On Fri, Sep 18, Andrew Shadura wrote:
> Karl E. Jorgensen wrote:
>> ''You should include "event" here, but you must not do so.'' ????
>> Huh?? To include or to not include.... I'm confused, and I think most
>> other readers would be too :-)
>
> I'm not sure I understand what exactly is confusing you :)
I'm not the bug submitter, but I see what's confusing him. Just as he
says, the manual page for brightd has some rather confusing text in the
description of the -e option:
# -e n
#
# Filter used event sources by POSIX extended regexp n (for example,
# use "i8042.+event" on intel platforms to avoid having HDAPS taken
# into account) You should include "event" here, but you must not do
# so.
Look at that last sentence.
# You should include "event" here, but you must not do so.
Grammatically, that clause "you must not do so" can be interpreted
only one way: it's saying
# you SHOULD: include "event" here, but
# you MUST: not include "event" here.
In other words, this action is both requested and prohibited.
I suspect that the writer of the manual got confused about the correct
way to negate a "must", and intended it to mean:
# You should probably include "event" here, but it is not mandatory.
The secondary problem is that it's completely unclear how I would
"include 'event' here" - does "event" mean the literal string 'event'
or the name of an ACPI event type or what?
The whole man page is full of grammatical problems, but the rest of
them are less serious than this.
--
JBR with qualifications in linguistics, experience as a Debian
sysadmin, and probably no clue about this particular package
