Jump to content

Talk:Killian documents controversy/Archive 7

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
< Talk:Killian documents controversy

This is an old revision of this page, as edited by 71.212.31.95 (talk) at 18:05, 8 October 2006 (→OT discussion about authenticity: evidence?). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.


OT discussion about authenticity

Hi. I took your advice and in from the Little Green Footballs wiki. That's a very curious date on your posting, since that's just after when I showed how the DoD records acutally do contain a proportionally printed document, the only one in the entire database of Bush's records and strangely and quietly released just a couple of days after CBS backed away from authenticating the memo: http://www.defenselink.mil/pubs/foi/bush_records, "Documents Released on September 24, 2004," page 6 (also note the funky fonts in the documents on pages 3&4). If you're little fuzzy eyed about recognizing proportional spacing, go here: http://www.aheckofa.com/FoolMeOnce/BushRecordPSClipRuler.jpg

Also, I don't understand why you're using that same murky instance of superscripting, that "Report to 111th F.I.S." when I had already pointed out in the LGF thread that there are much nicer, cleaner examples here: http://www.defenselink.mil/pubs/foi/bush_records, "Part 6," page 45. It's undated, but it's rather obvious that it's not a Word document.

But you are indeed right that "a feature of modern computer word-processing programs such as Microsoft Word automatically changes "th" to superscript characters when following numerals," but if you look at the superscripting in *all* of the memos, you will note some strange discrepencies. A comparison of the superscripting pattern in the memos to what happens when you type them up in Word can be found here: http://www.aheckofa.com/FoolMeOnce/SScriptsCompared.jpg

Hmmm, not only are only some of the th's following a number not superscripted, but none of the st's. There is a strangely inserted odd space that can explain away some of this, but not others.

The discrepancies aren't at all strange. They look like exactly what you'd expect if somebody were trying to avoid the auto-superscripting. Not that that proves anything. Also note that there are multiple ways to defeat the auto-superscripting in Word, including editing afterwards, using an el instead of a one, and IIRC just hitting undo after typing. But what the inconsistencies do suggest is that it's unlikely that a skilled typist familiar with the device they were using made these memos in the normal course of their work. - Thomas Phinney 10:32, 30 September 2006 (UTC)

Your argument makes no sense. We're not talking about a pile of documents, just 6 memos in total, some very short. If you figure out how to get around Word's auto-superscripting, then why not get rid of all the superscripts instead of just leaving them off and on in such an odd, random pattern (except for the "st's," none of which are superscripted)? The hypothesis is that the forger used a modern PC or Mac to create the memos, so if he or she makes a mistake, like leaving in superscripts, it would only take a few seconds to fix it and print it out, so why leave them? Lack of time, maybe? But obviously the forger would have had to have spent an awful lot of time to have achieved the detailed matchup in the contents and dates of the memos to whose of the DoD records. There was even a reference to "Bath" in a very short memo USA Today had, meaning James Bath, who was also suspended from flying, evidently again by Killian, just one month after Bush was, but this information had been redacted from the DoD records in 2004, and it was only by chance a researcher had already copied seen the pre-redacted version: http://sugarinthegourd.com/redacted.html

And better yet, why not just use one of the many, MANY typewriters still common in offices to type up forms and labels? Was it that the forger was somehow both very, very smart and very, very stupid, including being completely illiterate with using Word (not that Word has been shown to be able to convicingly recreate the memos)?

As far as whether typewriters from the early 70's can do this sort of stuff, for some strange reason, all these typewriter experts ignored IBM Executives, or just mentioned them in passing without any demonstrable tests, despite these models being pretty common all over until being gradually replaced by businesses and agencies with the more reliable Selectric models. Those clunky old Executives could do a few tricks that the sleek Selectrics couldn't do, though. These are from a "Model C" Executive: http://www.aheckofa.com/FoolMeOnce/ExecSuperscript2.jpg and http://www.aheckofa.com/FoolMeOnce/ExecSuperScript3.jpg. A sharp eyed person will note that both examples are also proportionally spaced.

But undiscussed by all these so-called typewriter experts is the Diablo Systems daisywheel printer, made since 1969, and evidently common by 1972 when Xerox bought Diablo for $28 million. Diablo's cofounder, David Lee, left Xerox in 1973 to found Qume. Qume also made daisywheel printers that were code and plug-compatible with Diablo's to the point that ribbons and print wheels with interchangeable. It's important to note the the early Diablo printers were strictly for the OEM market, and hence sold under a large variety of names. See http://goldsea.com/Innovators/Digital/digital.html, http://mfelker5.tripod.com/printwheels.htm, http://www.okoffice.com.au/product.asp?pID=44193&cID=879

The Diablo is a big issue because all Diablo models could proportionally print with same spacing shown in the memos, and well as being being fully capable of full super/subscripting (via half line feeds). This is a code set for the Diablo's: http://www.nefec.org/UPM/dblomain.htm And besides having a variety of standard printwheels, evidently it was easy to get customized ones if a special character or two was desired. A PDF file with some reprints of computer-related articles from the 1980's has some backgound on this stuff,starting on page 14: http://www.tinaja.com/glib/atg1.pdf

The only serious dispute for the memos being created on a Diablo has been a claim by some die-hard pro-forgers that, yes, maybe Diablo's could only do all this, but in only 1974+ models. This is apparently related to some patent applications found on the Internet. However these patent apps seem related to a dispute that formed between Xerox and Qume when Qume started selling its Diablo compatible printers in 1973/74, and which eventually resulted in a lawsuit: Qume Corp. v. Xerox Corp., 207 U.S.P.Q. 621 (N.D. Cal. 1979). However all the information available indicates that all Diablo's, from their inception, came with a core code set that included the proportional and super/subscripting commands (they could also function as high-precision plotters.)

Hope this clarifies. And if someone wishes to dispute any of this, please do me the courtesy of providing some relevant countering evidence in kind instead of opinion or links to opinion. Obviously I've tried at least to be very thorough in presenting my case. I'm obviously in a very small minority with my view on the memos, but to paraphrase Anatole France, if fifty million people believe a foolish thing, it is still a foolish thing. -BC 209.6.203.244 16:39, 28 September 2006 (UTC)

Joseph Newcomer considered and rejected the Executive, for several reasons, along with the Composer, at http://www.flounder.com/bush2.htm (click on the 18 Sept 04 update for the page about the Executive; each of the colors leads to a different multi-anchored page.) I do not remember him addressing the possibility of a HyType machine. --htom 17:41, 28 September 2006 (UTC)

Ah yes, Joseph Newcomer, PhD, long on words (and words, and words....), but a wee bit short on actual demonstration. As you might have guessed, I've been to his site and found it a bit lacking in actual scientific methododogy. You do not remember him addressing the Diablo HyType because he never did, as was the case with all these alleged experts that came out of the woodwork back then, like Peter Tytel, the "typewriter expert". I guess we can cut Tytell some slack since a daisywheel printer technically isn't a typewriter. But one can't really do the same Newcomer -- he makes an awful lot of claims about his expertise. From http://www.flounder.com/bush2.htm: "I am one of the pioneers of electronic typesetting." and it just goes on and on from that. Also, he boldly states, "The probability that any technology in existence in 1972 would be capable of producing a document that is nearly pixel-compatible with Microsoft's Times New Roman font and the formatting of Microsoft Word, and that such technology was in casual use at the Texas Air National Guard, is so vanishingly small as to be indistinguishable from zero."

But as I've more than enough demonstrated, by 1972 there were two widely available devices capable of printing proportionally, and in superscript, and with a small font: the IBM Executive typewriter and the Diablo daisywheel printer. Both devices have been obsolete for decades, but would expect any sort of real expert to show due diligence in researching all the devices available around that time that might have possibly created the memos, discover and examine what their capabilities were, try to find print samples and/or specifications, and so on before rendering any sort of conclusion. In Newcomer's case, the Diablo is not mentioned once on his entire site, and the Executive typewriter is not even examined -- and even then, his sample is only a wedding program -- until until about a week, Sept. 18, 2004, after he makes the claim of forgery on or before Sept, 12, 2004. If that doesn't sound very scientific, that's because it's not: http://www.pages.drexel.edu/~bcb25/scimeth/intro1.htm

I seriously considered and dicarded the Executive, for multiple reasons, by the end of the first weekend after the Killian memos came out. Sure, it's a proportional device. However, in my mind the real question (1) whether the device was capable of the particular particular spacing in question (the IBM Executive was not, as I mentioned earlier on this page back in July), and (2) whether anything very similar to the Times Roman used on the memos was available for that device(again, not for the IBM Executive). See http://www.creativepro.com/story/feature/21939.html?cprose=5-39. - Thomas Phinney 10:38, 30 September 2006 (UTC)

I also had seriously considered the Executive. I did some research and got my hands on a copy of a technical manual that the author remembered being created on a Model "C" Executive. It was proportionally spaced and had full super/subscripting in small fonts. But as evidently with you, I found its proportional spacing to not match up with that shown in the memos. I don't think you can say much about the font or typeface since the Executives came with a choice of typefaces.

The point I was making is that there were at least two devices available around 1972 capable of proportional spacing and full super/scripting: the IBM Executive typewriter and the Diablo printer. Considering that the forgery charge initially with the false claim by the likes of "Buckhead" that there were no devices then that could proportionally space, this was not a trivial matter. Think about it -- the whole "there were forged" business got started with some then anonymous blogger making nonsensical (that is if you had bothered to check) claims about typewriters then and even things like Wang word proccessors could or could not do. And then have these self-proclaimed experts like Newcomer coming out of the woodwork discussing Selectrics and Veritypers and also claiming forgery before even looking at an Exexcutive and apparently utterly unaware of the Diablo. Of course our diligent media didn't exactly help to clear up matters... -BC 209.6.203.244 14:50, 30 September 2006 (UTC)

Newcomer actually inadvertently provides strong evidence that Word was NOT used for the memos. Go back to his site here: http://www.flounder.com/bush2.htm#12-Sep-04 There are two things to look at: one is the poor matchup character-by-character between his Word-created "from" and a memo "from". If you do further character-by-character matchups, you will find that the Word Times New Roman font is consistently inconsistent with the font used in the memos, especially in terms of height.

