Hi,
is anyone using DTP to store APIs (e.g. from programming languages like java, c++, ruby, frameworks, …) and/or programming documentation?
if so, how would one get started? i created a new database for documentation and added the entire api/documentation of sqlite (database i was working with at the time) to it. didn’t work. search never revealed what i wanted and was actually worse than sqlite’s own site search.
also, how do you keep it up to date?
thanks
How did you exactly add the information? E.g. one huge document or many small ones? And what kind of format did you use, e.g. PDF or HTML?
i added many small html files.
for example, this is one page i added: sqlite.org/lockingv3.html
Static elements (e.g. footers, headers, navigation etc.) might bloat the search index and therefore cause poor results. It will probably be much easier to create a focused, relevant and automatically updated search index using DEVONsphere Server.
i am not sure what you are saying here. does devonsphere server have better AI than DT?
It’s basically the same but it will be possible to optimize/focus the search index of DEVONsphere Server by stripping unnecessary stuff (e.g. like footers, headers, navigation) whereas DEVONthink is always indexing everything.
That’s also the reason why See Also/Classify results for web archives or web PDFs are often worse than the results for taken plain/rich notes which contain only the interesting article/text without any additional bloat.
somewhat related. how good are you at filtering out the fluff? if i import web sites with instapaper or readability, can you actually at least remove the small header/footer from those two (or similar services)?
Well, probably we could but these small footers/headers shouldn’t cause any major bloat or search issues.