|
|
Top-posting on Usenet
Top-posting is that annoying habit of writing a reply above
the question it's answering. People do it because they're too
lazy to move the cursor to the right place, and they tend to be
poor snippers too. (A good snipper removes all quoted material
from the reply that is not immediately germane to it, leaving
just enough context in place to let the recipient(s) understand
what the reply is actually about.)
I wrote the following Usenet article on 4 January 2006 in an
attempt (which was, apparently, partially successful) to illustrate
to a top-poster just why it's a bad idea. It's a reply to
his article, which attempted to defend the indefensible, but it
looks at first as if... well, you'll find out!
Message-Id: <dpfs0a$n74$1@nwrdmz02.dmz.ncs.ea.ibs-infra.bt.com>
From: Richard Heathfield <invalid@invalid.invalid>
Subject: Re: This is the end
Newsgroups: comp.lang.c
Reply-To: invalid@invalid.invalid
Date: Wed, 04 Jan 2006 07:01:02 +0000
References: <1136215822.522731.219410@g49g2000cwa.googlegroups.com>
<1136216129.229371.30140@g14g2000cwa.googlegroups.com>
<1136256053.820488.114050@g14g2000cwa.googlegroups.com>
<dpe3sc$1vn$1@news1brm.Central.Sun.COM>
<OpBuf.83077$vl2.53510@fe2.news.blueyonder.co.uk>
<dpep0j$nqb$1@nwrdmz03.dmz.ncs.ea.ibs-infra.bt.com>
<1136354026.175771.27550@o13g2000cwo.googlegroups.com>
Lines: 171
Organization: sigh - KNode crashed again
User-Agent: KNode/0.7.1
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7Bit
clayne said:
Sure it is. [Actually clayne didn't say that, but bear with me here. You'll
soon work it out, I think.] Are you not flexible enough to cope with text
that is the other way around to the way you usually see it? Don't you think
you have a duty to put up with it because I happen to like doing it that
way? Or would it make more sense for me to make my text easier to read if I
wish to communicate effectively with people rather than jar them off all
the time?
> Yea, now c'mon. This isn't even a reasonable comparison.
No, it's not the same at all. One reply is in the right place, the place
that makes it easy to follow the discussion - and the other one isn't.
> Look, whether
> someone replies on top or on bottom, it's the SAME thing.
Scope isn't all that important if you reduce the amount of quoted text to
the minimum necessary to remind people of the context of the discussion. By
reducing the amount of scope people have to deal with, you'll reduce it to
the point where it's hardly relevant. But when you start putting things the
wrong way up, that doesn't make things easier for people; it makes them
harder.
In fact, if you have a particularly long paragraph or series of paragraphs,
it may well be that, by the time you get to the end of it, the stuff you as
a top-poster left in might seem very strange indeed, since if it bears any
relevance at all to your reply, that relevance - that connection - is much
more likely to be associated with the beginning of your paragraph rather
than the end.
As an example of this, please observe the following statement (to which this
set of paragraphs is a reply), and note how your mind is jarred back to the
beginning of this set of paragraphs, rather than having the reply flowing
smoothly /from/ the prompting text.
> Scope is what matters.
Well, no, it isn't a question of thinking outside of the box; the real issue
here is that we have one person who thinks he knows best and a whole bunch
of people who have years of experience of helping people to learn better C,
and the whole bunch of people have for many years settled on a style which
they think works efficiently, allowing them to answer a large number of
questions with considerable speed.
Anything which *slows them down* is going to reduce the usefulness of
comp.lang.c - and we don't want that. (Please remember that we don't get
paid for this. We want to help people. Lots of people. But we can't spend
all day doing it.)
Top-posting certainly slows /me/ down, which is why I don't like it, and I
don't suppose other regular contributors here are much different to me in
their reasoning. And one last thing about your statement: did you not, in
your upside-down world, mean to write "the preceding" rather than "the
following"?
> The real issue here is that we have a good amount of just
> plain inflexible people who cannot think outside of the box for just 5
> minutes to realize the following:
Yes, whoever wrote that is right; it does indeed make a lot more sense to
see the question before you see the answer.
> "When I read a reply to someone... Let me first read the part with the
> quote-symbol delimited content first (">"), and then read the unquoted
> new content section."