And further up the page, he makes an extremely damning admission: "I was a bit annoyed that the experiment dealing with the 18-August-1973 memo was not compatible, until I changed the font to an 11.5-point font. Then it was a perfect match, including the superscript 'th'". And then a little further down he tries to explain this away with, "However, this might be an accident of the many levels of transformation from the original (wherever that is) and the photocopying, scanning, document conversion, and re-printing. The 11.5-point font could represent a reduction to 96% of the original size in the various transformations."

Well....

First off, MS Word uses long established character spacing for Roman-style fonts. WordPerfect for DOS used similar spacing with its "CG Times" font, and it was also the default proportional spacing used in Diablo and Qume daisywheel printers (and don't forget that many of the early laser printers has a Diablo mode to "letter quality" printing). So you would expect at least a modest matchup between Word and a Diablo in terms of spacing, but an *exact* match? No, and the letterheads on May 4th and Aug 1st memos prove this. As well as eliminate Word.

"First off, MS Word uses long established character spacing for Roman-style fonts. WordPerfect for DOS used similar spacing with its "CG Times" font..." What do you mean by that? Both applications simply use the spacing built into the fonts (whether it was a system font or a printer font, depending). I have similar difficulty with the rest of your comments on the printers. I did a very careful analysis of the spacing; in order to match the memos, the font spacing would have to be a lot closer than any general non-font-geek's idea of "similar." (No offense intended, but if you think the font could be Palatino, you're not looking at the same level of detail you need to be.) - Thomas Phinney 11:13, 30 September 2006 (UTC)

There was some research with different "Roman style" fonts, both with Word for Windows and Mac, as well as WordPerfect for DOS, to see how they compare to the font used in the memos: http://www.aheckofa.com/FoolMeOnce/CBSBushMemos.html#FunWithFonts Note how all the different PC-generated fonts both match and *don't* match the memo font in some aspect, including even Times New Roman, although they all indeed show "similar" spacing as I claimed. I can't return to this until tomorrow, but I would be interested in your opinion since you apparently genuinely looked at this stuff (I am just a troll after all.) -BC 209.6.203.244 14:50, 30 September 2006 (UTC)


You can demonstrate this on your own if you are handy with a PaintShop type program. First grab the May 4th and Aug 1st memos from here: http://www.cbsnews.com/stories/2004/09/08/60II/main641984.shtml Now replicate one of the letterheads as best you can using Word, print it out and then scan it in. Now then superimpose your Word replica as best you can on either of the letterheads and feel free to resize all you want. You should end up with what you might think is a pretty good match at first sight.

But now try superimposing one of the memo letterheads on the other. You will then find a true dead-on match that is noticeably much, much more accurate than your best efforts with creating a Word replica. You can see one such experiment here: http://www.aheckofa.com/FoolMeOnce/HeadersCompared.jpg The two upper samples were made from overlaying a Word-recreation of the letterheads over each of the two memo letterheads; the lower sample is from overlaying the two memo letterheards over each other.

But if there was all this claimed distortion and smearing that occurred when the forger tried to make the memos look old, how can that be that the memo letterheads are dead-on copies of each other?

If you believe the forgery scenario, then the May 4th and Aug 1st memos were originally created in Word and then were deliberately distorted to make them look old. But then how come they still end up as perfect matches to each other but not to a fresh Word replica? Where is "The 11.5-point font could represent a reduction to 96% of the original size in the various transformations" that Newcomer uses as an excuse to for using a very non-standard font size?

All I can say is tsk, tsk... -BC 209.6.203.244 00:14, 29 September 2006 (UTC)

I don't think there's much need to disprove the Diablo HyType, as the first version of this venerable daisy wheel printer appears to have been introduced in 1973 (apparently in the fall), rather than having been in use in, say, the spring of 1972. At least, that's what every source I can find suggests. Do you have any actual evidence that the HyType was available in early 1972?
There were certainly a few proportional-spacing devices available at the time, but the Executive was not capable of doing the fine level of proportional spacing shown in the memos. It offered a much coarser set of predetermined widths than the memos, and its most similar available typeface was much wider than the Times Roman used in the memos.
As for the memos looking similar to each other, they all went through the same set of operations as far as photocopying, faxing, and being scanned into Adobe Acrobat. So it is perfectly reasonable that they would suffer the same kind of overall scaling distortion. None of the experts have claimed that they "were deliberately distorted to make them look old" - that is a straw man argument you created, not any kind of reasonable interpretation of your opponents' views. -Thomas Phinney 10:10, 29 September 2006 (UTC)

Please provide at least one alleged source for the statement "I don't think there's much need to disprove the Diablo HyType, as the first version of this venerable daisy wheel printer appears to have been introduced in 1973 (apparently in the fall), rather than having been in use in, say, the spring of 1972. At least, that's what every source I can find suggests. Do you have any actual evidence that the HyType was available in early 1972?"

Define "actual evidence" -- I already showed that Diablo Systems as an independent company began making daisywheel printers in 1969 and was bought by Xerox in 1972 for $28 million, and that one of the co-founders of Diablo left Xerox in 1973 to form a rival daisywheel printer company called Qume. What do you think Diablo was doing between 1969 and 1972 when Xerox paid all that money for it? When Xerox bought Diablo, they got the HyType I. If that isn't enough proof, I found some ribbon crossreferences for the HyType I ribbon, which was mostly, but not completely replaced at about 1974-75 with the HyType II ribbon, and they link to a bunch of old, forgotten dedicated word processors dating back to at least 1972. If a 1972 word processor uses a Diablo HyType I ribbon, that kind of suggests that perhaps, perchance HyType I's were being used then. The best ribbon crossreference I found is no longer online, but this Swiss one isn't so bad: http://onlineshop.jpd.ch/sp/farbband-multistrike/qume-1145-farbband-multistrike-original-schwarz.asp

Note how you won't see either "Diablo" or even "HyType" mentioned -- Diablo Systems was originally strictly an OEM supplier.

Here's the promotional brochure from Diablo that came out in September 1973. http://archive.computerhistory.org/resources/text/Xerox/XEROX.Diablo_HyTypeI.1973.102646256.pdf#search=%22Diablo%20HyType%22. But as you point out later, even if 1972 [Edit: oops, meant 1973]was the year it was introduced under that name, it wouldn't matter if the HyType was available earlier as OEM equipment. So what we'd need is some specific model of printer that was made in that way, to research. Do you have any specific names of printer models that one could research? - Thomas Phinney 11:42, 30 September 2006 (UTC)

Well, I Googled some of the CPT models in that Swiss ribbon crossreference and found this computer timeline mentioning the CPT model 4200 word processor as being introduced in 1972: http://www.computermuseum.li/Testpage/01HISTORYCD-Chrono1.htm

Since the Swiss ribbon crossreference for the HyType I ribbon lists the CPT 4200.... -BC 209.6.203.244 03:36, 1 October 2006 (UTC)

My comments about the Executive were related to how it was not examined, along with the Diablo, by any of these alleged experts before they plunged ahead with their forgery charges. I strongly suspect that only reason why all discussions focussed on Selectrics, at least initially, was simply because there were and are Selectrics still in operation all over (primarily for typing up labels and filling in preprinted forms) and hence handy and not requiring much work. Since almost all functioning daisywheel printers and IBM Executives have long ago been scrapped, Selectrics became the default representives of all 1972 office technology. Until Marian Carr Knox mentioned that she used an Olympia typewriter. So then it was Selectrics and Olympia typewriters that represented all 1972 office technology, with a side discussion of Selectic Composer, a typesetting machine that would never have been used to create just memos. Executives made a belated appearance here and there, but Diablo's not at all. All of this represents inexcusably shoddy journalism by the media, and laughably incompetent "research" by all of these alleged experts.

As far as the aging of memos go, your claim that "None of the experts have claimed that they "were deliberately distorted to make them look old" - that is a straw man argument you created, not any kind of reasonable interpretation of your opponents' views" is demonstrably false -- I never said "experts" and many claimed this. Simply Google "Killian memos aged" for proof. And that excerpt off of Newcomer's site I used had this comment, "However, this might be an accident of the many levels of transformation from the original (wherever that is) and the photocopying, scanning, document conversion, and re-printing," basically claims that the memos were distorted through muliple processes. Indeed the first, original blog post that started the forgery bandwagon states (from http://www.freerepublic.com/focus/f-news/1210662/posts):

Did you fail to complete your quote here? I did some browsing, but I have yet to find any expert who claimed the memos were "deliberately aged." Certainly I did not, and neither did Newcomer. - Thomas Phinney 11:49, 30 September 2006 (UTC)

?? This is a complete quote of what I wrote: "If you believe the forgery scenario, then the May 4th and Aug 1st memos were originally created in Word and then were deliberately distorted to make them look old." Did I mention "expert"? No. Was this a central part of the forgery claim? Yes -- from the original Buckhead comment: "I am saying these documents are forgeries, run through a copier for 15 generations to make them look old." Did I say that Newcomer also claimed this? No, but I pointed out that he justified using an oddboall point size, 11.5, to *supposedly* dupe a memo in Word (he of course didn't show his results, choosing only to just talk about them): "However, this might be an accident of the many levels of transformation from the original (wherever that is) and the photocopying, scanning, document conversion, and re-printing." And you -- are you an expert? -- I never brought you up in any of this.

You say that "the first original blog post... states (link):" and then you don't have a quote after the link and the colon. That's why I asked if something was left out.
And yes, I am an expert on typography. I gave a lecture on forensic typography at the 2002 international typographic conference in Rome, and a lecture on the typography of the Bush memos in particular at the 2004 St Bride conference in London (St Bride is a library dedicated to the history of printing). I'm writing this from Lisbon, where this year's international typographic conference (ATypI) is just wrapping up, and I did four talks in the "Type Tech" section, and one in the main conference. If you Google "Phinney Bush memos," follow the links to the Washington Post, CreativePro.com and Newsroom-l for coverage of my thoughts and quite detailed analysis of the spacing of the Bush memos. (I recommend the CreativePro article in particular, as it has the most detail.) I concluded it was some version of Times, with the 18-unit-to-the-em spacing that is not possible with any device so far suggested as having been available and plausible in 1972 (we're not counting actual phototypesetting or metal typesetting as plausible here). - Thomas Phinney 15:17, 1 October 2006 (UTC)


"Buckhead": To: Howlin

Howlin, every single one of these memos to file is in a proportionally spaced font, probably Palatino or Times New Roman. In 1972 people used typewriters for this sort of thing, and typewriters used monospaced fonts. The use of proportionally spaced fonts did not come into common use for office memos until the introduction of laser printers, word processing software, and personal computers. They were not widespread until the mid to late 90's. Before then, you needed typesetting equipment, and that wasn't used for personal memos to file. Even the Wang systems that were dominant in the mid 80's used monospaced fonts. I am saying these documents are forgeries, run through a copier for 15 generations to make them look old. This should be pursued aggressively."

