Re: Instruments and models Vs. Accuracy
Re: Instruments and models Vs. Accuracy
- Subject: Re: Instruments and models Vs. Accuracy
- From: Marco Ugolini <email@hidden>
- Date: Thu, 22 Sep 2005 12:27:19 -0700
Hello, Mitchell.
I may have misunderstood what Mark told me, though I know what I heard (in
its essence, if not word by word), and that is what I reported. Of course
it's my responsibility alone if what I reported turns out to be incorrect.
My question was (and I paraphrase): "Is CIECAM being currently used in
commercial color management applications, or is it going to be soon?" And
what I reported is what I heard -- paraphrasing: "No, it's not available
commercially quite yet. It will need further study, refinement and testing
before it is used successfully in commercial applications." These quotes are
not verbatim, but just the essence of what I recall having heard from Mark
after I raised my hand in class to ask him.
I categorically and emphatically deny that I ever meant my quote to imply in
any way that Mark said that the CIECAM model is "precluded from being used
in practical applications." That would be a gross misunderstanding of what I
wrote.
My insinuation that the MCSL may have been "working for Microsoft all along
and were prohibited from revealing that by a binding NDA" was uncalled-for
and infelicitous, but it was facetious, tongue-in-cheek. My stupid sense of
humor was heavy-handed in this case. I should be more careful not to say
stuff like that so freely.
But I clearly didn't claim to KNOW that this was a fact. Just being a
smartass and shooting my big mouth (or keyboard!), I guess, though there was
no ill intention in it. Still, I'm sorry for writing that, and I apologize
for hurting anyone's feelings.
But Mitch, why do you have to say that I "WANTED to hurt" Mark? That is
truly unfair, at least as much or perhaps more than my stupid remark,
because mine was not INTENDED to hurt, and yours clearly is intended to hurt
ME. I only have great respect and admiration for Mark, and you truly have no
right to say or imply otherwise.
I did my best to quote what I heard, but I apologize if in any way I failed
to understand Mark's answer in its intended proper context. I certainly do
not mean to embarrass anybody, just as much as I will not let anyone
embarrass me unfairly, and I do confirm that any errors on my part are in
perfectly good faith.
FYI, Mark Fairchild is also receiving a copy of this message directly from
me. No need to forward on your part.
Best regards.
--------------
Marco Ugolini
Mill Valley, CA
> From: Mitchell Rosen <email@hidden>
> Date: Thu, 22 Sep 2005 09:55:13 -0400
> To: ColorSync Users Mailing List <email@hidden>
> Subject: Re: Instruments and models Vs. Accuracy
>
>
> I came across this on the list from about a week ago:
>
> In a message dated Wed, 14 Sep 2005 01:53:44, Marco Ugolini wrote:
>
>> I was at the Munsell Color Science Lab in Rochester last June, and, when
>> asked, the very people who are perfecting the CIECAM model (Mark Fairchild,
>> in particular) specifically said that CIECAM is still in the research phase
>> and NOT ready for the consumer market.
>
> This seemed odd, so I asked Mark himself. Below is what he replied.
>
> Mark Fairchild wrote:
>
>> I don't recall the conversation or the context from the summer
>> course. I certainly do talk about color appearance modeling being
>> an area of active research, but I stress that more for the iCAM
>> stuff that is even more complex. I know for sure that I have talked
>> about CIECAM02 as being very mature compared to CIECAM97s and that I
>> am unaware of any immediate issues that require the CIE to be
>> working on a revision. There is no committee doing such work. A
>> few difficulties have been identified (like mathematical misbehavior
>> when unreal colors are sent through), but I don't think those
>> preclude the model being used in practical applications. I might
>> well have said that it should be used with care to make sure that
>> the extra parameters aren't used to do more harm than good. I have
>> no doubt that there are plenty of people out there that can use it
>> with care and I would suspect that Microsoft is making the
>> simplifying assumptions required to make it work in practical
>> applications. You could look at Rod's recent work (to be presented
>> at CIC on perceptual color gamuts and how CIECAM02 can predict them)
>> as an example of how well it can do.
>
>> Bottom line ... I have no concerns with anyone using CIECAM02 in
>> color management applications as long as they use it with knowledge
>> and care. (BTW: Knowledge and care are also required to use CIELAB
>> in similar applications!!).
>
> -------------
>
> As a student in our lab last summer, Marco Ugolini is certainly
> welcome to quote us in as accurate a way he can and as he did above,
> we expect it will be done in good faith. We are happy to clarify
> when appropriate.
>
> But, we were saddened to see the rest of Marco Ugolini's email of
> Sept. 14 where he went on to accuse Mark of purposely misleading on
> CIECAM02 perhaps due to a hypothetical financial relationship with
> Microsoft????? The absurdity and hurtfulness of this conjecture is
> many-fold. For one thing, in the Munsell Lab's 22 year history, we
> have never had a financial relationship with Microsoft - not so much
> as having them buy lunch for a faculty member. Second, Mark
> Fairchild is the most outspoken individual I have ever met on the
> topic of absolute separation of the interests of corporate sponsors
> from the interests of intellectual integrity. If you really wanted
> to hurt him, you chose the right words. And thirdly, if we were to
> have Microsoft become a project sponsor in the future (it could
> happen any time - they are interested in our research and in our
> students) we would never accept terms from them or any sponsor that
> would prevent our research from being publishable or to encroach on
> our free intellectual expression. If we ever felt uncomfortable
> talking about a subject, we would tell you.
>
> There is never a need to guess as to who our project sponsors are.
> They are publicly acknowleged in our publications and in our annual
> report (available at web site below).
>
> - Mitchell Rosen
> Munsell Color Science Laboratory
> Rochester Institute of Technology
> http://www.mcsl.rit.edu
> _______________________________________________
> Do not post admin requests to the list. They will be ignored.
> Colorsync-users mailing list (email@hidden)
> Help/Unsubscribe/Update your Subscription:
>
m
>
> This email sent to email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden