# Letters to the editor, November 2008

This a traditional letter column. You are encouraged to write a letter of comment on anything that you find worthy of comment. It will (may) be published in this column along with my reply. As editor I reserve the right to delete material; however I will not alter the undeleted material. E-mail to me that solely references the contents of this site will be assumed to be publishable mail. All other e-mail is assumed to be private. And, of course, anything marked not for publication is not for publication. Oh yes, letters of appreciation for the scholarly resources provided by this site will be handled very discreetly. This page contains the correspondence for November 2008.

### Other Correspondence Pages

From: Beat Gfeller
Date: 11 November 2008
Subj: Algorithm for Tracking the median on a sliding window

Looking for an efficient algorithm for computing the median on a sliding window, I came across the page http://richardhartersworld.com/~cri/2004/median.html where you sketch an algorithm. Unfortunately, many details of the algorithm you describe are unclear to me from the description. Is there any research publication on which your description is based? Or is there any other more detailed description of the algorithm than the mentioned page?

There is good news and bad news. The bad news is that there is no other description; that page is all there is. The good news is that I looked at it and immediately realized that there was a much simpler algorithm that was also more efficient. I will rewrite the page. In the meantime here is an outline.

Let m (odd of course) be our window size. We maintain the window in an ordinary circular buffer W. Load the initial m elements into W.

Let k ~= sqrt(m), k odd. We create a circular buffer C that can hold 2*k elements. We find the central k elements of W, sort them, and load them into C. The median is the center element of C. Maintain an index pointing at the median.

We are now initialized.

Here is what happens on a slide:

Call an element small if it less than the smallest element in C and large if it is larger than the largest element in C.

The slide will delete one element and insert one element.

If the deleted and inserted elements are both small or both large we do nothing.

If the the deleted element is large and the inserted element is small we move the index two locations to the left if we can; likewise if the deleted element is small and the inserted element is large we move the index two locations to the right if we can. In all cases, if we can't move the index further to the left or right we reload C. Note, this is an O(m) operation.

If the inserted element is in C, i.e., if it is neither large or small, and the deleted is not, we add it to C if we can (we can't if C is overflowing) and move the index left, right, or not at all - again, if we can.

If the the deleted element is in C and the inserted element is not, we delete the element from C and move the index, again if we can.

Finally, if both the deleted and inserted elements are in C we delete and insert as appropriate, and move the index, if we can.

The average time cost of this algorithm is O(1) and the worst case is O(sqrt(m)).

I hope this helps. If you need more details please let me know.

From: sirje
Date: 8 November 2008
Subj: Whatever

Hello!

What is Doda Powell`s e mail?

That's for me not to know, and you not to find out.

From: Jamie O'Toole
Date: 2 November 2008
Subj: roger and elaine

I had a hard copy of this someone gave me years ago and I lent it to someone else who lost it. I was desolated. Then, bless the Internet and my sometimes OK memory, I checked for Roger and Elaine on Google and there it was. Thank you for posting it, for keeping it, and maybe for writing it. Still makes me laugh.

You're welcome. As it chances I didn't write it and have no idea who did; it is one of those things that circulated in email many years ago. And yes, it's a wonderful piece.

From: Peter Neilson
Date: 1 November 2008
Subj: Paris Hilton for President

You allowed in your October editorial (resurrected today by your all-too-unforgiving web host) that you favored some hotel for president. What did you say, Sheraton Boston? No, it was the Paris Hilton. You said, "Why not?" A good choice, but to answer your question, it's not a US citizen, having been born and built in France.

I agree with your sentiments entirely, with the minor proviso that I disagree with the whole of their content. The Paris Hilton is the child of American hotels, and, as such, is an American citizen by birth.
There is also a person who has the same name as the hotel. But she also is not eligible, being only 27 years old, so you could not have meant her.
That is a consideration of course, but, one must take into account that the lady in question has led a rather fast life and is rather older than her chronological years.
By the time the public at large read this letter, the Presidential Matter will have been decided, and it will be obvious (regardless of the outcome) that the Republican ticket was inverted. Indeed, you already suggested same, more or less.
The term, bizarre, best describes the politics of the twentyfirst century.

From: sirje
Date: 1 November 2008
Subj: Whatever

Hello!

What is Donn Powell`s email?

I dunno, have you asked Dona?

From: Anthony R. Lewis, PhD, FN
Date: 2 November 2008
Subj: Back again

Welcome back to the web, Mr. Harter. Anathema upon Earthlink; dig up the idolators bones, I say. Fewer than 3 days to Ralph Nader becoming president--he used his money to buy a majority of the electoral college.

My kitchen renovation continues apace.

The cats say hello, especially to Jimmy who seems sweet and affectionate.

It's good to hear that the kitchen remodeling proceeds apace. It seems a shame to replace the old one - if I recall correctly the style was, ah, unique.