The font used was neither Palatino nor Times New Roman, but Times Roman. http://blogs.adobe.com/typblography/2006/08/bush_guard_memo.html - Thomas Phinney 11:42, 30 September 2006 (UTC)

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! I did not know Hailey had this second report, http://imrl.usu.edu/bush_memo_study/supporting_material/bush_memos.pdf!! This report independently comfirms some of the discrepencies I had long ago noted -- the memos *were* done on an impact device and that they were only "Roman like" and neither Times New Roman and very, VERY unlikely Times Roman as well (Macs use Times Roman and they have the same problems as Windows Word, so your contention that they were done in Times Roman is unconvincing without real life samples). And despite all the prior attacks and criticisms, Hailey still contends that some sort of impact printing device with platen was used. Hmmm, what impact printing device with a platen could print proportionally, I wonder...) To quote the first part of Hailey's conclusions:

No, Hailey is wrong about the typeface: it is Times Roman, as cited in my blog post above. There is not even a shred f doubt. Hailey is not a typographer, and the fact that he did not know about the two biggest versions of Times in the world and therefore look at both of them is evidence of that. (There's the Linotype/Adobe/Apple version, and the Monotype/Microsoft version). My post above explains exactly where he went wrong. I challenge you to find a typographer who will look at that evidence and disagree.


I believe the memos were typed for the following reasons:

1. They cannot have been done in Times New Roman, so the argument that they were done digitally has no logical support.

2. The evidence of character damage is no longer in question; the "t," "e," "a," "c," "R," "o," "M," and "N," are all clearly defective, and in each case the character has unique defects. Other characters not discussed also show signs of being defective.

3. I found good evidence that characters interacted with each other, something only possible with a typewriter or other device that produces characters one at a time and involves physical impact.

4. Spacing in the memos is consistent with using a platen and not consistent with Word or similar digital processes: spacing of the heading is not centered; headings do not align; fractional returns are consistent with adjusting with a platen; left edge of several memos appears to have drifted to the right causing characters to penetrate the left margin

I especially liked his bit about the centering bit, because as I had posted earlier, if you go here, http://www.aheckofa.com/FoolMeOnce/HeadersCompared.jpg, you will find the two upper samples were made from overlaying a Word-recreation of the letterheads over each of the two memo letterheads; and the lower sample is from overlaying the two memo letterheards over each other. Note how the upper samples don't align, while the bottom one does so precisely. If there were differences with how the centering was done between some 30 yr old mechanical device and a modern PC/Printer combo, this would be exactly the sort of discrepency you would expect.

I'm going to have to leave this merry little discussion for a while to look into a few more things, including revisting whether "HyType I" was just a low end daisywheel model incapable of the precision needed for proportional printing (this is another reference to the Adam Coleco and note the ribbon type: http://www.myoldcomputers.com/museum/comp/adam.htm) But I gathered some extremely useful stuff here. Thanks. -BC 209.6.203.244 13:48, 1 October 2006 (UTC)

Please note all of the false claims about technology -- either "Buckhead" had no clue or was deliberately lying. But this misinformed nonsense is what got the forgery charges rolling.

Also demonstrably false and/or disengenuously misleading is your claim that, "As for the memos looking similar to each other, they all went through the same set of operations as far as photocopying, faxing, and being scanned into Adobe Acrobat. So it is perfectly reasonable that they would suffer the same kind of overall scaling distortion." If the memos were created in Word and then copied or faxed X amount of times, deliberately or not, if there any distortions at all, they would vary from copy to copy. You create documents A, B, C and so on, and then recopy and refax them enough times to noticeablely distort them from the originals, they're going to distort from each other just as much -- it's either a random process or else there is a defect in one of the devices that would introduce distortion on a particular area on the page. So if you start off with two identical letterheads, run each one through a series of copying/faxing/scanning operations and the final copies are still dead on perfectly aligned with each other, with no discernable ghosting or smearing, that means your equipment introduced no significant distortion.

No, you were right the second time with "or else there is a defect in one" (or more) "of the devices...." This is not unusual among faxes and photocopiers, and pretty well understood. - Thomas Phinney 11:59, 30 September 2006 (UTC)
Also, distortion is not necessarily "on a particular area on the page." The more common distortions include non-square reproduction (causing very slight stretching or compression) and uneven skewing where the skew increases as the original is pulled into the machine's auto-sheet-feeder. - Thomas Phinney 08:05, 1 October 2006 (UTC)

Now suppose after you do all that, someone hands you an original document with a very similar looking letterhead, and it's claimed that it was created on the same computer/printer combo that you used to create the documents you ran through all that processing on. Now, you're already determined that the processing introduced no significant distortion, so if the letterhead on that new document doesn't match up identically, then it was NOT created on the same computer/printer combo. And that's the case here: the memo letterheads line up with each other perfectly, with not even the slightest bit of discernable ghosting. The same letterhead created on Word will misalign and ghost in some area regardless of how much you adjust the scale and do your darndest to best fit them. There isn't a lot misalignment, but it's definitely there and very, very much like that shown in Charles Johnson's infamous CYA animated gif: http://littlegreenfootballs.com/weblog/pictures/Pictures/aug1873-pdf-animate.gif

If you take the non-forgery scenario where the documents were created on a very different device that used the same standard proportional spacing and a Roman-style font, and you then overlayed Word replicas and adjusted the scale and positioning for best fit, you should expect to get a reasonably good match but with some misalignment and ghosting at least. Which is what you would end up with if the memos had been created on daisywheel printer, with proportional mode turned on, and a little "th" character used on one of the extra printwheel spokes (and without an additional extra spoke for an "st").

Perhaps I am misunderstanding you, but it really sounds as if you think there is a single "standard proportional spacing." That is generally very untrue, whether it's typewriters or digital fonts we're talking about. With proportional typewriters of the era, each had its own system of possible letter widths that each font had to be adapted to fit to. In the case of the Selectric Composer, each particular letter had only a single possible width for, so every font was set to the same set of widths on that device. But all of the typewriter class devices for which I have specific information, which existed in 1972, were not capable of the 18-units-to-the-em system used for Times Roman in the memos - Thomas Phinney 12:14, 30 September 2006 (UTC)
The Diablo Systems Inc. Model 1200 HyType I Printer Maintenance Manual (Pub. No. 82003, 2nd ed., Nov. 1974) indicates that the printer supported a print line of "132 Columns @ 10 characters/in. (3.95 char/cm)" and "158 Columns @ 12 characters/in. (4.76 char/cm)" with column spacing of "60 Positions/in., 1/60th in./increment (23.6 pos./cm 152.4 mm/increment)" (Table 1-1, p. 1-1). The I/O interface included 11 data lines to carry BCD information representing carriage movement values. The high order bit represented the carriage movement direction. The ten low order bits represented the carriage movement distance, "in increments of 1/60th of an inch. Six increments equal 1 character column at 10 characters or columns per inch, while 5 increments equal 1 character column at 12 characters or columns per inch." (p. 4-2). This indicates that the Model 1200 HyType I printer was not capable of producing the 18-units-to-the-em system used for Times Roman in the memos. 71.212.31.95 20:11, 30 September 2006 (UTC)
Great, thanks! So for 12 point type, that would be 10-units-to-the-em, which is not good enough, as you say. That would eliminate that particular device as being capable of producing the memos. Is this information from the manual available online somewhere that you can point to? - Thomas Phinney 00:52, 1 October 2006 (UTC)
A PDF of the manual can be found here: http://www.bitsavers.org/pdf/diablo/82003_Hytype1Maint_Nov74.pdf 71.212.31.95 01:03, 1 October 2006 (UTC)

I'm not sure what that eliminates. Apparently Diablo made different daisywheel models, and the HyType I may actually have been just a low end model. All of the known Diablo code sheets indicate that at the very least most Diablo models can indeed do proportional printing in two ways: with with a simple ESP P command or via a much trickier character mapping using horizontal indexing and control -- see: http://www.nefec.org/upm/dblofrm.htm The web info is really sketchy on pre-Xerox Diablo stuff. What's needed is manuals for the OEM printers Diablo made for the likes of CPT. It can be verified, though, that by 1975, the dedicated word processing market was huge: 1975 Business Week Article Note the number of "Redactron" systems sold, and note also that Redactron models are also listed in the Swiss ribbon crossreference.

It eliminates the possibility that the Diablo HyType I daisywheel printing mechanism was used to produce the memos. Of course, you have never produced any evidence that any system using the HyType I was commercially available in 1972. Or that any such system supported proportional fonts. Or that any other system reasonably available to Killian at that date could have produced the proportional spacing exhibited in the memos. Your material concerning products from 1975 and later has no relevance. 71.212.31.95 16:07, 1 October 2006 (UTC)

Actually it doesn't now. I just found two sites with further interesting info.

This tech document discusses the proportional print characteristics of old devices -- apparently unrelated to Killian stuff -- including Executive typewriters, (which had resolutions of 1/32" to 1/36") and how some daisywheel printers at least were proportionally printing with just the 1/60" resolution, which is the same as the HyType I: http://www.quadibloc.com/comp/propint.htm

The table near the end of the document clearly shows the difference between the 18 units-per-em spacing of Monotype Times New Roman and the 9 unit system of the Selectric Composer, the 5 unit system of the Executive, etc. You don't seem to understand the implications of this. It shows that these machines cannot reproduce the spacing of typeset-quality TNR. As already noted, the 60 positions per inch spacing capability of the HyType I mechanism would support only 10 units-per-em for 12-point type. (The 60 positions per inch spacing of the HyType was undoubtedly chosen because 60 is the LCM of 10 and 12 and therefore the smallest value that allows both 10- and 12-pitch monospacing to be expressed as an integeral number of units.) 71.212.31.95 20:27, 1 October 2006 (UTC)

This is a belated response to that particular comment, but I've started looking more into the mechanics of having a daisywheel with either a 1/60" & 1/120" resolution printing specific proportional typefaces like Times Roman versus Times New Roman versus one that supposedly the Redactron model "Redactor II" used called just "Times". When I went to look up if just plain "Times" is a trademarked font, I found that it was, thereby stongly implying that the Redactor II would actually print it as such. So I thought to see how this "Times" compared to Times Roman and Times New Roman, and went to this site: http://www.myfonts.com/fonts/agfa/times/morelikethis.html This conveniently allows you type in text and see how it look in Times Roman, Times, and Times New Roman all at once. Cool. But I noticed something very curious right away: The "Times" and "Times New Roman" matched up rather closely, but not "Times Roman" -- its proportional spacing was a little wider. Hmmmm.... When I looked into that, it looks as though that the widths and such have been a bit conflated in recent decades for "Times Roman" and "Times New Roman", especially so with the introduction of TrueType on both Apple and Windows. But the plain old "Times" evidently has not been. What's especially curious it that the last modified date year for "Times New Roman" is 2000, while for plain old "Times," it's 1974! Yet they seem evidently identical enough to pass for each other, especially if, say, you only had some coarse copies to look at. My, my....

