• Re: Usenet quoting conventions

    From Alan Mackenzie@acm@muc.de to news.software.readers on Sun Apr 5 14:12:09 2026
    From Newsgroup: news.software.readers

    Thomas 'PointedEars' Lahn <PointedEars@web.de> wrote:
    Alan Mackenzie wrote:
    Thomas 'PointedEars' Lahn <PointedEars@web.de> wrote:
    Alan Mackenzie wrote:
    Thomas 'PointedEars' Lahn <PointedEars@web.de> wrote:
    [Please put quotation prefixes for empty quoted lines, too. Fixed
    below.]

    You don't give any reason why, beyond, presumably, your personal
    preference. Is there any other reason?

    The reason is that with all newsreaders that do not follow *your*
    (newreader's) convention (you[rs] appear[s] to be the unconventional one >>> here), empty lines in quoted text end up as empty lines in a first-level >>> quotation instead of the quotation level to where they belong.

    You didn't address the points I made about deleting paragraphs and
    readability. That's disappointing.

    I have addressed there the question that you asked. I do not see the logic in your subsequent argument: Paragraphs can be trimmed in any case, and you are not actually making it easier, ....

    In vim, the editor I use for Usenet, the command to delete a paragraph
    forward is d{. This is far easier than marking the lines of a paragraph
    by hand, then deleting them. This only works when the paragraphs are
    delimited by blank lines. Possibly some vim guru will tell me how to set
    the paragraph delimiter to ^>[ >]*$, but that doesn't seem helpful.

    .... but *harder* as one has to look even more carefully who wrote
    what, with the interleaved quotation levels that your modification
    produces.

    I dispute this. The quotation levels one looks at are those of the
    paragraphs, not the delimiting lines between them. Surely?

    All you seem to be saying is that (some) other people quote differently.

    In my (about 25 years of) Usenet experience, people who remove already
    posted quotation level indicators are in the minority instead.

    I've also been posting over 25 years on Usenet. I suspect the real
    reason few people trim the inter-paragraph gap is that their editors
    aren't up to the job. With vim, it is no trouble at all.

    You aren't citing an RFC, or some canonical written standard somewhere.

    I am not aware of a network (quasi-)standard regarding this; it is based on what I observe as "best current practice".

    You may be confusing "best" practice with "most common" practice.
    (Anybody with a few decades experience of software development
    understands the difference.) You've been trying for the last few posts
    to come up with a reason why the common practice is better. I don't
    think you've managed it, yet.

    But these two manuals are pretty convincing:

    <https://www.netmeister.org/news/learn2quote.html> <https://einklich.net/usenet/zitier>

    (From your message headers and your signature I presume that you can read German without a translator, too.)

    Yes, indeed. I spent some time perusing these documents. They didn't
    seem to touch on the issue we're discussing.

    By removing quotation level indicators that others wrote, you are
    *modifying* quotations without good reason and without indication.

    No. Others didn't write these quotation marks. Their newsreaders
    inserted them automatically for them. I'm not removing the >s from
    people's text. I'm removing them from the delimiters between the pieces
    of their text.

    This, I think, is the main difference between us. You regard the gaps
    between paragraphs as being an integral part of the following paragraph
    (why not the previous one?), whereas I regard them just as separators.

    Compare this with the space between adjacent sentences. The full stop at
    the end of the sentence is an integral part of it. The one or more
    spaces between it and the next sentence are not. You'll note that before
    the first sentence of a paragraph and after the last one, there are no
    such spaces. They're pure separators between sentences.

    I put it to you that the blank lines between paragraphs are likewise just separators. When you are answering a Usenet post, and trim the first few paragraphs, you presumably don't take care that the first remaining
    quoted paragraph starts like:


    Text ....

    . Instead you would trim the separator, leaving just:

    Text ....

    I don't see there is any need for these separators to be quoted.

    Regarding mathematics, occasionally you might accidentally remove "(much) greater-than" signs than have been word-wrapped to the beginning of a line. They are of course a problem in and of themselves as they *could* be
    mistaken for a quotation level indicator by the reader; but this situation *can* happen. [Depending on what you remove from the beginning of a line, occasionally you also might remove vertical bars from "divides" statements
    or set definitions (which, as you can see below, are used as indicators for
    a quotation from a third source or to avoid ambiguity -- in my example, the latter).]

