Message-ID: <4e4f98b9.0806e30a.0505.ffffbdb3@mx.google.com>
Content-Type: multipart/mixed; boundary="===============0355875053052718243=="
MIME-Version: 1.0
In-Reply-To: <87hb5duraz.fsf@freedomboxfoundation.org>
References: <8739gyw0zh.fsf@opensourcematters.org>
	<87sjoxxmcw.fsf@servo.factory.finestructure.net>
	<87hb5duraz.fsf@freedomboxfoundation.org>
Subject: Re: [PATCH] Added --initial-index and --last-index to search/show
To: notmuch@notmuchmail.org
From: Patrick Totzke <patricktotzke@googlemail.com>
Date: Sat, 20 Aug 2011 12:21:26 +0100
Precedence: list

--===============0355875053052718243==
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable



Hi!
A very good idea indeed! This could become quite handy speeding up my inter=
face.
One question: What do you (intend to) do to ensure that accumulated results
'add up well'? =

Consider the following: you query some 10 threads, the user tags one of them
so that it doesn't macht anymore and afterwards you list the next 100 threa=
ds.
After the tagging, the number of hits decreases, so if you query
from hit number 101 to 200, you will not return the 100th (now 99th) thread=
 at all.

Ok, one could always list all threads up to the one one is interested in bu=
t that
would make refreshing the list slower and slower when you keep scrolling do=
wn..
best,
/p

--===============0355875053052718243==--