Sharp as a Marble - The Memos Do Not Show Kerning

<< Losing the Buzz | Home | Happy Birthday Georgia! >>

Politics

This post has been rescinded - My analysis was proven to be incorrect by Dale at Q and O, therefore I concede error. The memo does indeed kern looking at the letters f and j.

Besides, here's someone who knows a hell of a lot more about doing proper research than I.

Final thoughts? Hoax.

posted @ 9/10/2004 9:38:00 PM

Comments

Passerby says:

on 9/10/2004 11:03:00 PM

I think I have to agree with you that "kerning" is not in play here. As I understand it, computer-font kerning strictly refers to a limited set of letter-pairs which get special treatment at printing time. You can download programs to view the kerning info of your fonts: my standard MS Times New Roman has only 116 kerning pairs defined, for combos like YA, Yu, and rg. If kerning were turned on, the letters CYA would probably be closer together.

But that's not the end of it. The individuals letters of the font have horizontal spacing information which is *not* kerning. Part of this info tells our word processors "after printing this letter, move x/1000s of an inch to the right before you start printing the next letter". Now in Times New Roman, the lowercase "f" is a curious beast--it tells the word processor to move slightly to the *left* before printing the next letter. That makes the second letter--which in most cases will be a small lowercase figure--nestle comfortably under the loop of the f. To repeat, this happens whether kerning is on or off.

(To see an example where this leftward shift is bad typographically, type fW in your word processor--the W overlaps the loop of the f. But most of the time, it works nicely).

In the suspect memos, there are several combos like fe and fr. If they were produced on a modern computer with TNR or any similar font, the left edge of the second letter should be distinctly underneath the right edge of the f's loop. If they were produced on any 1970-ish office equipment, it is inconceivable to me that so subtle a bit of typography would be present.

I *think* I can see the typical computer-generated TNR f-spacing in the memos--the following e's and r's seem to be underneath the loop. But frankly, the quality of CBS's PDFs is too poor for me to be sure. I hope they will release versions of their best copies scanned at high resolution so we could make better tests.

By the way, I'm not wedded to the font being Times New Roman; I was just using that as an example. Many other bookish fonts, e.g. Garamond, also have a lowercase f that places the following letter under its wings.

Robb Allen says:

on 9/10/2004 11:12:00 PM

Yes, I've retracted my original statement because of discrepancies in my methodology. There's plenty more to look at though, and I'd still like to hear W tell Danny Boy to shove it where the sun has no business shining.

Steve Poling says:

on 9/11/2004 12:56:00 AM

Kerning is a digraphic thing, it depends upon the pair of adjacent letters. I used Word with kerning turned on and off and tested CYA. With Kerning on, the letters overlap, with it off, the letters do not. The letters CYA in the 18 Aug 1973 memo show overlap.

alfonso says:

on 9/11/2004 2:17:00 AM

I disagree with the previous post. I did the same experiment (turning kerning on & off for capitals "CYA" in MS Word) and the supposed 1973 memo matched the "no kerning" setting. The key is to look for the spacing between Y and A in CYA. Kerning is not the smoking gun. Centering is.

Rock Lobster says:

on 9/11/2004 4:21:00 PM

Excellent article. I checked the "f"s in the forged memos, and put together some images and notes here: http://www.freerepublic.com/focus/news/1213112/posts

F15C says:

on 9/11/2004 10:34:00 PM

Mighty big of you to admit when you've made a mistake. You have my respect and continued readership. Others in the blogsphere would not have courage to do so.

Stay true.

Robb Allen says:

on 9/11/2004 11:59:00 PM

I'm in this for the truth. Truth cannot be obtained if your pride comes first. Even if I was able to determine these were factual documents, I would concede that as well.

Besides, there are so many other glaring issues that whether an 'f' overlaps on pixel or not isn't even that important.

The centering of the header is even more damning if you ask me.

PilgrimFarAway says:

on 9/12/2004 4:35:00 AM

This is the difference between the Old Media and the New Media. The Old Media buries it's retractions on page D17 and quietly pulls the article. Honorable bloggers run the retractions and corrections right in front of God and everyone.

I respect your honesty. If I can find a tip jar, I'll hit it.

Robb Allen says:

on 9/12/2004 10:28:00 AM

I appreciate the gesture as I don't have a tip jar. If your so inclined, I'd prefer a donation to a worthy cause to help the victims of Charley and Frances as I was in the path myself and know there are lots of people who need the help more than I (heck, I only lost power for 2 days and had minimal damage to a fence.)

But above all, keep looking for the truth. Even if the truth goes against what you want it to.

Dr. Dave says:

on 9/12/2004 2:56:00 PM

I think that what is so fascinating about this is the experimental tack it has taken. The site

http://shapeofdays.typepad.com/the_shape_of_days/2004/09/the_ibm_selectr.html

shows the results of attempts to recreate the documents in question on the IBM Selectric Composer, putatively the best candidate machine available in the early 1970's for producing all the effects in the memos. People should look for themselves. I am a research chemist, and this is the most fun I've had following a news story ever. Show me the data! I'd especially like CBS to pony up the originals.

I think it's grand that the blogosphere is handing the MSM their butt, and I think that this could usher in an era where a watchdog internet-based community would keep an eye on stories coming from either side of the political spectrum. While the MSM has said in effect, "How do we know who is writing blogs" as if the journalistic profession has some mysterious way of finding truth. What I am impressed by, and this thread demonstrates this, is that people posit hypotheses, and if the evidence weighs against them, they abandon them. It's a lot like how science is done- someone looks at data, makes a stab at explaining it, and the community beats on the idea until it either collapses or is thoroughly examined and found compelling.

Comments have been closed on this topic.

Comments

Archives

Categories

Comments

Archives

Categories

Second Amendment

Sharpest Marbles

Blogroll

Funny, as in Ha-ha

Good Reads

Extended Blogroll

Industry Links