it's getting better all the time

/tuning/files/perlich/xoomer.gif

---------------------------------
Do you Yahoo!?
Y! Web Hosting - Let the expert host your web site

> /tuning/files/perlich/xoomer.gif

Paul, there's an anomaly in this gif file. Photoshop balks,
and ACDSee shows a Cubist version.

-Carl

--- In tuning@y..., "Carl Lumma" <clumma@y...> wrote:
> Paul, there's an anomaly in this gif file. Photoshop balks,
> and ACDSee shows a Cubist version.

Yep, it's munged. Also, are we *really* supposed to be able to easily see light yellow lines on a white background? (if this .gif was meant for printing, then it would be a different matter, to some extent...)

Cheers,
Jon

--- In tuning@y..., "Jon Szanto" <JSZANTO@A...> wrote:
> --- In tuning@y..., "Carl Lumma" <clumma@y...> wrote:
> > Paul, there's an anomaly in this gif file. Photoshop balks,
> > and ACDSee shows a Cubist version.
>
> Yep, it's munged. Also, are we *really* supposed to be able to
>easily see light yellow lines on a white background? (if this .gif
>was meant for printing, then it would be a different matter, to some
>extent...)

if you saw the yellow lines, then what, exactly, was "munged"? i
suppose you could answer off-list, i could try making another one off
the .bmp or just sending you the .bmp itself, and we could go from
there . . .

p.s. i thought yellow was a good color because i didn't want it to be
obtrusive, but you can still read off the errors of the consonant
intervals of any ET if you so desire . . . also, yes, this page might
very well end up in a XH article, in which case printing to black and
white would be a must . . .

Paul,

--- In tuning@y..., "wallyesterpaulrus" <wallyesterpaulrus@y...> wrote:
> if you saw the yellow lines, then what, exactly, was "munged"?

Carl's description was perfect: "cubist". It is as if someone took square sections of the piece and cut them out, pasting them on top of each other and destoying any of the material lining up. IIRC, the bottom info has been truncated as well. You can see numbers, lines, etc. but they are jumbled.

> i could try making another one off the .bmp or just sending you
> the .bmp itself, and we could go from there . . .

You do need to make another, or try to redo a .gif from the .bmp. I pointed George Secor to the following small and free bmp2gif tool:

http://download.com.com/3000-2192-870646.html?tag=lst-4-24

> p.s. i thought yellow was a good color because i didn't want it
> to be obtrusive, but you can still read off the errors of the
> consonant intervals of any ET if you so desire . . .

It is *not* a good choice because of the amount of light coming out of both the white and yellow pixels. Many people (according to web [as well as computer/screen] usability studies) find this very hard on the eyes. Many other pastels would work, but it *does* need to stand out just a tad more.

> also, yes, this page might very well end up in a XH article, in
> which case printing to black and white would be a must . . .

Not with a grant or independent funding! :) Well, if it was color printing then the yellow would work, otherwise you'll have to grey it or use some kind of dotted/dashed line.

Cheers,
Jon

--- In tuning@y..., "Jon Szanto" <JSZANTO@A...> wrote:
> Paul,
>
> --- In tuning@y..., "wallyesterpaulrus" <wallyesterpaulrus@y...>
wrote:
> > if you saw the yellow lines, then what, exactly, was "munged"?
>
> Carl's description was perfect: "cubist". It is as if someone took
>square sections of the piece and cut them out, pasting them on top
>of each other and destoying any of the material lining up.

this is very strange. when i look at the file,

/tuning/files/perlich/xoomer.gif

using internet explorer, either today or on the day i posted it, it
looks just fine! you guys must be using something else, then? what
could be happening? (offlist)

> > p.s. i thought yellow was a good color because i didn't want it
> > to be obtrusive, but you can still read off the errors of the
> > consonant intervals of any ET if you so desire . . .
>
> It is *not* a good choice because of the amount of light coming out
>of both the white and yellow pixels. Many people (according to web >
[as well as computer/screen] usability studies) find this very hard
>on the eyes.

the number of yellow pixels is less than 1% of the number of white
pixels. so i doubt that changing from yellow to some other color
would significantly change the amount of light coming out.

>Many other pastels would work, but it *does* need to stand out just
>a tad more.

wait until you view the non-munged version, and then tell me the
yellow lines, though subtle, don't convey the necessary information.