    But if nothing else convinces you, then it should be that you should not modify quoted text without good reason; and if you do modify it, you should indicate that (which you are not).

    Again, I am not modifying the text. I am modifying the delimiters
    between the pieces of it. (I also refill text reasonably often when the
    level of quoting has made it too wide.)

    This is particularly (but not only) confusing when the quotation level
    is indicated by colors.

    Where is the confusion?

    Here, depending on how often a quotation has been quoted, and who participated in the discussion:

    | qrs wrote:
    | > klm wrote:
    | >> abc wrote:
    | >>> xyz wrote:
    | >>>> tuv wrote:
    | >>>>> text
    | >>>>> text
    | >>
    | >>>>> text
    | >>>
    | >>>> text
    | >>>> text
    | >
    | >>> text
    | >>
    | >>> text
    | >
    | > some text

    I don't think the paragraph separators are any less easy to read than if
    they had more >s in them. They would be easier still to read if they
    were just blank lines.

    You can also see a real-world example at the top, where I have deliberately not restored the quotation level indicators that you removed.

    If there is any, surely it would be best solved by not colouring lines
    which are empty bar the quote marks and spaces.
    By contrast to what you are doing, that behavior cannot be modified in the majority of cases. But AISB, ....

    AISB???

    .... even if the lines are not colored differently, there is still the
    issue that the purpose of a quotation level mark to indicate to the
    reader what belongs to which quotation level is lost by your
    modification: The quotation level is "jumping" pointlessly from one
    paragraph to the next even when adjacent paragraphs belong to the same
    level.

    How does that compare with the optical confusion caused by what should
    be distinct paragraphs being merged by vast bars of ">>> >>
    ...."?

    | qrs wrote:
    | > klm wrote:
    | >> abc wrote:
    | >>> xyz wrote:
    | >>>> text
    | >>>> text
    | >>>>
    | >>>> text
    | >>> text
    | >>>
    | >>> text
    | >> text
    | >> text
    | > text
    |
    | text

    My eyes, at any rate, can easily follow this way from top to bottom who
    wrote what (aided by the coloring, too). What about yours?

    No problem, but I would find it easier with blank lines. Where quoted separators become more tedious is when the level of quoting gets
    excessive, say 40 or 60 deep. This happens when people neglect to trim
    the posts they're answering.

    [ .... ]

    You are welcome, and thank you, too.

    Like the rest of a message, signatures are not to be quoted unless referred to. A proper MUA/NUA automatically trims signatures like mine which have been separated from the rest of the text by a line starting with, and containing only, "-- ". Again, I am not aware of a network (quasi-)standard regarding this (RFC 1855 "Netiquette Guidelines" is not it; it talks about signatures, but only that they should be short), but it is widely
    implemented so.

    In a moderately long post (say ~150 lines), I think it can be useful to
    see who the previous poster was when near the bottom of the post. That's
    why I normally leave one level of sig in.

    I think that if we continue this discussion, we should continue it in <news:news.software.readers>. F'up2 set accordingly.

    Accepted.

    --
    PointedEars

    Twitter: @PointedEars2
    Please do not cc me. / Bitte keine Kopien per E-Mail.
    --
    Alan Mackenzie (Nuremberg, Germany).

    --- Synchronet 3.21f-Linux NewsLink 1.2
  • From Carlo XYZ@carloxyz@invalid.invalid to news.software.readers on Mon Apr 6 08:19:41 2026
    From Newsgroup: news.software.readers

    Alan Mackenzie wrote on 05.04.26 16:12:
    Thomas 'PointedEars' Lahn <PointedEars@web.de> wrote:

    [..]

    I've also been posting over 25 years on Usenet.

    Then I suppose you can guess what (not) to expect
    from your opposite number.

    Yes, indeed. I spent some time perusing these documents.

    Ouch.
    --- Synchronet 3.21f-Linux NewsLink 1.2