I think that everything important has been addressed; I don't have a dilemma. Some readers have understood the point that I was trying to make, and some haven't. I can't ask for more than that. I certainly never expected that _everyone_ would understand it.

When I started this project, I thought that I might be able to find a mathematical clue that would help to answer a question that was asked, but never answered, in 1998: Did this particular mistake result from a lack of ethics, or from a lack of knowledge?

I didn't find such a clue. The algorithm that I found, and described in my first two messages, could have been chosen because of a lack of knowledge, or it could have been chosen because of a lack of ethics. The mathematical evidence answers the "how" question, but it doesn't answer the "why" question.

I started writing a Bad Astronomy Story, explaining _how_ the image had been very badly processed. I tried to follow, conscientiously, the principle of "innocent until proven guilty," attributing every mistake to a lack of knowledge, rather than to a lack of ethics, whenever possible. I was able to convince myself that all of the mistakes of April 6 could have been, conceivably, innocent mistakes.

But then I got to April 7. The hypothesis of "presumed innocence" began to fail. Something happened, during the rest of that month of April 1998, for which I just couldn't find a 100% innocent explanation. The retraction, which surely would have been issued if everyone had been acting according to Feynman's principles of "utter scientific integrity," was never issued.

So I rewrote my Bad Astronomy Story, with the "how" part of the story still focused on something that happened during April 6 (the inappropriate processing of an image), but with the "why" part of the story now refocused on something that didn't happen during April 7-30 (the issuing of a retraction).

Your suspicion is correct: it was meant as a request for his professional opinion. It was meant to serve as an "open letter," i.e. a message that is addressed to one person, but is meant to be seen by, and followed-up by, anyone who has an interest in the relevant subject area. These open letters are very common in the "blogosphere."

I don't believe that the original data, as returned by the spacecraft, were ever suppressed.

But I do believe that the filter that they used was so grossly inappropriate, and their explanation of the effects of that filter was so grossly inadequate, that a retraction was required, as a matter of normal scientific integrity.

So this accusation isn't about missing information (from the spacecraft) but, instead, it's about a missing retraction.

The words "save face" may belong in the blank. No one likes to admit a mistake. Scientists, however, _should_ admit mistakes, for obvious reasons.

Perhaps it will help if I provide a new explanation of the "slip" itself.

They described the algorithm in question as a "long, narrow high-pass filter" and they stated that its purpose was "to help reduce some of the [...] streaking that is noticeable in the original data."

The "long, narrow high-pass filter" is a tool which can be used, in effect, to destroy whichever features are the "longest" in the direction of the spacecraft's movement across the landscape. This isn't unreasonable, per se, because the streaks really are the longest features: each streak is 4,800 pixels long (*), while the "face" landform is nowhere more than 350 pixels long. They had _plenty_ of room for successful use of this technique. If they'd targeted this technique more appropriately, e.g. if they'd targeted it for the selective suppression of features longer than approximately 4,000 pixels, then they would've suppressed only the streaks, and we wouldn't be having this discussion now. Instead, seemingly inexplicably, they targeted it for the selective suppression of objects longer than approximately *31* pixels.

Thirty-one pixels!!

What a slip!

My original animation can be described in a way which is relevant here: the animation demonstrates the effect of a "targeting" parameter. In the first frame, only the longest features, the streaks, are targeted. In subsequent frames, the broadness of the targeting is gradually increased, targeting more features in each frame. In the last frame, the targeting parameter is set to 31, the same value that was used to create the press release image. As you can see in the animation, this value makes the targeting _much_ too broad.

(*) (originally 9,600, before the image was sized down, by a factor of two)

No, I don't believe that the original data, as returned by the spacecraft, were ever suppressed.

Yes.

This is mostly answered by my new explanation of the "slip" itself, above. Their description acknowledges the selective suppression of "long" features, but it says that this suppression was targeted at the streaks. In reality, as I've explained above, the targeting was overly broad-- enormously overly broad-- and, as a result, the suppression included major features of the "face" landform itself.

Has this message expanded enough, and clarified enough, to answer your questions? If not, let me know.

I don't know of any discrepancy that hasn't already been addressed.

I think that I've answered all of the short questions in this list, while answering the questions in the longer overlapping list, above. Let me know if you think that anything still needs to be answered.

P.S. Thank you for your explanation of the "atmosphere" that prevails here. Dr. Plait, if you read this some day, you might want to grab some of motm's excellent introductory material, above, for a FAQ entry.

Yes, yes, yes! This is what I was looking for, hopefully true for others!

Check. I'll stop worrying then . Oh, and very true about not everyone. Well, hopefully everyone here, but in real life...forget it :P

Check. Common...hmm. I could see that, I'll apologize for sounding rough there. Common, perhaps, just none of us are used to seeing it here. Threw me off guard at least. Thanks for clarifying.

Check. Quite true! (Emphasis mine).

Check. Thank you, that was much clearer.

Check

Yes, far beyond what I would expect if you were really one of the conspiracists that comes through from time to time, they invariably dodge, dance, and eventually leave for various reasons. Check out some of the other threads in the conspiracy section to see what I mean. Thank you for letting me pin you down for a second. It has helped leaps and bounds. In my mind at least, and hopefully for others too.

Good. I had thought there was; I took your later posts trying to "further clarify" as "still looking for an answer". Thanks!

I'm happy. :applaud: [i](That's supposed to be one of those clappy guys, but I don't remember the code apparently!)

Mmm, glad I could help? That's not quite what I was after, but if it helps it helps...*runs off.

Thanks a thousand times for doing this Bad Guardian. And welcome to the board again. Hopefully there isn't this much confusion in the future, good to get that out of the way at the beginning. And if there is confusion...well, it happens to the best of us. If you have the interest and the eye for this sort of thing (analysis, understanding of photo-rendering, astronomy...) you could be very helpful when the really wild ones come by. If you're sticking around that is. If not, well, thanks for taking the time to clarify (and doing so before you leave makes it all the better in my book--a very rare trait indeed!).

Edited to add: also thanks for doing it all in one post. I can see bits and pieces in your other posts, but they've been scattered. This helps a ton.

__________________
None to speak of