But the librarian at www.cbi.umn.edu go back to me about the contents of that "Specifications for Redactor II Printer System" document dated 1973, but apparently the system was forthcoming at that point and not yet in producion. That pushes its manufacturing date back at least a year, and it was pointed out that the Business Week had mentioned that it was delayed for a year after it was announced, so we're probably looking at a 1975-76 manufacturing date at the earliest, darn it.

But...the libririan gave me other info that could be useful. Stay tuned.... -BC 209.6.203.244 18:45, 4 October 2006 (UTC)

I didn't mention the issues with the font mechanics of using a daisywheel with 1/60" resolution or another with 1/120". I reluctantly agree that 1/60" would not be enough for a non-right justified line. For true 12-point Times text, you would need a resolution of 1/108" by my calculations. The standard Times XX width ratios could be fudged a bit with 1/60", but would be noticeably off on lines of non-justified text compared to a Word created lines of Times. With right justification, depending on the sophistication of the spacing algorithm, it may be passible given the coarse quality of the memo copies, as PC Magazine demonstrated with this comparison of a text created on an IBM Selctric Composer, which apparently had a resolution of 1/72", to a Word recreation: http://www.pcmag.com/article2/0,1759,1644869,00.asp

But a daisywheel with 1/120" resolution can obviously do 12 point Times accurately, but if you want to keep the math simple and deal with integer proportions, the typeface used by the Daisywheel would likely have an equivalent point size of 10.8, which may have been rounded off to 11 point. Which makes more sense since the IBM Selectric Composer, which used comparable typeface sizes, was cited as being in 11 point by this earlier cite of mine: http://www.quadibloc.com/comp/propint.htm

But unlike current digitized fonts, which spacing and size is automatically scaled, the type size wasn't tied in with the spacing setting on daisywheels -- if you selected "Times" spacing on your Redactron or whatever word processor, it'll proportionally print whatever printwheel you stick inm even if it's something like a Courier. FYI. -BC 209.6.203.244 13:46, 5 October 2006 (UTC)

Looks like we're making progress. You now understand why the 1/60" positioning of the HyType I is significant. You also understand that the "proportional spacing mode" built into later printers is different from true proportional spacing based on font-specific metrics. And you understand that in unjustified text (e.g. the Bush memos) discrepancies in character widths become readily apparent (even in copies too poor to identify the font from the glyph shapes) because they accumulate along the lines of text. This is all good. Also keep in mind that the Selectric Composer was not really a typewriter. It was a very sophisticated and expensive machine designed specifically for low-end typesetting applications. It is not a reasonable possibility that Killian would have used such a machine to produce the memos. And note that even with a daisywheel mechanism capable of 1/120" positioning, sophisticated controller electronics and/or software is needed to implement typeset-quality proportional spacing with such a device. That really only became feasible with the availability of microprocessors. 71.212.31.95 16:01, 5 October 2006 (UTC)

I was thinking you could fake it a bit with the 1/60" resolution, depending on the sophistication of the computer coding, but realistically and practically, that was only a possibility with right justification, and even that would likely fail with close scrutiny of the high rez copies that Hailey is suppose to have in his possesion. The big question now is whether the HyType I, with its 1/60" resolution, or some immediate forebear prior to the Xerox acquistion, was the only daisywheel mechanism in wide use. If there was indeed 1/120" rez model as at least an option as well at that time included with word processing systems, then the forgery thing is dead. That Swiss ribbon site shows a huge number of Diablo models that I've never heard of, and I suspect that the bulk of those are pre-1980, so it's very, very likely there were indeed 1/120" rez models in common use around 1972. As far as printer control goes, that was not at all a primitive time techwise, and it doesn't exactly take a lot of memory and CPU power to map out character spacings. The CBI library has detailed specs on the Redactron Redactor II, and I'll get copies of that one way or the other, and that should confirm the capabilities. That should push back the date of when the memos could have been created as is to about 1976. While that's not 1972, it's definitely within striking distance now. All I would then need is a good reference to either an earlier Redactron model or one from a competing product with similar capabilities, but from 1972. Bear in mind that having the memos being able to be created on a circa 1972 device would completely fit in with all of the other evidence, including explaining away the superscripting and letterhead discrepencies, as well as the archaic document formatting, and so would thereby completely destroy the forgery hypothesis.

I'm also now getting a pretty good idea of what the office tech was like back then. It's a very intriguing, almost completely forgotten about era. The word processing market evidently took off like crazy in the very early 70's with lots of startup companies with very innovative products, and reached its peak about 1975-1976. But by the late 70's, it had quickly collapsed into a handful of large vendors, like IBM and Wang. There was probably also pressure from the surging PC market at that point from the lower end. I do remember a reference from a while back about how the CP/M-based word processing systems, and even the first IBM-PC based ones, were far less sophisticated and capable than the much more expensive (and apparently high maintenance) dedicated word processors that they replaced. The latter Wangs and IBM Displaywriters were used mostly as fancy electronic typewriters that could store full documents and apparently could do little of the typeset-type stuff that products like the who-ever-heard-of-it Redactor II could do.

Again, research, research.... I might have a cute, fun thing for you typographers to look at later. -BC 209.6.203.244 19:49, 5 October 2006 (UTC)

Obviously, you still don't get it. Thre's no evidence that daisywheel printers were commercially available in 1972, much less in common use, even with 1/60" resolution. Here's a 1975 brochure on an early Wang word processor: http://www.wang2200.org/docs/Wang%20Word%20Processing.pdf And here's a 1974 product guide for a Wang office computer that supported word processing: http://www.wang2200.org/docs/System%202200%20General%20Product%20Guide.pdf Both systems used Selectrics for letter-quality output at that time, although Wang later introduced a daisywheel printer with 1/60" horizontal positioning. http://www.wang2200.org/docs/2281WC%20Printer-Plotter%20Data%20Sheet.pdf - 71.212.31.95 21:27, 5 October 2006 (UTC)

Ummmm....I had already linked a 1972 CPT 4200 [Fixed typo -BC 209.6.203.244]to a Diablo ribbon and showed how a *later* model, a CPT Cassetype 4200 from apparently 1978, used a Selectric mechanism. And Wang wasn't even mentioned as a player in the word processing market in that 1975 Business Week article covering the word processing business then: 1975 Business Week Article And as I said, and as counter-intuitive as it might seem, the earlier word processing systems by the likes of Redactron evidently had much more sophisticated document creation functions than later devices like the Wang. Look at the description of the Redactor II, a machine that came out about 1976, here: Redactor II Description and compare it to what your Wang can do. Again as I said, later word processers like the Wang and the IBM Displaywriter functioned more like big electronic typewriters with full page storage. By contrast, the early word processors looked more to be meant as do-all composition machines.

But this is all surmising and such. What we really need is actual data and specs on those mysterious early 70's daisywheels. Which I'm working on (which means I should spend less times here.) Well, I'm glad at least someone else here is making use of the Internet for research.

What if I do, and I'm certain I will, find a circa 1972 word processor using a Diablo model with 1/120" horizontal resolution -- do I win then or will I still have to do more work? -BC 209.6.203.244 22:18, 5 October 2006 (UTC)

You are hopelessly confused. 71.212.31.95 23:06, 5 October 2006 (UTC)

http://www.bookrags.com/Word_processor http://users.ipfw.edu/jehle/deisenbe/compartics/History_of_Word_Processing.pdf

From the "Bookrags" cite: "In the early 1970s Linolex, Lexitron and Vydec introduced pioneering word-processing systems with CRT display editing."

That "History of Word Processing" cite is just dumb: it discusses IBM stuff and then skips over to microcomputers and CP/M. Not very germane to the discussion, is it? (Not very good history, either.)

Since we're citing stuff, here's another one from me: http://blogs.msdn.com/murrays/default.aspx Around 1978 I got a Diablo daisy-wheel printer to go with my IMSAI Z80 microcomputer. Not only was it much faster than the Selectric, it had many daisies some of which were proportionally spaced and it was designed to work as a computer printer. I had gotten into microcomputing thinking that by computerizing my house I'd learn something about experimental physics, since a real physicist surely needs to know something about both experiment and theory. To handle the proportional spacing, I wrote a printer driver. My colleague Rick Shoemaker, another microcomputer addict, and I decided we'd write a book called Interfacing Microcomputers to the Real World, and we "typeset" it using my printer driver and a daisy wheel printer. Addison-Wesley published the book, just as it had published Laser Physics, but this time using our nice proportionally spaced camera-ready proofs.

A library near me still had that book, so I popped on over to take a look, he used a Diablo 1345A, which is a HyType II -- not bad, eh?

I now suspect HyType I's are 12xx models, while HyType II's are 13xx. The first model HyType II would then likely be the 1300. In between that and the 1978 1345A used to typeset the book and according to that nifty Swiss ribbon crossreference, there was the Diablo 1300, 1330, 1330 WP, 1334, 1334 WP, 1335, 1335 WP, 1340A, and the 1340 WP. Hmmm....

