More on the short and long compilation method and clauses
Well, we all knew Instant Text was a deep program - just didn't know how deep! I have wondered for a very long time why certain phrases didn't come up - but it didn't occur to me to try multiple setting compilations to get them, till just last week. I do think this method is one that is well worth trying out. For each doctor I have applied it to, I notice a distinct improvement in the number of phrases that appear and in shorter continuations - and a small but distinct jump in the productivity stats. I have settled in at 3 and 5 for the "short compilation" and probably 6 and 9 for the "long compilation." I Include the short compilation above the long compilation when filling the shell glossary (Dr. Repeto's glossary). My folders of material to be compiled vary quite a bit in size but are generally in the range of 200 KB to 1 MB. If the file is over 1MB, those setting may produce too large a glossary (at least it put my old 400MHz 128MB RAM computer into slo-mo on one such monster glossary). I should mention that I have compiled all these files with the Options/Continuations level of continuations set to "All" and a minimum continuation length of 9. I haven't experimented with fewer than 9 for that setting lately but now would be a good time to do so. :) As to the figures on "clauses". For those like me who forgot what Jean and IT mean by "clauses", the IT V manual says that "Clauses are additional phrases created by Instant Text, and a large number of clauses ensures rich continuations." (p 49; the manual discusses clauses again briefly on page 99). The number of clauses and the other figures below are viewable in the Glossary Viewer at the bottom of the screen. Dr. Repeto 2 10
Dr. Repeto 2 5
Dr. Repeto 6 10
Repeto 2 10 vs. Repeto 2 5 and 6 10
So the combo has a lot more entries but not that many more clauses. Again, not sure how many of these are duplicates. Back to you, Jean! Jon
| ||||||||||||