FYI: my FISL slides

Joshua Judson Rosen rozzin at geekspace.com
Wed Jun 29 22:58:24 EDT 2011


Werner Almesberger <werner at almesberger.net> writes:
>
> Joshua Judson Rosen wrote:
> > For a safer pallet, cf.:
> > 
> >     http://jfly.iam.u-tokyo.ac.jp/color/#pallet
> 
> Hmm, seems that there are no really "safe" colors :-(
> I've changed the shapes and uploaded the new version.

The table works *much* better, now--thank you! :)

Looking at XFig.., I see what you mean--there are 8 safe colours
defined in the guide, but it looks like there's no way to tell XFig
to use even the 5 that you need; it looks like the FIG file-format
apparently supports user-defined colours, but I don't see a way
to access that via XFig. Inkscape can apparently open  the fig file
for editing, but can't write data back out in fig format.

> looks more "nervous" now. But should be unambiguous.

Yes. I should have remembered an example that I could give:
Consumer Reports uses quite a nice set of colour-safe icons
(with some redundant colour-coding), all of which are
uniformly circular, for their ratings; this image is from
the front page of consumereports.org:

    http://static1.consumerreportscdn.org/etc/designs/cro/application-resources/modules/nav/images/home_find_ratings_header.png

The circles indicate, from left to right, something like:

    `excellent', `above average', `average', `below average', `terrible'

(hey--that's 5..., but I bet there's a trademark issue or something...)


Though, in the particular case of your chart, I actually like
the `nervousness' of it--not in a `makes me happy' sense but in
the sense that it effectively communicates the state of things:
the movment from entirely-closed systems to entirely-open ones
is a turbulent process, but things `calm down' again as you win.

Of course, if there were *no* progress and everything was closed,
the chart would be uniform then too--uniformly red squares.
It's that initial `comfortable' state of status quo that lulls
people into inaction, of course :)

So, don't just `fix' the *chart*--let's fix the *reality*
that it depicts, and then the next chart will be more happy ;)

-- 
"Don't be afraid to ask (λf.((λx.xx) (λr.f(rr))))."




More information about the discussion mailing list


interactive