I'm planning to vote for Nader myself, for no better reason than that I've not seen any TV ads for him.

The cats lie to you, but, then, that's what cats do.

From: sirje
Date: 1 November 2008
Subj: Whatever

Hello!

What is Dona Powell`s email?

I dunno, have you asked Donn?

Date: 17 October 2008
Subj: Frisky Femme Fatales

Dear Ms. Rinehart:

Recently, I spoke with Richard Harter, Highmore's Scribe of Scandal, to receive a urinary update on the frisky femme fatales of Grand Lodge Infanmy. I must say, madam, that I was chagrined to discern a distinct lack of compassion on Mr. Harter's part for the couch's creative crotch crouchers. He certainly failed, in my opinion, to appreciate the devastating social and financial repercussions meted out by the court.

Obviously, these luckless lasses will need a means of support to deal with their restitution tributlations, and I believe I have come upon the perfect solution. I enclose an article which will reveal all. I believe that these pure-as-driven-white-ladies are now in the perfect position, if you'll pardon the reference, to offer their services to the WBC. Pugilists everwhere await the purity of their product. Please pass along this suggestion to the ladies - but don't tell the Miller journalist that I sent it.

United in sisterhood,

### Klitschko used baby pee to fight hand swelling

Vitali Klitschko used his son's wet, used diapers to keeps his fists from swelling up after winning his WBC heavyweight title bout against Nigeria's Samuel Peter, the Ukrainian told a German newspaper on Tuesday.

Klitschko said he wrapped them around his hands and they helped him recover.

"Baby pee is good because it's pure, doesn't contain toxins and doesn't smell," the 37 year old boxer told Bild after he won back the WBC title with a technical knockout on Saturday.

"I wrap nappies filled with my 3 year old son Max's wee around my fists," Klitschko said, adding that he got the idea from his grandmother. "The nappies hold the liquid and swelling stays down."

Klitschko said that Peter should try the nappy trick as well.

Peter chose not to come back out after eight rounds that saw him weaving and occasionally wobbling as Klitschko landed a persistent stream of heavy left jabs to Peter's head in front of more than 12,000 people at the new OZ World Arena.

The 37-year old Ukrainian firghter (36-2, 35 KOs) used his height -- he's 6-foot-7 -- and reach to keep Peter crouched and defensive.

The above missive was addressed to Our Lady of the Large Black Dog. As it chances, it came into my possession by means that are best left to the imagination.

I am reliably informed that N. Adams once was a resident of Hyde county, and that she left the environs of Highmore under circumstances that I do not feel should be aired in public. It is good to see that she has taken an interest in the affairs of her one time time residence. Her suggestions have all the value that we have come to expect from her.

From: Chris Ogle
Date: 18 October 2008
Subj: Character names in Tom Godwin's "The Cold Equations"

On your webpage http://richardhartersworld.com/cri_d/cri/1999/coldeq.html you state that the girl's name is "Amanda". In my copy of the story ("18 Greatest Science Fictions Stories" aka "Masters' Choice" edited by Laurence M. Janifer), the name is mentioned four times as "Marilyn Lee Cross". I don't know if this is a typo or if the name changed in different printings.

Bad Richard; it's a typo. Marilyn Lee Cross it is. I haven't the foggiest idea how Amanda got it in there, but I will fix it forthwith. Thanks for catching it and letting me know.

From: Gary Pringle
Date: 21 October 2008
Subj: personal God

Now I believe in God , you do not .One of us is wrong.Supose I am wrong when I die I will end up the same as you. However if I am right ,and ther is a God and a hell where non believers go I win either way ,but you on the other hand if I am right will spend eternity in hell .You do the math ,I believe you are an inteligent person

This is one of the classic arguments of believers, sometimes called the fire insurance fallacy. The problem with it is that it is a false dichotomy, which is a fancy way of saying that you are only considering two possibilities and ignoring all of the rest. For example:

Maybe the Romans were right about their Gods, and all of you Christians, being non-believers in the true gods, are going to hell. Maybe the Hindus are right and you will be reincarnated. As a proselytizing Christian your karma is bad and you will be an insect in your next life. Perhaps, as the gnostics teach, the creator of the material world is an evil demiurge who created the world to provide an arena for pain and suffering. If you come to His attention, you might suffer for eternity; best to avoid coming to His attention.

From: Peter Neilson
Date: 19 October 2008
Subj: Unavailable

Mr. Well-loved Harter!

You are too popular and thus unavailable until the first. Sheesh. Maybe http://www.hostadfree.org/ can help?

It's my fault, really. There are certain pages on my site that are, so to speak, toxic. By toxic, I don't mean that they are a source of malware. Rather they are pages that every so often are discovered by major websites and become the amusing page of the day/week/month. I exile these pages off to Annex D, the home of overly popular pages (these are web site pages, not pages in King Arthur's court who entertain the ladies of the court whilst the knights are out jousting with dragons). One of the pages escaped my notice and received about 100,000 hits or so.