I may be many things, but confused is not of them (well, at least when it comes to science/geeky/logic stuff...) -BC 209.6.203.244 03:02, 6 October 2006 (UTC)

I photocopied some pages out of that Sargent/Shoemaker book, scanned two of them into PNG/s and posted them here: http://www.aheckofa.com/FoolMeOnce/Murray.html FYI. -BC 209.6.203.244 13:33, 6 October 2006 (UTC)

Your reference pertains to a Diablo 1345A printer from "around 1978". One of the book pages you scanned indicates that the printer had a built-in microprocessor. You don't seem to understand that microcomputer-related technology advanced rapidly during the 1970s, especially after microprocessors became available. You cannot infer the capabilities commercially available in 1972 from a device dated to 1978. 71.212.31.95 17:44, 6 October 2006 (UTC)

What I'm essentially trying to do is move backwards in time towards 1972 by using provable checkpoints in regards to what common office technology was capable in an given year. The Swiss ribbon crossreference indicates there were daisywheel printers in use back in 1972, but without much indication of specific model and capability. That 1977 description of the Redactron Redactor II made it sound very much like it could have created the memos, and that 1973 reference to it suggests it used technology available at least very close to the time of the memos, but that's not exactly hard proof either (although more info is forthcoming on that). Having a book typeset on a Diablo in 1978, with not only the specific model number of the Diablo used, but with actual pages to look at -- c'mon now, that wasn't such a bad catch for a troll, no? Did you catch the clean superscript? That was totally by accident -- I went skimming through the book (it was late, I had a hard time finding a parking spot, and it was an archived book that I had to wait to be retrieved) and I looked first for any Diablo references, and then for any superscripting. I didn't notice any superscripting and was getting antsy to leave, so I just made photocopies of the Diablo references. It wasn't until later when I was looking it over when I saw the little "tm" superscript. Cool.

What this does is bolster the Redactor II as being capable of doing the memos since that lawyer who describes it bought it around 1977, just one year before Sargent said he got his. And the Redactor II is linked to 1973 in some fashion at least. Also bear in mind that his IMSAI/Diablo combo was likely a much less powerful device, at least in regards to word processing, than the Redactron. That "Scroll Systems Retroscroller (tm) Text Output" controller sounds especially very intriguing, but there is no reference to it in Google. Was that a new gizmo then or an old one? All that lost info.... -BC 70.90.98.226 19:52, 6 October 2006 (UTC)

As I understood it, what you're trying to do is show that the memos could have been produced in 1972 using a word processor with a daisywheel printer. Thus far you have failed to offer any valid evidence to support that claim. You say that the Redactor II "is linked to 1973 in some fashion at least". Beyond the fact that 1973 is too late to support your claims, your "link" to 1973 is a specification document which according to CBI shows that the Redactor II was then not yet in production. This supports the 1977 Business Week article, which indicates that the Redactor II was not delivered until 1976-7. Moreover, that original version cited by Business Week could not be the one described in your "1977" reference, because it used magnetic tapes rather than floppy discs. Yet you continue to refer to this Redactor II as if it had some relevance to the issue of 1972 daisywheel printing technology. It doesn't. 71.212.31.95 18:38, 7 October 2006 (UTC)

Ah, but as I pointed out, 1972 is now within striking distance -- devices capable of replicating the memos were available around 1976 at least, and these were not at all, as had been claimed by the likes of Newcomer and Phinney, vanishingly improbable and/or ultra expensive typesetting machines. They were indeed actually regular office stuff that even lawyers in small firms were buying back then. Even though there is a lot more research and investigating to be done, and even though there is frustratingly hardly any records left of those fascinating early-mid 70's word processing systems, the bits and pieces of evidence that has popped up so far have all been undermining the fundamental assumptions and beliefs of the pro-forgery people, expert or not. And the vast bulk of the external evidence aside from the appearance issue very much points towards to the memos being real and created by some circa 1972 device.

You continue to make claims without evidence. You have offered no evidence that any office products capable of producing the memos were available in 1976. Much less 1972. You have not in any way undermined the expert conclusion that the memos are forgeries. 71.212.31.95 18:05, 8 October 2006 (UTC)

While I've been pretty good at going public here rather quickly with stuff I find out about, there were a few things I've kept to myself, mostly because I need to know what they mean. For instance, I was informed that the 1973 document on the Redactor II compares that to three other competing word processing systems, presumably on or about to come on the market at that point, and each used either a Diablo or Qume daisywheel printer: NBI, OCI-Veritext, and the AB Dick Magna I. I'm not too sure to make of this tantalyzing info so far (some people at CBI are trying to get more detail, though). The "NBI" reference is too vague. A Google on "AB Dick Magna I" gets pieces of a pay-to-read reference that starts off as, This unit was an AB Dick Magna I automatic "daisywheel" typewriter that is compatible with IBM Mag Card II equipment. We equipped the Magna I with an... And all I can find out about the "OCI-Veritext" so far is that it used a HyType II(!)/Qume ribbon. Very tantalyzing, but some more detail is absolutely needed. There's a good chance, though, I can push the memo-capable tech back to 1973 within a week. And then....well, we'll see, shan't we? -BC 209.6.203.244 23:26, 7 October 2006 (UTC)

I think you overestimate your chances. But I encourage you to continue your research, because I think the early history of word processing technology is very interesting, and not well documented. 71.212.31.95 18:05, 8 October 2006 (UTC)

Just before heading out the door, I found a passing reference to the OCI-Veritext, but while it doesn't say much about it other than to make it seem more like a smart typewriter, the article it appears in gives some more insights into early word processing tech: 1984 Creative Computing Article -BC 209.6.203.244 00:30, 8 October 2006 (UTC)


The second indicates that at least some of the Diablo models from the 70's had a HyPlot *option" which doubled the resolution to 1/120" http://gopher.quux.org:70/Archives/usenet-a-news/FA.printers/81.08.11_ucbvax.2620_fa.printers.txt

This is from 1981. It has no relevance to the capabilities of products commercially available in 1972. 71.212.31.95 20:27, 1 October 2006 (UTC)

The basic question it whether just 1/60" resolution for proportional spacing would be sufficient to dupe the memos at least as well as the Word recreations. Bear in mind that the Word recreations only approxtimate the memos and are not at all identical, most noticeably with the two Killian memos with letterheads. I know that some of the typography experts think a higher resolution would be needed, but even an appoximate match by circa early 70's device should be enough to severely undercut the forgery claim since there really isn't anything else supporting it in a verifiable way. -BC 209.6.203.244 18:23, 1 October 2006 (UTC)

And the basic answer is no, as already explained. You just don't understand the technical issue. 71.212.31.95 20:27, 1 October 2006 (UTC)

The fundamental problem is that are dealing with technology that was obsolete and scrapped so many years ago that very, very little evidence is remaining for even their ever existing, nevermind nice think manuals filled conveniently with specs and codes. As I think I demonstrated enough, the "created by Word" scenario has too many easily demonstrable discrepencies and issues to be taken seriously, especially so when the contents issue is factored in.

The fundamental problem is that your claims concerning the technology available in 1972 are based on uninformed speculation unsupported by any evidence. 71.212.31.95 16:07, 1 October 2006 (UTC)

Of course life would be much easier if Bush had simply fessed up to whether the memos were true or not. His not commenting is more support for the memos being genuine.

There is one more test I'm looking at that looks promising: PC Magazine use to have (or maybe they still do) annual printer issues where they tested and rated different printers, and they included print samples. I can get my hands on some early-mid-80's issues that had daisywheel tests that had samples of proportional print mode in addition to Pica and Elite. Since Pica and Elite are fixed pitch, they can be used as a reference for the type of proportional spacing used in the daisywheels, most of which were Diablo compatibles and had identical proportional print modes. If the proportional mode of the Diablo-compatible daisywheel matches the spacing shown in the memos, would that be sufficient to at least kill the forgery charges? It may not completely clear up the origin of the memos, but it would be nice to finally put the always very, very dubious forgery claims to rest. -BC 209.6.203.244 05:23, 1 October 2006 (UTC)

It is very doubtful that printer evaluations from PC Magazine in the 1980s would have any relevance to this issue. 71.212.31.95 16:07, 1 October 2006 (UTC)

Oh yeah, I came across this, but I'm not too sure what to think of it yet: https://secure.fixyourownprinter.com/?search=cr&q=HyType But it looks as though the HyType I was indeed the low-end model since the Coleco Adam used it and the Coleco wasn't made until 1983: http://oldcomputers.net/adam.html So I may have been mistaken in thinking that HyType I was the early version of the HyType II -- they may instead have been contemporaneous models, with one being cheap and the other deluxe. My bad. -BC 209.6.203.244 05:41, 1 October 2006 (UTC)

You're now claiming that the HyType II was available in 1972? And that the HyType II was capable of producing the proportional spacing exhibited in the memos? What evidence do you have to support this claim? None, of course. 71.212.31.95 16:07, 1 October 2006 (UTC)

Nooo....if you have been paying attention, what I've done is establish via a printer ribbon cross-reference that there were indeed word processing systems in 1972 that used HyType I daisywheel ribbons. So evidently there were word processing systems using Diablo-manufactured daisywheel printers at the time of the memos. What's murky now is what models they were and what exactly were they capable of. Like I said, at that time Diablo was strictlyan OEM printer supplier to word processor makers like CPT and Redactron, and so any documentation would have been provided by them and would refer to the printer by their model numbers and very likely not Diablo's (think Toyota Matrix versus the Pontiac Vibe, which is also a Matrix). There is some documentation after Xerox took over regarding the HyType I Model 1200, and I had been thinking that the Model 1300 HyType II replaced it a couple of years later, but apparently now the HyType I may have been around in one form or the other until the early 80's. So this to be looked further into. The important thing is that we now have word processing systems in place with some model of Diablo daisywheel printer at the time of the memos. That's something, no? -BC 209.6.203.244 18:23, 1 October 2006 (UTC)

No. As noted above, the HyType I mechanism could not have produced the proportional spacing in the memos. And what evidence do you have for the claim that there were word processing systems available in 1972 that used HyType 1 ribbons? 71.212.31.95 20:27, 1 October 2006 (UTC)
"...daisywheel tests that had samples of proportional print mode in addition to Pica and Elite..." This makes the assumption that there was only one system of proportional widths available to the HyType I. Although this is possible (the Selectric Composer had this limitation), there is no reason to assume it's true without more evidence. A single sample that doesn't match would not prove anything. However, knowing that the HyType's horizontal advance was in units of 1/600 of an inch does (though I'd still like to know where that manual was found, so I can verify it myself). [edit: Oops, that link was given earlier, sorry.]
"PC Magazine... annual printer issues... If the proportional mode of the Diablo-compatible daisywheel...." The Diablo 630 daisywheel printer was introduced in 1983. The term "Diablo compatible" apparently referred to the Diablo 630 in particular. Although the HyType I was still in use as a super-low-end printer at this time, there isn't much reason to assume that it had the same level of proportional spacing capability as a model introduced a decade later. (Perhaps it did, but that would certainly need some evidence.)

