Re: top-replies was Re: [PLUG] kernel parameters for memory usage?

[gabriel rosenkoetter <gr@eclipsed.net>]

[Dropping reference to Paul's Message-ID so that my response doesn't
get added to the kernel parameters thread by MUAs that pay attention
to the In-Reply-To header. Keeping Bob's Message-ID because his is
already broken off.]

This is getting to be rather petty and, while top-replies really do
irritate me, I don't care this much.

I complained in that email because I was replying to five or six
emails at the same time (hint, use tag before reply in mutt) and
was having to delete a LOT of useless text to compose my reply.
Sure, that's my problem... but do you really want to irritate people
who you hope will reply to a question you've asked?

On Tue, Nov 18, 2003 at 04:19:04PM -0500, Paul wrote:
> >1. Good MUAs (and news readers) provide a "skip past quoted
> >material" function.
> Nice feature.  I wonder if Mozilla can skip quoted text. 

I did say "good MUAs". Mozilla is, at best, a web browser of
questionable quality...

> I use blank spaces for clarity.

Well, actually, the blank spaces you use come up with a quote
character preceding them when I read them (maybe because you just
started adding newlines right at the end of the quoted text?).
I suppose that's actually kind of handy for skipping quoted text,
but it distracts my eye while the quoted text is still on the
screen.

> How many bytes does a blank space contain?

I was referring to eyeball bandwidth, not network bandwidth.

> I'm guessing one or two bytes.

Just to be academic, you're right. It's one for the >, one for the
newline. Not enough to matter here.

(It's the "quoting more text than what you're actually replying to"
that's a waste.)

> The goal is communication and expression as opposed to 
> efficiency and absolute correctness.

I'd argue that the first two are hard without the third and an
attempt at the fourth. But whatever.

Bob, I don't think any of your arguments hold water.

On Tue, Nov 18, 2003 at 05:39:40PM -0500, Bob Schwier wrote:
> 1.  If the person reading the reply needs info from the
> original, it's there.

It's in the mailing list archives if they need it. And if they want
to reply to the text of the original message, then they should be
replying to the original message.

We have a problem if the person doing the quoting didn't quote (or
cite!!!) sufficiently, but that's a different error and one that
shows an equal disrespect for future readers as quoting too much or
in an inconvenient place.

> 2.  If the individual doesn't need it, he or she can get the new
> info and delete quickly.

Unless that individual's trying to reply (and, say, be helpful), in
which case they have to slog through all that extra BS in an editor,
unless they want to perpetuate the long trailer of useless cruft.
It's disrespectful at best to lay that on your correspondents.

Also, there's more than just what you're responding to at the end of
those messages, especially on PLUG: there's the mailing list footer.

> 3.  Some of us have precious little time to get through the messages.

All the more reason that the information should be presented
concisely and in an order sensible for reading. The onus is on the
one sending the message to trim the quoted material to being what's
relevant.

> If the message is purely text, this is not really a creator of
> bandwidth problems.  The problem comes from those messages include
> attachments.

That's simply not true. For one thing, I don't recall at any point
specifying "network bandwidth". The problem is also visual and
mental (more data to process for the receiver).

Also, this unnecessary few lines of text builds up over time, both
over the course of a thread (the last message in the kernel
parameters thread that I replied to was about eight times as long
as it needed to be, all bottom-quotes) till it actually *is* a
relevant quantity of data, and bottom-quotes waste space long-term
on the mail server storing and forwarding the mailing list. (Let's
say a message that could have been 40 bytes is, because of bottom-
quotes, 360 bytes. That's 320 bytes times the number of PLUG
subscribers, which we'll pretend for now is 300. Now we're up to a
megabyte in network bandwidth... and that's just one message.)

But, above all else, bottom-quotes are CONFUSING! We don't naturally
read documents from the bottom to the top, we read them from the top
to the bottom. The problem is concisely summed up in this classic
.signature:

A: No.
Q: Should I include quotations after my reply?

-- 
gabriel rosenkoetter
gr@eclipsed.net

Attachment: pgpABVAwvqamE.pgp
Description: PGP signature