bug-lilypond
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Barline scans


From: Marnen Laibow-Koser
Subject: Re: Barline scans
Date: Wed, 10 Jan 2007 10:22:32 -0500
User-agent: Mozilla Thunderbird 0.9 (Macintosh/20041103)

Han-Wen Nienhuys wrote:
Marnen Laibow-Koser escreveu:
[...]
Nope, wrong again.  My original judgement came in fact from viewing the
PDF file.
[...]
The fanciest computer displays are 142 dpi (1600x1200 14" laptop
screen).  This is not good enough for judging typography.

Augh! The moment I posted my earlier remark, I knew you were going to say this. I neglected to mention that I did in fact magnify the PDF file to a point where I could be sure that the proportions I was seeing were accurate. I often do this to judge typography, and it has served me very well in other cases.



To summarize:

 - Yes, LilyPond uses heavy barlines, at approx. 1.9
stafflinethickness (20 pt staff height, 0.5pt stafflinethickness)

 - Whether you think that's ugly or not, is a personal opinion.

Yes. "Ugly" is a personal opinion. "Out of the mainstream", however, is not a personal opinion, but is a measurable statistic.

Unfortunately, our samples of the "mainstream" have so far been too small to be conclusive -- Jan's sample shows one thing, while mine shows another. Clearly more research is needed.


 - On the factual side, this follows Baerenreiter's house style, a
publisher whose material widely regarded as excellent.

I have mentioned several times during this discussion that I am among Bärenreiter's fans. I hope I can find a suitable Bärenreiter edition to measure.

<devil's-advocate>
However, it is also true that Bärenreiter is very distinctive -- not to say unusual -- in many aspects of its graphical appearance. While beautiful, it may be *too* idiosyncratic to be a suitable model. Perhaps an edition such as Henle or Wiener Urtext would make a better model. But that's a whole other discussion.
</devil's-advocate>


 - We aim for high quality *printed* output; some of the finer details
of the font (eg the stem of the flat symbol) require a 1200 dpi
printer to be visible.

I realize that, and I very much appreciate such attention to detail. But that doesn't mean that lower resolutions should be completely ignored. The state of modern digital typography is such that we can do both.


 - Therefore, to assess output output quality, we use laser
print-outs.  If you want to partake in a discussion, I invite you to
do so based on laser printouts, preferably printed at 1200dpi or
better.

Of course laser printouts are better for this purpose, but not everyone is lucky enough to have a 1200-dpi laser printer sitting around. My inkjet printer produces excellent output with other software (and I'm pretty picky); I have no reason to think it magically loses this capacity with Lilypond. Sure, ragged edges will show up in a 2400-dpi scan, but that's not the main point here.

Besides, you yourself have produced measurements that confirm that the measurements I took from my inkjet output are quite accurate (i.e., you confirmed my 191% figure and my staff-line measurements pretty exactly). So your implying that my figures are somehow invalid is rather disingenuous, I think, given that they correspond exactly with yours. My printer is certainly not ideally perfect, but neither is it ideally imperfect.

The issue here is not the quality of my printer, but Lilypond's typographic decisions, which I have also confirmed through high-resolution examination of PDF files, as I mentioned above. I thought we had a good substantive discussion of these points starting; methodological quibbles can't make *everything* go away.

Best,
Marnen





reply via email to

[Prev in Thread] Current Thread [Next in Thread]