Re: Image quality on the web

Daniel W. Connolly (
Wed, 16 Nov 1994 22:29:11 +0100

In message <199411161523.AA06184@tap.pcd.Kodak.COM>, Chris Dobbs writes:
>A colleague of mine recently offered the following observation:
>> [It was] discovered that Mosaic imposes a 50 color/image stipulation.
>> So, if an image has more than 50 colors in it, the Mosaic
>> browser reduces the number of colors to 50--it utilizes a
>> median cut algorithm.
>I believe he was referring to NCSA Mosaic, but I have yet to
>confirm this.
>My question to this list: Is this a widespread limitation of
>web clients? Due to some shared source library which enforces it?

It is a widespread limitation of X clients. The X color model is, to
be polite, inconvenient. As an X client, you can either share the
public colormap, and hope that you can allocate the colors you need,
or you can install your own colormap, and invoke the technicolor

Most commercial X apps try to use the public colormap, and use it
conservatively. For example, I worked on the AVS project a while ago,
and when AVS started up, it would try to allocate a 6x6x6 colorcube,
and if that failed, I think it would try to get a 4x4x4, and if that
failed, it would say "you gotta be kidding!" and give up. There
was a command-line option to instruct AVS to use a private colormap.

I think netscape does something similar. And FrameMaker, as I recall.

I don't think NCSA Mosaic's color policy is very flexible. But then, I
don't expect folks that give away source code to be all things to all
people ;-)

>More generally, I would be interested in any reference material, specs,
>or guidelines for the handling of image display in clients.

For X specs,
See: Information about X

Meanwhile, the Mac and Windows platforms have completely independent
color policies. And while these are the three main platforms, nobody
said that they are the only platforms in the world.

For guidelines, there's the Motif styleguide and Apples User Interface
Guidelines, thought I haven't read their color recommendations carefully.
I expect that Microsoft provides some specs/guidelines in this area too.

Hmmm... I bet there are some papers in the Computer-Human Interface
journals on this sort of thing. There are several excellent techreport
indexes on the web that I don't have ready access to right now. Sombody
want to a little surfing and see what there is to see?

>I would also be interested in any such material on image quality standards
>for the web. My sense is that this does not exist.

My sense agrees with your sense.

> While there may be
>agreed to standards for file format, the issue is larger that that. If
>I get confirmation that this is relatively uncharted territory, I may
>start an effort to address it.

Go for it! I STRONGLY encourage you to scope out a set of issues and
start an open forum to address them. Start here on www-talk, if you
like, and then migrate it to a separate mailing list (please archive

>The prevalence of imaging on the web is large and likely to grow. Given
>all the attention paid to the quality of text display, it seems appropriate
>that we should do the same for images.
>Chris Dobbs
>Imaging Systems Engineer
>Eastman Kodak Research and Development
>p.s. If I should address these questions to another list, please educate me.

It sounds to me like you'll be doing the educating on this topic.