Used Commit all to send all translations from the TXLF to the TM. Added file TXLF file from the previous project to the new projectĤ. Created a new WFP5 project, with an empty TMģ. Made a copy of the original source file (identical, under different name)Ģ. I contacted Wordfast tech support who helped me solve the problem :-) Like you said John it was probably down to tags, the TXLF might have become corrupted along the way because the idml file was very long and had a lot of tags. Post title: Corrupted TXLF fixed by creating a fresh project It should not matter which OS or which version of WFC anyone is using, since all versions of WFC use the same TM and glossary format. If I understand correctly, you can share TMs and glossaries via the LAN, and those are simply text files. How to enable "up to 20 colleagues via LAN", would different WFC versions (5+) and W8/7/Vista/XP 32/64 get along? You can set concordance to find only exact expressions, or not, or to find only exact words, or not, but concordance search will always find results with all the words that you search for, so it's never fuzzy. It's not possible to set a fuzzy percentage for concordance searches, as far as I know. The settings are standard and Concordance manually does find 75%+ fuzzy matches properly. What is your maximum number of results (in PB e.g. I don't use the latest version, but concordance searches always search the TM first, then the BTM, and then any sibling TMs, up to the maximum number of results. Do you mean that WFC creates its own BTM and then searches it? Or what do you mean by "test BTM"? I assume we're talking about concordance searches here, sincec you mention "sibling TMs".
The latest user manual doesn't mention anything about a "test background TM". I noticed that the latest version uses a test Background TM, ignoring two sibling ones at the same folder. In such a case, your colleague might use WfAnywhere to get a taste of the features. Did you try to select one of the existing TMs as BTM? if such selection is impossible, then it is because of the 500 TU limit of demo mode. I don't understand your mention that Wf " ignor two sibling ones at the same folder". The procedure is explained (succinctly, I must add) on page 18 of the current manual ( ) Links to Machine Translation (MT) will only work for two or three segments in a session.Ģ) Wordfast web site actual quote is "An unlimited number of users can share the same glossaries over a LAN (Local Area Network)." (see #of). When the limit is reached, a message will tell you at that point, you can create a new, empty translation memory(1) and keep translating. "The Translation Memory (TM) will hold only about 1000 translation units. Post title:"start using Wordfast Classic with all its features activated, except these demo limitations"ġ) Quoting from Wordfast web site ( ) the only limitations of the demo versions are: 9/10 times it is a tag but could be something special about the IDML file that corrupts the reconversion back. I would contact tech support to see what the issue is. So don't be confused by an incomplete progress bar. This is confusing many users and I believe it will be changed in the next version. Currently, if you want it to progress, you can either add a space, remove it, then commit or use the "Verify Segment" icon (shortcut) or click on the green arrow in the far right column before committing. It happens quite frequently that you don't have to edit a fuzzy match (e.g. The progress bar does not increase if you commit an unedited fuzzy or unedited machine translation segment. I, too, have encountered that issue before, and your answer is the first explanation that I see. Is this mentioned in the documentation/Help?
Topic: Wordfast Pro 5 Clean up unsuccessful