> > also, yes, this page might very well end up in a XH article, in
> > which case printing to black and white would be a must . . .
>
> Not with a grant or independent funding! :) Well, if it was color
>printing then the yellow would work, otherwise you'll have to grey
>it or use some kind of dotted/dashed line.

already dotted/dashed except for the 0-cent-deviation-from-just-
consonance lines.

Paul,

> using internet explorer, either today or on the day i posted it, it
> looks just fine! you guys must be using something else, then? what
> could be happening? (offlist)

Looks like Yahoo was originally mucking up the download (not unknown). I re-downloaded and it is OK.

> the number of yellow pixels is less than 1% of the number of white
> pixels. so i doubt that changing from yellow to some other color
> would significantly change the amount of light coming out.

The point is not how many are yellow, it's that the ones that are are so close in the light spectrum to the white. It is tiring on the eyes and more difficult to see than a slightly darker hue. I couldn't even see that they were dotted/dashed til I zoomed in quite a bit.

One thing to keep in mind: while the light in a room may vary, a printed page will look (essentially) identical when distributed to many people, but the many variations on how a person will have set the controls on their monitor (brightness, contrast, and especially the gamma correction) makes color decisions for these images difficult. It could be that your monitor is set darker than most.

Anyhow, it is up to you; I could certainly make samples with other colors, as it is easy to alter the .gif palettes and resave. But if you ignore the studies of people who study people looking at monitors, you run the risk of having less impressive results than you seek.

Cheers,
Jon

> this is very strange. when i look at the file,
>
> /tuning/files/perlich/xoomer.gif
>
> using internet explorer, either today or on the day i posted it, it
> looks just fine! you guys must be using something else, then? what
> could be happening? (offlist)

In my post I explained that result. ie doesn't enforce standards
very well. Did you do anything differently between the first
zums and this one?

> the number of yellow pixels is less than 1% of the number of white
> pixels. so i doubt that changing from yellow to some other color
> would significantly change the amount of light coming out.

I don't know what Szanto is talking about. It looks fine on-screen.
Print may be harder, but not as hard as Jon says. Tho IMO b&w
would look best on XH. Though it's up to you/John, of course.

-Carl

Carl,

--- In tuning@y..., "Carl Lumma" <clumma@y...> wrote:
> ie doesn't enforce standards very well. Did you do anything
> differently between the first zums and this one?

As I mentioned to Paul, I think the problem may have been Yahoo munching bits during download. I re-downloaded it and it seems fine.

> I don't know what Szanto is talking about. It looks fine on-screen.

And very screen-settings dependent. I'll stand by my years of hanging out on web design lists, where how information is portrayed in browsers (and therefore on-screen) is a paramount issue. Which is why is may look fine to you, and still have problems in terms of wider distribution.

> Print may be harder, but not as hard as Jon says.

Oh, not hard. Just harder than sending a color image and expecting it to automatically come out fine when reduced to black and white.

> Tho IMO b&w would look best on XH.

Just out of curiousity, why? I was being cute when I talked about funding for the color graphics, but in what situation would this much information, especially where the eye needs to track things, be better in b/w than color?

Cheers,
Jon

Jon wrote...
>As I mentioned to Paul, I think the problem may have been Yahoo
>munching bits during download. I re-downloaded it and it seems
>fine.

In that case it was probably still yahoo (I dl'd it twice and both
were broken).

>And very screen-settings dependent. I'll stand by my years of
>hanging out on web design lists, where how information is
>portrayed in browsers (and therefore on-screen) is a paramount
>issue. Which is why is may look fine to you, and still have
>problems in terms of wider distribution.

Yeah, if he wants to place it on a web page he should stick to
the old Netscape palette, as I think I saw you mentioning to him.
But I assumed he didn't care about that.

>> Tho IMO b&w would look best on XH.
>
>Just out of curiousity, why? I was being cute when I talked about
>funding for the color graphics, but in what situation would this
>much information, especially where the eye needs to track things,
>be better in b/w than color?

It would require John to change the process/look he's had for a
while, which I was protesting on behalf of nostalgia. But change
can be good, too.

-Carl

--- In tuning@y..., "Carl Lumma" <clumma@y...> wrote:
> > this is very strange. when i look at the file,
> >
> > /tuning/files/perlich/xoomer.gif
> >
> > using internet explorer, either today or on the day i posted it,
it
> > looks just fine! you guys must be using something else, then?
what
> > could be happening? (offlist)
>
> In my post I explained that result. ie doesn't enforce standards
> very well.

huh ? ? ?