We have. But our automatic response is to look /below/ the question for the
answer. When we don't find it there, we have to search for it consciously.
It may only take a moment longer. But some of us can do a lot in a moment.
Like, we could be answering other people's questions. (Incidentally, the
only reason I'm bothering to explain all this to you is that I'm far too
ill today to focus on writing good code.)
> If you haven't formed this into a motor function by now, I don't know
> what to tell you.
That's right, that's precisely what we don't want - but for good reasons.
> Honestly, I think the real beef is that people just
> don't want the new content showing up to their eyes before the
> context..
Well, it's not so much that, and more because it's common sense to have the
question appear before the answer.
> some kind of perverse control thing...
Is it not incumbent on the technical writer to present the reader with the
material in the order that the reader expects it, so that he will gain the
most value from it in the shortest possible time?
> ("no, let ME determine
> the order in which I read new things!" type of deal).
Try popping down to your local garage to get free tips on what you're doing
wrong in car maintenance. While you're there, straighten the guy's tie -
or, if he's not wearing one - take one and tie it on him yourself. Whenever
he says something, make sure you're standing on your head before replying.
If he thinks you're nuts, tell him to stand on his head too. Tell him he's
got Asperger's Syndrome, and is not being flexible. Tell him to learn a new
adaptive method - like standing on his head, for example. And don't forget
to act all surprised and hurt when he tells you to drop dead and refuses to
give you any more personalised free advice.
> Learn a new adaptive method,
Wouldn't it be easier for you to learn a new coping mechanism instead of
banging on about top-posting when it's obvious we're not about to adopt a
less efficient convention just because somebody asks us to?
> learn a new coping mechanism instead of pumping the NG
> full of replies telling people who to reply/quote when the issue is
> insignificant
In fact, I hardly mention it at all. (Are you wondering what I mean by "it"
yet? You wouldn't be, if I'd written this reply in the traditional way.)
And the reason I hardly mention it (what?) at all is that I know from
experience that arguing with top-posters is almost always a waste of time.
And so, instead of arguing with them, I generally *don't bother answering
their questions*. I'm not religious about it, but I generally find myself
skipping past articles that don't provide context in the place I expect to
find it.
So now what are you going to do? Insist that I engage with top-posters?
Remember you're not paying me for this. It may not sound very fair to you,
but I don't recall seeing anything about "fair" on the box lid. There's
nothing in the rules to force me to engage with clueless people.
Maybe you'll argue that, by refusing to engage with top-posters, I'm not
learning anything from them. Well, that's fine by me because I tend to be a
giver of advice here rather than a recipient, and in any case all the
clueful regular contributors here know how to write Usenet articles
properly, so if ever I do have a question about C, there are plenty of
knowledgeable people ready to answer it, without my having to rely on the
(generally flawed) knowledge of people who can't even learn which way round
a question and answer should go. So - no loss there either, from my point
of view.
Maybe there are other people in this newsgroup who have the same strategy as
me: silently ignoring most top-posted articles. I don't know. But if there
are, then top-posting in comp.lang.c becomes a way to reduce significantly
your chances of getting good-quality help and advice.
Keith Thompson is actually doing newcomers a service by drawing their
attention to the customs and mores of the regular contributors to this
group. It's a thankless task, on the whole, and one which he has been
performing quietly for quite a while, as well as helping people to
understand C better. Although I have never met Keith, I feel as if I have
come to know him quite well, and I would just like him to know how much I
appreciate his contributions to this newsgroup.
> (granted, I don't see you as a significant contributor to
> this, I see Keith Thompson as the main curator).
I don't know. I never heard of anyone by that name. <shrug> For your
information, though, Stefan Wilms was the founder of the campaign against
excessive grumpiness in comp.lang.c.
> But probably not with Dick Heathfield, right?
That was exhausting and confusing. I hope I never have to write another
reply like it.
--
Richard Heathfield
"Usenet is a strange place" - dmr 29/7/1999
http://www.cpax.org.uk
email: rjh at above domain (but drop the www, obviously)
And that's it, really. Apparently at least one person thought
this response sufficiently eloquent to be worth referring to,
so to save you looking it up in Google Groups, I've copied it here.
|
|