In other words, some more research needs to be done. Your viewpoint is that of a typographer, but at a computer firm, Adobe Systems, so you are familiar with computer fonts, but not so much with old printers and their capabilities, including type characteristics. By your own admission, you've only recently concluded that the memo fonts are not Times New Roman. Hailey's expertise is in document authentication and the type characteristics of old devices, apparently primarily typewriters. I make no claims to expertise (but nor do I claim I have none), but I have looked carefully at the methodologies that you two and others have use and found some flaws.

You don't provide nearly enough samples to convincingly demonstrate your point that the fonts are Times Roman when individual memo characters are evidently not. You seem sincere, but you should have spent more time doing more word-to-word, character-to-character comparisons between the memmos and Wrod recreations and then demonstrated how this supports your conclusion. Your pdf "samples" are grossly insufficient by any scientific research standard.

Hailey has does far more sampling and had demonstrated his results much more extensively than any of the other supposed experts involved in this, so he get more brownie points towards his expertise level. He seems, though, to have no idea about daisywheel printers -- instead he extensively describes a strange typewriter-like device with a platen that can print proportionally in a font that sort of looks like Times New Roman, but isn't (perhaps he should have looked at these: http://www.aheckofa.com/FoolMeOnce/Printwheel3A.jpg and http://www.aheckofa.com/FoolMeOnce/Printwheel2A.jpg.)

And obviously my interest is in debunking this forgery nonsense, and it is nonsense -- you put *all* the pieces on the table -- fonts, contents, DoD docs and so on -- throw out all the easily refuted bogus bits, and then when you put all them all together, forgery is not at all what you end up with. If anything, what you do end up with is far worse. -BC (this really will be my last say on the matter until I have more concrete info to share) 209.6.203.244 16:04, 1 October 2006 (UTC)

"As I think I demonstrated enough, the "created by Word" scenario has too many easily demonstrable discrepencies and issues to be taken seriously, especially so when the contents issue is factored in." I do not think your concerns demonstrate that, nor do most other experts who've been looking at the documents (nor most of the typographers who've brought it up with me in the last couple of days at the annual international typography conference, running in Lisbon right now - and being a random selection of international folks, most of them can't stand Bush any more than I can). I for one take no stand on the truth or falsehood of the contents; not my area of expertise. But the typography (font and formatting) is so incredibly improbable that until/unless a machine is found that could do such a thing, I will continue to consider the memos almost certainly forgeries.

- Thomas Phinney 07:45, 1 October 2006 (UTC)

I suggest you do your homework better. You alleged experts have not helped at all to clear up matters here. You, like Newcomer, have offered a lot of opinion but with very little in the way of demonstration to back it up. You claim now that the memos are in Times Roman font (MAC) and not Times New Roman (Windows), but you haven't demonstrated that with any real sampling either. By comparison, I took all 3 instances of the word "Harris" from the "CYA" memo and compared them to a Word-created one -- the characters don't match up that well. In your interview with Creative Pro, http://www.creativepro.com/story/feature/21939.html, you make the statement:

"So, every letter is different. The baseline wavers a bit. That's no surprise. We're not looking at an original printed directly from a computer. This is an effect of the image degradation from being scanned twice at low resolution. This is why many of the arguments based on letter shape are pretty dubious.

If you want to test it, type a memo in Times New Roman. Fax it to yourself (make sure the fax guides are loose enough that the paper can twist just ever so slightly). Get it scanned back into a computer, and then drop the resolution to 120 dpi on the final image. It will look a heck of a lot like the CBS memo at this point. Absolutely no need to use a typewriter (or a degraded font) up front!"

But as I pointed out to you, there was a guy did that with a Word-created CYA copy, http://www.poweroftheindividual.org/blog/2004/09/fauhxed-bush-memos.html, and I again took out his 3 instances of "Harris" and compared those to Word-created one. While the three Harris's are much more distorted than the CYA memo ones, they still match up better to their Word origin than the CYA ones. Also in regards to the baseline wavering, the the deliberately distored Word-created one wavers in groups of letters, and *not* individually as show in the CYA memo, which would be the case with an impact printer. And I made the effort to actually demonstrate this: http://www.aheckofa.com/FoolMeOnce/CBSBushMemos.html#FunWithFonts.

Have fun in Lisbon. ;) -BC 209.6.203.244 21:17, 1 October 2006 (UTC)


Veritas vincit. -BC 209.6.203.244 18:42, 29 September 2006 (UTC)

What's the evidence for your claim that a daisywheel printer with proportional spacing was available to Killian in 1972? 71.212.31.95 05:14, 30 September 2006 (UTC)

That was already covered: this is a Swiss HyType ribbon crossreference listing the CPT model 4200: http://onlineshop.jpd.ch/sp/farbband-multistrike/qume-1145-farbband-multistrike-original-schwarz.asp

This is a computer timeline mentioning the CPT model 4200 word processor as being introduced in 1972: http://www.computermuseum.li/Testpage/01HISTORYCD-Chrono1.htm (That's only one example, but I had referenced a Business Week article from June, 1975 showing that the word processor market was in full bloom by then.) As far as whether this was available to Killian, that's like asking if a 1972 Chevelle was available to him -- that's a pointless question. The thing that started the forgery charge in the first place was the claim by "Buckhead" that there were no such devices in 1972 that could proportionally print, and then Charles Johnson did his little, cute, and highly misleading CYA/Word animation and so "Rathergate" was born, and truth and journalism took a holiday. All the other claims of evidence for forgery have turned out to be utter nonsense, so it's really back to figuring out what device printed out the memos. If the best evidence is that is was indeed a daisywheel, then not only is the forgery claim dead and buried, but it also demonstrates the risk of relying on the very sketchy blogosphere for real information, how deplorably lazy and sloppy the mainstream media has become, and what sort of a man George Bush is, since he could have put this to rest a long time ago, sparing the long, drawn-out, and often vicious smear campaign aimed at Rather, CBS and Mapes, if he had any integrity whatsoever. -BC 209.6.203.244 21:41, 1 October 2006 (UTC)

The original CPT 4200 word processor used a modified Selectric typewriter: http://www.compmuseum.org/index.php?title=CPT_Cassetype_4200 It could not possibly have produced the typography exhibited by the Killian memos. You really have no idea what you're talking about. Your claims cannot be taken seriously. 71.212.31.95 22:34, 1 October 2006 (UTC)

Sorry, but I've produced a ribbon cross-reference showing the CPT 4200 used a HyType ribbon, which does not fit a Selectric -- maybe CPT offered each as an option. We don't know for sure because the bulk of this stuff, including what would have been very helpful manuals, has been long ago obsolete and trashed. As I've mentioned before, it's very murky about what the technology was like back then. All that's known for sure is that the word processing market was in full bloom by 1975 with many venders now long since gone; Diablo was making daisywheel printers from about 1969 and they had to have been selling enough of them to get Xerox to buy the company for $28 million in 1972; there more than a few problems with the "created by Word" scenario; and that the more you look into it, the more the memos don't look like they were forged. -BC 209.6.203.244 23:00, 1 October 2006 (UTC)

As I said, you really have no idea what you're talking about. (You obviously didn't look at the CPT 4200 documentation at the link I provided.) Phinney has shown conclusively that the memos are forgeries. You just don't understand his (fairly simple) technical analysis, which clearly excludes the daisywheel printing technology available in 1972. 71.212.31.95 01:31, 2 October 2006 (UTC)

Gawd...let me type this really slow so you can understand. The CPT 4200 is referenced, along with an awful lot of long forgotten word processors in this ribbon cross-reference for HyType ribbons: http://onlineshop.jpd.ch/sp/farbband-multistrike/qume-1145-farbband-multistrike-original-schwarz.asp Go look, it's there. I didn't make this up. Now as far as the CPT 4200 in your link, that now looks to be a different 4200, the model 4200-i. Why? Go here: http://www.tonsoftoner.com/products/cpt/3044128519.htm That uses a Selectric ribbon. So I was right to suggest that the CPT 4200's came with either a Selectric or HyType option, which trashes whatever point you were trying to make. In regards to Mr. Phinney, I can only point out that, unlike Hailey, Phinney has shown very little in the way of demonstrable evidence to support his claims, the current one being that the font shown in the memos is Times Roman (which is what Macs use) and not Times New Roman (which is what Windows uses). It isn't enough to say something like, "I'm an expert, these are my credentials, therefore you must believe this opinion of mine" without going into at least some verifiable and demonstrable detail at how that opinion was reached.

The CPT 4200 described in the link I provided is the original CPT 4200. The one that was introduced in 1972. The one that was referenced in the chronology you cited. That's the only model that's relevant to the issue. Your ribbon cross-reference website (hardly an authoritative source) may simply be in error, or it may refer to a different machine with a similar name of later date. Since it gives no date, it does not support your claim that a word processing system using the HyType mechanism was available in 1972. With regard to Mr. Phinney, see below. 71.212.31.95 15:19, 2 October 2006 (UTC)

You might want to consider using this rather useful Internet search site called "Google.com". Perhaps you may have heard of it. Let's say you were curious about, oh say, the ribbon an old CPT mode 4200-i word processor used. You think it might use a Selectric ribbon, but you're not sure. To find out, all you need to do is do a Google search on: CPT 4200-i Selectric and you will get results like http://onlinecomputersareus.stores.yahoo.net/ibmseiirit3.html, http://www.intimecatalog.com/supplies/details/RIBBON_DATAPRODUCTS_DATA_PRODUCTS_R5180.phtml?camp=Froogle&subcamp=DATAPRODUCTSDATA_PRODUCTS_R5180 and so on.

