I'm glad you enjoyed the detail. As I mentioned, I bailed on the company and the project in early 1980, as the software work was being moved to Los Gatos CA, and I had neither an invitation nor the desire to work in California; I just went back “home” to Toronto.
The only thing I still have is a brochure for the DHL1000, but at the moment I don't know where it is. I never kept any hardware. I'm going through a lot of "archives" in my house now, so if/when I come across the brochure I'll scan it and post a copy here.
I did most of development / adaptation of the printing software for the Arabic version (after completely rewriting the standard English version, the original code was pathetic), but had no involvement on the Arabic side of the hardware work, so can't tell you about display issues. AFAIK we just substituted an Arabic keyboard for an English one (USA not UK) so the internal storage would have just been the ASCII it generated.
In Arabic, characters have up to four forms: initial, final, middle (i.e., between two other characters), and standalone. The Daisy printwheel for Arabic (not too sure, I think it was called Naksh) had all the elements to print all characters properly, plus diacriticals, but some printing characters had to be made up of more than one impression plus maybe diacriticals. The first processing of a print line was to create the set of print impressions from the ASCII characters, including up and down platen rotations for subscript/superscript diacriticals. Then, once the line length was determined, I had to take the current position of the print wheel [it was left wherever it ended up after each line was printed], figure out the least travel to start the new line from left to right or vv., reposition the wheel, and if necessary reverse the whole current line and change ups to downs and left travel to right and vv. as required.
I learned the tiniest bit of Arabic as I worked through this. First was a trip to one of the English-language bookstores in Kowloon to get a book (in English) on written Arabic. Then, as I progressed with the code, I'd print sample documents and run them by Nadia, our in-house consultant on Arabic, and let her point out issues. She was Lebanese, could speak and read Arabic but not write it, was never formally taught the language so could not give us formal spelling and grammar rules. Challenges! We occasionally had visitors from the Bahrain office, who also reviewed and provided comments as necessary.
As to Walt and "Waldo" - two different people.
Walt (Walter) Hillblom was Larry Hillblom's uncle, an electrical engineer, and had worked at IBM and at one of the disk drive manufacturers of the day (might have been Memorex, not sure). Larry set up a company in Los Gatos, and hired Walt and a bunch of his cronies to design and prototype the system. In some ways, and just coincidentally, I often knew and understood more about hardware and microprocessors systems than they did. Two glaring design errors were made: 1) there was no disk change interrupt, so when a user swapped floppies, the system didn't know, and kept merrily writing data to sectors as described in the FAT that had been loaded into memory from the previous floppy - you can guess the results…; and 2) there were no bypass capacitors on the power traces to the memory chips, so random memory errors were common due to minor voltage fluctuations and current draw. Both got fixed after a lot of arguing from me (just some know-it-all Canadian kid, in their view).
Alan ("Waldo") Walden was an Aussie who ran DHL's Middle East operations out of Bahrain. I think he got there courtesy his father (Bill? Walden) providing some financial or management consulting to DHL - I never got a clear story, this may not be quite right.
I have no idea what happened to the system(s?) in Bahrain, I suspect they were eventually junked, and I don't recall any being delivered into Saudi Arabia by the time I left. No idea about other locations / countries.
