I just realized the list I send from dux has page numbers I usually read in the first column of words so that doesn't matter if people want it with out page numbers I will remove and resend. It is just something dux adds by default. ken -----Original Message----- From: liblouis-liblouisxml-bounce@xxxxxxxxxxxxx [mailto:liblouis-liblouisxml-bounce@xxxxxxxxxxxxx] On Behalf Of Ken Perry Sent: Wednesday, August 27, 2014 7:07 AM To: liblouis-liblouisxml@xxxxxxxxxxxxx Subject: [liblouis-liblouisxml] Re: List of UEB words Ok I am attaching a list of 99149 words that I created from an old Linux aspell file. The words.txt is the original word list and the words.brf is the converted file from Duxbury UEB. Once my changes are accepted that I have in as a pull request our list matches this list exactly. There are still problems with back translation but at least the forward translation matches what Duxbury puts out. Note I did not just use Duxbury one of the translators here at APH went through the whole list and gave me a list of problems. All of the problems were fixed in only 3 changes to the old UEB table. It would be good to use these two lists as some kind of auto check but I am not sure how to best add it. What I do currently is run the list through lou_translate to forward translate the words.txt and compare it to words.brf. Then I run it through lou_translate to back translate what I forward translate then compare the result to the words.txt to see if the back translation matches. That is where I am still getting errors but most of the errors are exceptions that need to be added. I think even the old UEB tables can be made extremely stable even though I think the new tables will be better and easier to upkeep in the future. Ken Ken For a description of the software, to download it and links to project pages go to http://www.abilitiessoft.com