Now that you've confirmed that it was indeed the CPT Model 4200-i that used a Selectric ribbon, the next question may well be, what came first: a model called simply the 4200 (which used a different ribbon type), or one called the 4200-i. Hmmm.... -BC 209.6.203.244 19:55, 2 October 2006 (UTC)

Well, let's see. Since the museum webpage about the Selectric-based CPT 4200 that I gave you the link to above says: "It was the first product manufactured by the CPT Corporation", I guess that answers your question. Evidently it was too much trouble for you to look at that page and read as far as the second sentence. Yeah, Google is great, but you have to actually read the references it turns up. 71.212.31.95 21:05, 2 October 2006 (UTC)

And maybe you should have read the date on the last page of the "Training Manual" for that CPT model: http://compmuseum.org/upload/cpt%204200%20series%20manual.pdf Sorry, but ribbon manufacturing cross-references are very, very specific on models, and all of them have a HyType ribbon listed for the plain CPT 4600, and a Selectric ribbon for the CPT 4600-i.

Actually, I did read the date (2/78). What do you think it proves? That's presumably the date on which the manual was printed. It may well be later than the date on which the device was introduced. Or perhaps your chronology was wrong about the 1972 date. Remember, you're the one claiming that a word processor with proportional spacing was available in 1972, not me. You've provided no evidence that a CPT machine using a HyType ribbon was available in 1972, or that any such machine supported proportional spacing. But please keep trying. 71.212.31.95 01:33, 3 October 2006 (UTC)

If you don't like my CPT reference, you're going to hate my Redactron one: http://eet.com/special/special_issues/millennium/milestones/berezin.html According to this, Redactron started selling its first word proccesser in 1971. And According to that June, 1975 Business Week article I referenced earlier, Redactron had at that time sold its 10,000th word processing system. And that same Swiss HyType ribbon reference that has the CPT models also has *12* Redactron models listed.

My only problem with your Redactron reference is that, like your CPT one, it provides no support for your claim that a word processor with proportional spacing was available in 1972. This article says that Redactron's word processor started shipping in 1971. It goes on to say "Later machines, which included monitors, could do most of the important things—like arranging columns and margins and moving copy around—that were routine in later-day word processors." Please note the use of the word "later". And the article makes no reference whatever to proportional spacing. 71.212.31.95 01:33, 3 October 2006 (UTC)

But this is all still indirect evidence. I think I may have a lead on something slightly more direct, but we'll see. What's more certain is that these always foolish forgery claims are finally going to meet a way, WAY overdue, ignoble ending. -BC 209.6.203.244 22:12, 2 October 2006 (UTC)

When you can offer some real evidence for your claim that a word processor which could have produced the Killian memos was available in 1972, I'll be very interested to see it. 71.212.31.95 01:33, 3 October 2006 (UTC)

Some research in Redactron led to this page: http://www.cbi.umn.edu/collections/inv/burros/cbi00090-077.html Apparently the company folded up in 1977 after having been purchased by Borroughs a couple of years before then. There is a reference to the "Redactor II" word processor dated 1973, 8 lines down from the "Company Records" section. That wonderful Swiss ribbon crossreference site has the "Redactor II" using a HyType ribbon. 1973 is not 1972, but it's firmer evidence for there being word processing systems using daisywheel printers. By the way, you and others lurking out there my find it interesting to go to the search page on that Swiss ribbon site, http://onlineshop.jpd.ch/suchen.asp, and in the "Druckermodell" search field, put in "Diablo" then click on "Suchen". You should end up with the most comprehensive list of Diablo printer models you'll ever see. Since the Hytype I was also the model 1200 and the HyType II was also the model 1300, that list implies that they were contemporaneous models along with many variations. And logically, if you had a single initial product, you wouldn't put an "I" after it right away at least. In Diablo's case, it would be just the "HyType" if that was its first daisywheel printer. That there were a HyType I and II, and apparently existing at the same time, most likely means that the HyType II was a beefier, more expensive model, and not a sucessor.

But that's just guess work. Getting back to the "Redactor II" model, I also found this http://216.218.211.127/~forums/showthread.php?t=65765&page=13 To quote: "Burroughs bought out Redactor corporation (or at least its computer line from them), and sold a dedicated word processor with a large full letter size screen (portrait, not landscape) with dual floppy disk drives (RAM on a disc instead of linear on tape - HUGE time-saving improvement), for about $13,000. They gave me and another attorney a $3,000 discount if we would turn in a Redactor I so they could literally junk it at the city dump - they did not want to service them even for a $500 per year service contract. My buddy had one, and so we purchased a Redactor II (R-2) (which we promptly renamed R2-D2, of StarWars fame) for a mere $10,000. It used a Qume daisy wheel printer which had blinding speed that blew away the IBM Selectric, and we could attach different wheels for different fonts and sizes ranging from 10 pica to 12 elite to 15 fine print (great for attorneys ), Courier, Letter Gothic, and even "proportional spacing" Times (OMG!) print. It was almost like owning a print shop and having a sophisticated type composer machine. We could do anything...except graphics."

Now, you may note that the "Star Wars" reference puts the purchase at 1977, which also was apparently the last year that any Redactron model was sold. So the Redactor II apparently came out in 1973 with a Qume daisywheel printer (remember Qume was formed in 1973), was sold until 1977, and there is a reference to this system being able to proportionally print in a "Times" font, which is probably either what the Redactron manual said it was or else what the printhead was labeled as.

Your CBI reference says Redactron was acquired by Burroughs in 1976. Business Week (11/21/77) says: "Being acquired by Burroughs has not helped Redactron all that much either. 'You can hear the fizzle,' says the president of one competitor. Despite the Detroit computer maker's financial resources and years of related experience, Redactron delivered its first Redactor II, a display-screen text editor, a full year after the product was announced. It took longer than originally expected, explains President Berezin, to develop the new system." This indicates that the Redactor II was first delivered by Redactron after the acquisition by Burroughs, sometime in 1976-7, not in 1973 as you claim. 71.212.31.95 01:53, 4 October 2006 (UTC)

You go back to here, http://www.cbi.umn.edu/collections/inv/burros/cbi00090-077.html, and there is this reference: "Redactor II & Printer Specifications, Book 3," 1973. (Box 3, folder 13-14)

That's 1973, and not 1976-77. 1977 is apparently when Redactron went out of business. Also bear in mind that these were not PC's with model changeovers every several months or so -- something that came out in 1973 would be sold for the few years at least, with maybe some incremental enhancements every year or so.

While this doesn't place the system at 1972, it's awfully darn close and says a lot about the office tech around that time, and that "Times" font reference is at least intriguing, no? -BC 209.6.203.244 20:16, 3 October 2006 (UTC)

No. A reference to "proportional spacing" Times in 1977 is meaningless. Here's a pertinent extract from Desktop Publishing Skills (Felici and Nace, Addison-Wesley, 1987):
"Typewriter type and most impact printer type is handicapped by its monospacing -- the letters are designed so that each one takes up an equal amount of horizontal space on the page... The first efforts in the word processing world to reduce this problem resulted in proportional-spacing impact printers. These printers divide the letters of the alphabet into a small number of groups according to their widths. A typical proportional-spacing scheme may provide three units of escapement for every one on a monospaced machine. A wide character like an M would be accorded a width of three units, a narrower e would be accorded two, and an i would be measured as one unit wide." (pp. 22-24)
If you think the crude kind of "proportional spacing" provided by these early impact printers could have produced the spacing exhibited by the memos, it's simply a further demonstration that you don't know what you're talking about. 71.212.31.95 21:31, 3 October 2006 (UTC)

Sorry, but a typewriter reference (it sounds like it's talking about the IBM Executive) has absolutely no bearing in a discussion of daisywheel printers. And that 1977 cite was to a particular word processing system evidently introduced in 1973, and the Swiss ribbon crossreference confirms that it did indeed used a daisywheel printer, probably a Qume (the first Qumes at least were meant to be completely plug-compatible with Diablo's.) And as far as "crude" goes, the daisywheel printers then were precision devices, as I had already provided reference for via this pdf file, http://www.tinaja.com/glib/atg1.pdf, beginning on page 14. The only real issue is the question about what the horizontal resolution is for the different models at that time, which can be either 1/120" or 1/60". Some of the "em"-ty headed typographers (sorry...) have claimed that 1/60" is not enough for Times Roman, but we're talking about the memos and their coarse resolution may make this a moot point.

In any case, I guess I should mention that the good people here, http://www.cbi.umn.edu, are apparently going to look for references to daisywheel printers in their considerable collection of Redactron documents they conveniently have archived. I may be a troll, but I am a thorough one. ;) -BC 209.6.203.244 22:54, 3 October 2006 (UTC)

Where did you get the idea that the statements I quoted about proportional-spacing impact printers developed for the word processing world are a typewriter reference? And how could it be talking about the IBM Executive, which long predated word processing and employed a 5-unit system of character widths? And why do you think the reference you provided, which refers to the capabilities of an enhanced version of the Diablo 630 from the 1980s, is relevant to the question of the technology available in 1972? 71.212.31.95 01:28, 4 October 2006 (UTC)

???? You don't seem to be following the thread very well. Grant it, it's messy, so let me summarize: 1) Daisywheel printers existed from about 1969 and were certainly common enough by 1972; 2) All references indicate that those printers could proportionally print, either via a simple "proportional mode" command (Esc P) or through computer control of the characters mapping and spacing, with the horizontal spacing resolution of these printers was either 1/60" or 1/120" depending on the model (it's murky so far which models outside of the HyType I and a handful of latter models could do which); 3) Qume, a plug-compatible Diablo rival, began making its daisywheel printers around 1973; 4) around that time there were also computer-like devices called "dedicated word processors" that used both daisywheel printers and modified IBM Selectric printers, and the market for these, according to a Business Week article from that year, was flourishing by 1975 with many vendors, manufacturers and products, most of which long since gone and forgotten about; 5) Both Diablo and evidently Qume were primarily OEM suppliers at that time, so the references to the printers they made are mostly masked by the model numbers of the dedicated word processor manufacturers; 6) However, by using a *very* comprehensive ribbon crossreferences like this, http://onlineshop.jpd.ch/suchen.asp, one can link daisywheel printer ribbons to specific word processor models; 7) By Googling those word processor models using daisywheel printer ribbons, one can some dated references to those models, hence establishing a least a rough timeline despite the paucity of info even via Google since we are dealing with long, long obsolete and discarded technology; 8) There is a CPT word processor, the mode "4200" that evidently came out in 1972 and it's only ribbon reference is for a Diablo ribbon. There was another, presumably latter model, the "4200-i" listed as using a Selectric ribbon. This was objected to via a reference to a computer museum claiming that a "CPT Cassetype 4200" in their posession, which used a Selectric printer, was the first CPT word processor. However the training manual for this model, which is on their site in PDF format, is clearly dated as being printed in February, 1978, which is several years after CPT began making word processors. There is a Business Week article from 1977 that states some of the word processor vendors were "trying to sell compatible products into the giant IBM product base", which would suggests, given the date on the training manual, that the "Cassetype 4200" was only meant as a lower lost alternative to IBM's Mag Card Selectric; 9) There was a word processing company called "Redactron" that was started in 1969, was a market leader in 1975 when it sold its 10,000th word processor, was bought by Burroughs in 1976, and then apparently was folded in 1977. There are daisywheel ribbon references to at least 12 Redactron models, one of which is the "Redactor II". There are two other key references to the "Redactor II" found on the web -- one in an archived Redactron document dated 1973, and another from 1977 stating rather specifically that "It used a Qume daisy wheel printer which had blinding speed that blew away the IBM Selectric, and we could attach different wheels for different fonts and sizes ranging from 10 pica to 12 elite to 15 fine print (great for attorneys ), Courier, Letter Gothic, and even "proportional spacing" Times (OMG!) print. It was almost like owning a print shop and having a sophisticated type composer machine. We could do anything...except graphics."; 10) I asked the research librarians at the place where the Redactron files are archived, http://www.cbi.umn.edu, if they could look for dates and references in regards to daisywheel printers in the 1969-1974 time period, and I got a reply that they will.

