opinion. Vt320 (talk) 22:28, 4 January 2021 (UTC)

"...that is mostly a matter of opinion". Yup. I tend to agree with Linus Torvalds' opinion of the term "hybrid kernel". Having worked for many years in the Core OS group at Apple, I'd say XNU is not all that different from Boring Old Monolithic Kernels on which I've worked; yes, Mach messaging was used in some cases to communicate between kernel code and user-mode daemons, but if that makes it a "hybrid kernel", then RSX-11M had a hybrid kernel, what with F11ACP, right? (And autofs, to mention one project I worked on only used Mach messaging because we had it available but didn't have a good ONC RPC in the kernel; Solaris, which is generally not considered a "hybrid kernel", used the in-kernel ONC RPC to communicate with automountd.)

Perhaps more of a case could be made for Mica or NT, but I'm not sure how much of a case. My inclination is to call kernels such as L4 microkernels, as they are truly small kernels, even if a lot of the rest of the OS runs in privileged mode, not use microkernel for something as big as Mach, and avoid "hybrid kernel" entirely.

(BTW, is it "Mica" or "MICA"? Looking at the bitsavers documents, the engineering documents seem to use "Mica", but the Software Business Plan uses MICA.) Guy Harris (talk) 03:03, 5 January 2021 (UTC)

References

Disambiguation link notification for January 5

Hi. Thank you for your recent edits. An automated process has detected that when you recently edited DEC Prism, you added a link pointing to the disambiguation page Pascal. Such links are usually incorrect, since a disambiguation page is merely a list of unrelated topics with similar titles. (Read the FAQ • Join us at the DPL WikiProject.)

It's OK to remove this message. Also, to stop receiving these messages, follow these opt-out instructions. Thanks, DPL bot (talk) 06:24, 5 January 2021 (UTC)

I renamed VAX Macro to VAX MACRO

As you note, that's what DEC called it, so that's what Wikipedia should call it. Guy Harris (talk) 23:54, 26 January 2021 (UTC)

• Thanks, was considering doing it myself :) Vt320 (talk) 23:56, 26 January 2021 (UTC)

Orphaned non-free image File:Vms-arrow-logo.jpg

⚠

Thanks for uploading File:Vms-arrow-logo.jpg. The image description page currently specifies that the image is non-free and may only be used on Wikipedia under a claim of fair use. However, the image is currently not used in any articles on Wikipedia. If the image was previously in an article, please go to the article and see why it was removed. You may add it back if you think that that will be useful. However, please note that images for which a replacement could be created are not acceptable for use on Wikipedia (see our policy for non-free media).

Note that any non-free images not used in any articles will be deleted after seven days, as described in section F5 of the criteria for speedy deletion. Thank you. --B-bot (talk) 19:47, 7 February 2021 (UTC)

Disambiguation link notification for February 10

An automated process has detected that when you recently edited OpenVMS, you added a link pointing to the disambiguation page Erlang.

(Opt-out instructions.) --DPL bot (talk) 06:17, 10 February 2021 (UTC)

Orphaned non-free image File:SCO OpenServer.png

⚠

Thanks for uploading File:SCO OpenServer.png. The image description page currently specifies that the image is non-free and may only be used on Wikipedia under a claim of fair use. However, the image is currently not used in any articles on Wikipedia. If the image was previously in an article, please go to the article and see why it was removed. You may add it back if you think that that will be useful. However, please note that images for which a replacement could be created are not acceptable for use on Wikipedia (see our policy for non-free media).

Note that any non-free images not used in any articles will be deleted after seven days, as described in section F5 of the criteria for speedy deletion. Thank you. Vulphere 14:43, 11 March 2021 (UTC)

Disambiguation link notification for March 13

An automated process has detected that when you recently edited IBM System/38, you added a link pointing to the disambiguation page Kernel.

(Opt-out instructions.) --DPL bot (talk) 06:26, 13 March 2021 (UTC)

Disambiguation link notification for March 21

An automated process has detected that when you recently edited IBM i, you added a link pointing to the disambiguation page Kernel.

(Opt-out instructions.) --DPL bot (talk) 06:13, 21 March 2021 (UTC)

FS and S/38

The "Opel Task Force Report" draft from September 15, 1971 describes, on pages 10 through 12, some characteristics similar to what's in S/38:

The New Machine Interface is the lowest such complete interface defined. It is the focus for hardware CPU design analogous to (and quite similar to) the S/370 Principles of Operation.

The Execution Discipline Interface, which is built by components running on the NMI, is the "machine" on which most applications programs actually run. It is analogous to the S/360 Principles of Operation for non-privileged instruction, plus OS/360 Control Program Facilities. However, some of the facilities offered are at a much higher level than in OS (e.g., a one level store, logical source/sink I/O, extensive work control primitives).

The Application Development Interface is the lowest machine the application program sees. It is built by compilers which translate (and optimize) ADI programs to EDI programs, and/or by interpreters which execute ADI programs directly. ADI is the target interface for translators from high level languages (e.g., COBOL, PL/I, FORTRAN) and from special purpose application languages. It is, therefore, the most exposed, most durable of the FS interfaces.

The NMI sounds similar to the IMPI instruction set (down to the S/370ishness). The EDI sounds like IMPI plus the OS part of the VMC/SLIC (as opposed to the binary-to-binary translation or database parts). The ADI sounds like MI, with the "compilers which translate (and optimize) ADI programs to EDI programs" being the binary-to-binary translation part of the VMC/SLIC, without the "interpreters which execute ADI programs directly", which I don't think existed in S/38 or exist in IBM i.

The report also discusses the single level store concept.

I'm not sure whether that could be used as a reference for FS having similarities to S/38 or not. Guy Harris (talk) 01:53, 22 March 2021 (UTC)

Thanks for the reference. There is no doubt that the S/38 shares many of the characteristics of the FS project, and indeed some of the S/38 architects such as Soltis were involved in the FS project. In his book, Soltis claims that the use of object-based addressing in the MI was taken directly from the FS project, as was the term "Machine Interface" (although not necessarily the concept itself). He also mentions that there was a debate as to whether the FS NMI should be a System/370-style instruction set, or something which was higher level. Soltis was on the S/370 side, and given that he designed the IMPI, I'm sure some of the lessons from FS were in his mind.

The most interesting question to me is just how much of the S/38 was conceived in-house at Rochester, and how much was inspired by the FS. Vt320 (talk) 09:55, 22 March 2021 (UTC)

Disambiguation link notification for April 9

An automated process has detected that when you recently edited OpenVMS, you added a link pointing to the disambiguation page Azure.

(Opt-out instructions.) --DPL bot (talk) 06:13, 9 April 2021 (UTC)