Rockbox.org home
release
dev builds
extras
themes manual
wiki
device status forums
mailing lists
IRC bugs
patches
dev guide



Rockbox mail archive

Subject: Re: tomers: r24174 - in trunk/rbutil/rbutilqt: . base

Re: tomers: r24174 - in trunk/rbutil/rbutilqt: . base

From: Dominik Riebeling <dominik.riebeling_at_gmail.com>
Date: Mon, 4 Jan 2010 20:19:41 +0100

On Sun, Jan 3, 2010 at 9:17 PM, <mailer_at_svn.rockbox.org> wrote:
> Date: 2010-01-03 21:17:36 +0100 (Sun, 03 Jan 2010)
> New Revision: 24174
>
> Log Message:
> rbutil: Fix some messages

What on earth is this a "fix"? It doesn't fix anything. The strings
touched where completely correct, and the only thing changed is
punctuation and capitalization. Furthermore, from my point of view the
punctuation changed was absolutely correct, thus there wasn't even a
need to change it -- the change made it worse.

> -        emit logItem(tr("Download finished."), LOGOK);
> +        emit logItem(tr("Download finished"), LOGOK);

Why is the trailing dot removed here? This is a (rather short)
statement, so it should end with a punctuation character, i.e. a dot.

However, if such kind of messages should not end with a dot, then
please make it actually consistent. For example, there is
    emit logItem(tr("Download error: received HTTP error %1.")
which should, following the change, also not end with a dot. There are
also other log messages that (still) end with a dot. However, this
wasn't changed by this commit. Nevertheless, if we want log messages
to not end with a dot we need to agree upon it. Which hasn't happened
up to now.

> -        emit logItem(tr("done."), LOGOK);
> +        emit logItem(tr("Done"), LOGOK);

This "done" comes in a logical way: it completes a sentence /
statement. As in "Starting procedure <something> ...", then "done.".
Thus it (a) has to be lower case (it's still *in* a sentence) and end
with a dot as it ends the logical sentence. This also applies to all
other appearances of "done" that got changed.

> -    ui.themePreview->setText(tr("fetching preview ..."));
> +    ui.themePreview->setText(tr("Fetching preview..."));

Ellipses seem to be handled inconsequently in computer interfaces.
However, I disagree with this change as well. From my point of view an
ellipsis should *not* be "glued" to the previous word if it's
replacing a complete word (or part of a sentence). This is different
if it continues a word that has been cutted of -- for example as in
"What kind of c... is this?" as opposed to "This is really ...". The
ellipsis has two distinct functions here: replacing part of a word and
replacing part of a sentence. The changed string, however, used it to
replace a part of the sentence (that could be continued later, for
example with an error message or, which should usually be the case for
Rockbox Utility, a "done." message). Thus, from my point of view the
space before the ellipsis is completely correct. My Firefox tabs, for
example, behave exactly this way.

While it's a good thing to make strings more consistent this is IMO
the completely wrong direction. Nevertheless, *please* discuss such
changes first. It's creating more work (and is much more annoying) to
clean things up later.

 - Dominik
Received on 2010-01-04


Page was last modified "Jan 10 2012" The Rockbox Crew
aaa