And that's the status so far. Now are you up to speed? Also, have you ever thought about maybe being part of the solution? -BC 209.6.203.244 13:33, 4 October 2006 (UTC)


You tell me which of these shows more effort and research: http://imrl.usu.edu/bush_memo_study/supporting_material/bush_memos.pdf or http://blogs.adobe.com/typblography/bush/TNR%20vs%20TR.pdf Hailey makes a very strong case for the memos having been created on some sort of impact printer and he concluded that the "Shape and proportion of key characters (e.g., "F,""L," "g," "5") do not fit Times New Roman or any other digital typeface I have yet found."

Hailey is not an expert. Phinney is. Unfortunately, you're unable to tell the difference. 71.212.31.95 15:19, 2 October 2006 (UTC)

Really? I was under the impression that they were both experts, one in computer typography and the other in document authentication. No matter -- in science, the guy with the more extensive research and presentation of work and evidence wins. I Googled Phinney's "work" and this is the best I could find: http://www.creativepro.com/printerfriendly/story/21939.html That's hardly at all comparable in quality and depth to Hailey's work. Instead of spinning your wheels with easily dismissable, unsupported opinion, why don't you try some real science instead, like say confirming (or maybe even refuting) some of the stuff in Hailey's report. He gives a nice recipe on page 4 on to show some serious issues with trying to recreate the August 1, 1972 memo in Word. If you have a scanner, it should be easy (print on mylar sheets if you're no comfortable with Photoshop/Paintshop type programs. You do something like that and then you'll have something finally useful to contribute to this discussion. -BC 209.6.203.244 20:31, 2 October 2006 (UTC)

You may try to argue that this one expert's opinion versus another, but in science, homework counts, especially in showing how you reached your conclusions. Hailey shows his homework in detail, whereas Phinney has not. -BC 66.251.53.50 12:07, 2 October 2006 (UTC)

Phinney has explained his analysis quite clearly. The problem is that you are apparently unable to understand it. That's not Phinney's fault. You just aren't sufficiently well informed about typography and printing technology to understand what he's saying. For example, when Phinney refers to the 18 units-per-em system used by Times Roman, do you have any idea what he's talking about? Or does that just go right over your head? 71.212.31.95 15:19, 2 October 2006 (UTC)
Whoosh! 71.212.31.95 00:38, 3 October 2006 (UTC)

Oooo...I'm hurt. I don't pretend to be anything other than a troll, but let me put on this here thinkin' cap I have handy and turn its big ol' knob to 11.... Oh, yeah, when Phinney talks about "18 units-per-em system used by Times Roman", he's talking about digital typefaces that have nothing to do with the memos. The memos are of poor quality, meaning that their inherent coarseness makes such fine unit measurements laughably inappropriate. Also digital typefaces by their very nature are just digitized approximations of traditional typefaces -- they are to real type what MP3 encoding is to analog music. If he or anyone else wanted to really wanted to determine the true typeface used in the memos, a bunch of "A's" should be taken from all the memos and their characteristics averaged out, and then the same done for "B's," "C's," and so on. And then you compare these the defined characteristics of known/standard typefaces. I don't believe he's done anything like that, has he? But Hailey has, so again that gives him more credibility as a researcher and expert. Just pointing out "gosh, that's a narrow "E" -- it just gotta be Times Roman" is *not* sufficient. Sorry.

My head hurts now so I have to take my thinkin' cap off and maybe watch Fox News or somethin' like it to compensate. -BC 209.6.203.244 01:18, 3 October 2006 (UTC)

Sorry about the "Whoosh!". It was only meant to encourage you to respond to the question. But your response confirms my perception that you simply do not understand the technical basis of Phinney's analysis. If all you know (or think you know, and all you seemingly care to know) is that "he's talking about digital typefaces that have nothing to do with the memos", you are obviously in no position to objectively evaluate his analysis. The units-per-em character spacing is not an absolute measure but a relative one. It pertains to the positioning of the characters, not their appearance. And it applies not simply to a single character, but to an entire line of characters. So fine measurements are not involved. And, most importantly, the problem with the memos is not that they fail to match typeset-quality spacing, but that they match closely. This could not have been achieved with 1972 typewriter technology, nor could it be the result of degradation caused by copying and scanning. It therefore precludes the possibility that the memos are genuine. This is not really very difficult to understand. 71.212.31.95 03:03, 3 October 2006 (UTC)

Nooo....I had come up to speed enough on typography and type design topics way back (I do my homework). And I actually had a feeling that not mentioning spacing in my response would provoke some attempt at a gotcha, but it was not worth the revision. If the memos were created with an impact printer -- a daisywheel -- using a "Roman style" font and Times Roman spacing -- which appears to be awful close to the default spacing of "proportional mode" in Diablo-compatible printers -- then you would need a certain resolultion to say with genuine certainty that these letters and their spacing are without a doubt this particular font. Now if your only experience, as appears to be the case with Phinney, is with digitized fonts and the fine differences between patented fonts like Times Roman versus Times New Roman, and maybe how best to accurately display them and print them out on with current technology, he would be completely out his area of expertise when presented with documents of uneven, coarse resolution using Times Roman-*like* spacing and fonts. His natural instinct would be to look for a best fit for things he knows about -- like digitized Times Roman and Times New Roman fonts. But both of those fonts are based on old typefaces that were characterized and created very differently way back when, including during the 70's. And when you also stir in all these Roman-like fonts that were evidently common at the time, then you really need to do some research and show your work to justify your opinion and conclusion in the face of reasonable skepticism. Neither Phinney, Newcomer, or any other self-proclaimed typography experts claiming forgery has so far come even remotely close to what would be considered a sufficiently detailed scientific examination justifying their claims. It's so far been just one little spotty bit here, another spotty bit there, and lots of confused logic. That's not exactly enough. I personally pulled out repeating words from the memos to look for common characteristics and then compared them to Word recreations and found the same consistent discrepancies that Hailey did. What I did was hardly comprehensive, but it was still far more than what I've seen done by any of the so called typography experts.

As a matter of fact, I owned a daisywheel typewriter in the mid 80s. We had daisywheel printers (on dedicated word processors and PCs) in the office I worked at in the late 70s and early 80s. I have copies of the PC Magazine printer review special issues you were so excited about. (They don't help your cause.) So your idea that I am unfamiliar with this technology is completely bogus. 71.212.31.95 21:31, 3 October 2006 (UTC)

Was I talking about you or was I talking about Phinney in regards to old tech? Or are you Phinney with an anonymous IP? And didn't I mention something about justifying claims with some sort of proof? I'm glad you had a daisywheel printer in the mid-80's, worked at an office in the late 70's and early 80's that had dedicated word processors, and have some copies of the printer review issues of PC Magazine. Now, are you going to do something constructive with this info, or....? By the way, did you see my further up posts about the Redactron "Redactor II" word processor? Neat, eh? -BC 209.6.203.244 00:01, 4 October 2006 (UTC)

And in dealing with a real-life issue of document forensics, the font is still only part of the equation -- Phinney made a fundamental mistake in only discussing the font aspects of Hailey's research but not other key issues Hailey noted and discussed, like the poor centering of the letterheads and the evidence for impact printing and wear. In my case I'm looking at *all* of these aspects together with what the official DoD records support and show, what people who should know what the truth is have said (or mostly not said), combined with some research into what exactly was the capabilities of the word processing technology available at the time the memos were allegedly written. And when you do bother to do all this, the forgery claim becomes laughably inane. The absolute worst thing that the memos can be are faithful transcriptions of handwritten notes made within just a few years of the dates appearing on them. Which even then precludes them from being "forgeries" or fakes in any meaningful way.

So instead of you or anyone else glibly saying that I don't understand stuff like FUnits and em squares, you would be a bit more productive and have slightly greater credibility if you spent less time making clueless comments here and more time producing research and results that actually support and back up even a little bit of your claims. -BC 209.6.203.244 13:32, 3 October 2006 (UTC) 209.6.203.244 15:54, 3 October 2006 (UTC)

To be more specific, you claim that daisywheel printers based on technology developed at Diablo Systems were "made since 1969" and were "evidently common by 1972". What is your evidence for that? You also appear to be claiming that these devices were capable of the proportional spacing exhibited in the Killian documents. What is your evidence for that? 71.212.31.95 18:36, 30 September 2006 (UTC)

It's possible that Xerox was interested in Diablo System's disc drive business. --htom 01:03, 2 October 2006 (UTC)
Or maybe not. Xerox claims that the first Diablo printer released was the 1200, in 1973 (scroll way down, find "Diablo".) http://www.xerox.com/go/xrx/template/019d.jsp?view=Factbook&id=MajorProdCat&Xcntry=USA&Xlang=en_US

Not a thing likely to show up in a TexANG office. A Pentagon office, maybe. --htom 06:59, 6 October 2006 (UTC)

Retrieved from "https://en.wikipedia.org/enwiki/w/index.php?title=Talk:Killian_documents_controversy/Archive_7&oldid=80247769"