> Did you do anything differently between the first
> zums and this one?

no.

> > the number of yellow pixels is less than 1% of the number of
white
> > pixels. so i doubt that changing from yellow to some other color
> > would significantly change the amount of light coming out.
>
> I don't know what Szanto is talking about. It looks fine on-screen.
> Print may be harder, but not as hard as Jon says. Tho IMO b&w
> would look best on XH.

XH is a b&w only publication (photocopy).

> Paul!

Not that it matters, but both Chris Forester and myself have had instruments features on the cover in Color!

>
> From: "wallyesterpaulrus" <wallyesterpaulrus@yahoo.com>
> Subject: Re: it's getting better all the time
>
>
> XH is a b&w only publication (photocopy).
>
>

-- Kraig Grady
North American Embassy of Anaphoria island
http://www.anaphoria.com

The Wandering Medicine Show
KXLU 88.9 fm Wed. 8-9pm PST.
live stream kxlu.com

>>In my post I explained that result. ie doesn't enforce standards
>>very well.
>
> huh ? ? ?

ie = internet explorer. It makes a guess rather than giving an
error, like Photoshop.

-C.

--- In tuning@y..., Kraig Grady <kraiggrady@a...> wrote:
> > Paul!
>
> Not that it matters, but both Chris Forester and myself have
>had instruments features on the cover in Color!

of course that matters!! i'll have to see what john says about
this . . .

--- In tuning@y..., "Carl Lumma" <clumma@y...> wrote:

> >>In my post I explained that result. ie doesn't enforce standards
> >>very well.
> >
> > huh ? ? ?
>
> ie = internet explorer. It makes a guess rather than giving an
> error, like Photoshop.

so you're saying the .gif is flawed, but since ie makes a lucky
guess, i'm unable to detect the flaws when i use ie to view it? i'm
confused . . .

--- In tuning@y..., "wallyesterpaulrus" <wallyesterpaulrus@y...> wrote:
> so you're saying the .gif is flawed, but since ie makes a lucky
> guess, i'm unable to detect the flaws when i use ie to view it? i'm
> confused . . .

Sometimes you might just want to take us at our word! IE is a browser, not an image editing or viewing program. It is entirely possible that the image viewer in IE ignores some elements of a file that graphics programs show up as a flaw; all current browsers take liberties with the many elements of web presentation, causing much hair to be pulled by designers.

If at any point these graphics files will be needed for print, you simply *owe* it to yourself to make sure the file is OK. I would suggest having on hand either a low-cost or even no-cost image editing program to make sure you are preparing clean files, and that the results will be good for whoever you send it to.

Since this is completely OT, write me off-list if you want suggestions on these kind of programs; I've recently helped out George Secor in a similar area...

Cheers,
Jon

--- In tuning@y..., "Jon Szanto" <JSZANTO@A...> wrote:
> --- In tuning@y..., "wallyesterpaulrus" <wallyesterpaulrus@y...>
wrote:
> > so you're saying the .gif is flawed, but since ie makes a lucky
> > guess, i'm unable to detect the flaws when i use ie to view it?
i'm
> > confused . . .
>
> Sometimes you might just want to take us at our word!

i wasn't doubting anyone, just confused . . .

>IE is a browser, not an image editing or viewing program. It is
>entirely possible that the image viewer in IE ignores some elements
>of a file that graphics programs show up as a flaw; all current
>browsers take liberties with the many elements of web presentation,
>causing much hair to be pulled by designers.
>
> If at any point these graphics files will be needed for print, you
>simply *owe* it to yourself to make sure the file is OK. I would
>suggest having on hand either a low-cost or even no-cost image
>editing program to make sure you are preparing clean files, and that
>the results will be good for whoever you send it to.

so you're saying that microsoft paint and microsoft photo editor
don't qualify? i used both programs to view the .gif files *before* i
uploaded them (the first to convert from .bmp, and the second to
cross-check paint's result). i would have thought this would
definitely qualify as the kind of procedure you're now outlining,
but . . .

> Since this is completely OT, write me off-list if you want
>suggestions on these kind of programs; I've recently helped out
>George Secor in a similar area...

yes, please reply off-list.