It's not entirely my fault. Usually I would catch this sort of thing my monitoring my website with a tool that earthlink provides. However I recently upgraded to firefox 3 which has an undesirable feature. Earthlink never bothered to update the certificate on their tool page. Firefox 2.* asked if I wanted to look at it anyway. Firefox 3 just kicks me out. No doubt there is way to tell FF3 that I really want to look at the page, but I haven't figured it out. One of the rules of software is that the documentation for getting around bad design is always obscure.

What I end up doing is checking it on another machine that has FF2 on it. I've been remiss about doing that and I was caught by surprise.

Meanwhile, my commercial website, where I advertise the pony rides that I do, www.ponyspot.com, has been de-listed by Google for being a source of malware. Only it ain't. It's just the same as it always was, a bit of harmless html that displays text and a few photos. I'm to go through some dance with Google before I can gain back an unblemished reputation, if at all. Google for pony ride nc, and see what happens.

The only thing that I can think is a "problem" is that I have my e-mail address there, disguised in a chunk of javascript. Other than that, it's a straightforward piece of exceptionally non-interactive html. What's google for "sheesh!"??

Actually it might be compromised. I looked at their report which claims that some Russian site (ru is Russia, isn't it?) has snuck some malware onto the site. Who is hosting the site? They are supposed to be protecting you against intrusion - not that they necessarily do a good job of it. Have you downloaded the actual HTML and compared it with what its supposed to be there?

From: Arv Edgeworth
Date: 23 October 2008
Subj: Most mutations are not beneficial

I know you disagreed strongly with most mutations being harmful, although the current biology books indicate that (you never did respond to that either), in regards to mutations and their effects you said: "What is the consequence when one of these things happens? Most of the time the change either has no perceptible effect at all, or it is fatal." I noticed you indicated most changes are neutral or harmful, where do the beneficial changes come from?

You don't give up, do you? Well, I will play your silly game yet again. Shame on me.

I am not at all willing to take your word for it that the current biology books indicate that most mutations are harmful. Even if they did, and I doubt that they do, all that that would indicate is that the ones in question are out of date. Again, I'm not taking your word for it. If you can produce an actual study that looked at current biology texts (broken out by academic level and category, of course) that indicated any such thing, that would be a different matter. In the meantime you are just blowing smoke.

Don't most mutagens cause harmful changes, not beneficial ones? If DNA proofreads itself and corrects changes, wouldn't it be as likely to correct beneficial changes as harmful ones?
That's a good question. On first thought it would seem that it wouldn't make any difference since the proofreading doesn't "know" about beneficial and harmful. That is, one would expect the ratio of beneficial mutations to harmful mutations to be unaffected by proofreading. However this isn't quite right - major errors are easier to catch and are more likely to be harmful than minor errors. Ergo, DNA proofreading differentially eliminates a greater percentage of harmful mutations than beneficial. I don't know whether the difference is significant.
I still don't believe there is any proof of changes, except within a species. Consider these quotes:
I know this may come as a shock, but I don't really care what you believe.
In Paleobiology, vol. 6(1), in January 1980, p. 127, Stephen Jay Gould wrote: “The absence of fossil evidence for intermediary stages between major transitions in organic design, indeed our inability, even in our imagination, to construct functional intermediates in many cases, has been a persistent and nagging problem for gradualistic accounts of evolution.”
Shame on you. Gould has complained at length about persons of your ilk taking his words out of context and misusing them.
W.E. Swinton (British Museum of Natural History writing in Biology and Comparative Physiology of Birds, vol. 1, p.1, 1960 said, “There is no fossil evidence of the stages through which the remarkable change from reptile to bird was achieved.”
Note the date. In 1960 that statement was almost true. I say almost, because even then there was some evidence. Matters are substantially different today since the discovery of feathered dinosaurs. Incidentally the term "reptile to bird" is unfortunate.
David B. Kitts, Ph.D. (zoology) University of Oklahoma writing in Evolution, vol. 28. Sept. 1974 said, “Evolution requires intermediate forms between species and paleontology does not provide them.”
I don't know who Kitts was nor how much violence you have done to his argument by ripping his words out of context, but this is suspect. If he merely meant that it is rare to find finely graded sequences of intermediates, that is indeed rare, though not unknown. The reason is simple and obvious - the fossil record simply is not good enough to provide that level of detail.

-------

After I told him that I was done with him Edgeworth sent yet another email, the above. Quite unwisely I replied as above. Edgeworth replied with a long rambling letter filled with disinformation, misinformation, and self congratulation. I dare say I'm getting old, but sometimes life is just too short for this sort of thing.