- From: Adam M. Costello <png-group.amc+0@nicemice.net>
- Date: Tue, 3 Jun 2003 21:55:58 +0000
- To: png-group@w3.org, Susan Foord <susan.foord@ntlworld.com>
Glenn Randers-Pehrson <glennrp@comcast.net> wrote:
> change parts of PNG second edition, Clause 4.4, PNG Image, indicated
> by the square brackets:
>
> c. Truecolor: ... The alpha channel may be represented by
> a single [pixel] value. Matching pixels ...
> d. Greyscale: ... The alpha channel may be represented by
> a single [pixel value as in] the previous case. ...
>
> ... size of each sample, not the total [pixel size].
>
> to read:
>
> c. Truecolor: ... The alpha channel may be represented by
> a single RGB value. Matching pixels ...
> d. Greyscale: ... The alpha channel may be represented by
> a single greyscale value, similar to the previous case. ...
>
> ... size of each sample, not the total number of bits per pixel.
The "pixel value" phrase appears again in section 6.2, but the same
fix cannot be applied there, because the truecolor/grayscale cases are
combined, like so:
b. Truecolour, greyscale: A tRNS chunk contains a single pixel
value...
How about if we leave 6.2 as-is, but prepare the reader for it by
keeping the word "pixel" in section 4.4, like so:
c. ...by a single RGB pixel value. ...
d. ...by a single greyscale pixel value. ...
Also, I like the replacement of "size" by "bits per", so much that I'd
like to do it not only for "pixel", but also for "sample":
...the bit depth specifies the number of bits per sample, not the
total number of bits per pixel.
I've added these suggestions to my own growing collection.
AMC
Received on Tuesday, 3 June 2003 17:57:00 UTC