>>ie = internet explorer. It makes a guess rather than giving an
>>error, like Photoshop.
>
>so you're saying the .gif is flawed, but since ie makes a lucky
>guess, i'm unable to detect the flaws when i use ie to view it?
>i'm confused . . .

That's right. I checked it in ie here as well.

As for Paint... Microsoft has made it something of a design
principle to implement the least amount of error-checking
they can get away with. However, if Jon's intermittant yahoo-
munging theory is correct, you may never see the problem...

-C.

--- In tuning@y..., "Jon Szanto" <JSZANTO@A...> wrote:
> --- In tuning@y..., "wallyesterpaulrus" <wallyesterpaulrus@y...>
wrote:
> > so you're saying the .gif is flawed, but since ie makes a lucky
> > guess, i'm unable to detect the flaws when i use ie to view it?
i'm
> > confused . . .
>
> Sometimes you might just want to take us at our word!

well, just to rule out the possibility of a consistent cross-
program .gif-standard implementation failure among all the microsoft
applications i was using, i just downloaded paint shop pro (thanks to
your e-mail). the file looks just fine in that program, too!!

so i'm unable to accept carl's theory based on the current evidence
(carl can of course join in our offlist discussion if he so pleases),
and so i must tentatively conclude that there's nothing wrong with

/tuning/files/perlich/xoomer.gif

(which i dedicate to herman miller).

--- In tuning@y..., "wallyesterpaulrus" <wallyesterpaulrus@y...> wrote:
> --- In tuning@y..., "Jon Szanto" <JSZANTO@A...> wrote:
> > Sometimes you might just want to take us at our word!

...that the file caused problems on standard viewers on our machines. I think we've pretty much determined that it was bad downloading from Yahoo.

> i just downloaded paint shop pro (thanks to
> your e-mail). the file looks just fine in that program, too!!

That's what I used to do the other versions of it, as well.

OK, off-list for everything else on this one!!!!!

Cheers,
Jon

> well, just to rule out the possibility of a consistent cross-
> program .gif-standard implementation failure among all the
> microsoft applications i was using, i just downloaded paint shop
> pro (thanks to your e-mail). the file looks just fine in that
> program, too!!
>
> so i'm unable to accept carl's theory based on the current
> evidence (carl can of course join in our offlist discussion
> if he so pleases), and so i must tentatively conclude that
> there's nothing wrong with

Sorry I haven't been able to convince you. Paint and ie display
this fine, but not Photoshop (reports the error and displays
nothing) or ACDSee (cubist version, plus error).

You should be able to see, if you view the image in ie from the
location below, that even ie displays a black line down the
righthand side of the image, which was not present in your first
tree gifs, which were ok.

> /tuning/files/perlich/xoomer.gif
>
> (which i dedicate to herman miller).

Munged. Not something obvious, like a missing terminator or
bad header, as best I can tell based on:

http://www.dcs.ed.ac.uk/home/mxr/gfx/2d/GIF87a.txt

Why don't you try ziping it and posting it, or e-mailing it
to one of us?

-Carl

--- In tuning@y..., "Carl Lumma" <clumma@y...> wrote:
> > well, just to rule out the possibility of a consistent cross-
> > program .gif-standard implementation failure among all the
> > microsoft applications i was using, i just downloaded paint shop
> > pro (thanks to your e-mail). the file looks just fine in that
> > program, too!!
> >
> > so i'm unable to accept carl's theory based on the current
> > evidence (carl can of course join in our offlist discussion
> > if he so pleases), and so i must tentatively conclude that
> > there's nothing wrong with
>
> Sorry I haven't been able to convince you. Paint and ie display
> this fine, but not Photoshop (reports the error and displays
> nothing) or ACDSee (cubist version, plus error).

so why did paint shop pro do it fine?

> You should be able to see, if you view the image in ie from the
> location below, that even ie displays a black line down the
> righthand side of the image, which was not present in your first
> tree gifs, which were ok.
>
> > /tuning/files/perlich/xoomer.gif
> >
> > (which i dedicate to herman miller).
>
> Munged. Not something obvious, like a missing terminator or
> bad header, as best I can tell based on:
>
> http://www.dcs.ed.ac.uk/home/mxr/gfx/2d/GIF87a.txt
>
> Why don't you try ziping it and posting it, or e-mailing it
> to one of us?
>
> -Carl

i'll upload the .bmp to the same directory. tell me how that looks,
and make your own .gif i fyou like . . .