For the Esperanto phrase "Ni memoros instrui tiun lecionon", I gave the translation "We will remember teaching that lesson", which was rejected, and given the solution "We will remember to teach that lesson". The two English sentences are obviously different in meaning, but why would the given Esperanto sentence not cover both meanings? To disambiguate, I'd think you would have to explicitly say "Ni memoros ke ni instruis/instruas/instruu tiun lecionon". The PIV defines "memori" as "Konservi en la spirito la perceptojn antaue ricevitajn, la ideojn antaue akiritajn"; it does not mention intent. Perhaps this sentence here is a calque of the English idiom "to remember to + verb"? If you replace "remember" with "recall" the difference may be more obvious. This kind of reminds me of the confusion between "provi" and "klopodi" when translating the English "to try". Then again, I may be wrong; what do you think?
The sentence «Ni memoros instruinta …» does not have the instructing as an object, so the "instruinta" ("having taught") must go with the subject ("ni"); hence would it not mean something like: "We, having taught, will remember (...something)"?
In English, the meaning of "remember + V-ing" differs from "remember to V", with former implying past, and the latter implying (future) intent ("we will remember that we should be V-ing, and will in fact do so"), a distinction that is idiomatic and particular to that verb; for example, there is no such difference between "we like V-ing" and "we like to V". The phrase "I remember eating bread" does not mean "I remember that I have to eat bread". The dictionary sense of "memori" in Esperanto does not appear to warrant assuming that the same idiomatic distinction of the English uses of the verb "to remember" can be found in "memori".
This is a common fault in dictionaries. Good dictionaries will show how words are used so you can simply slot the words into a sentence.
For example, if you look up "stop" in English, it should define both "stop doing something" (ĉesi fari ion) and "stop to do something" (halti/ĉesi por fari ion). Likewise, an esperanto dictionary should show "memori fari ion" (remember to do something) and "memori ke oni faris ion" (remember doing something).
I agree, examples are very helpful. In this particular case the definition of "memori" is from the PIV, which is generally considered authoritative, and it gives about a dozen examples, none of which parallel the English "to remember to V" in the sense of "acting on intent". If the dictionary defines a verb to have a specific meaning, and does not give another (different) meaning, what basis do we have to assume the dictionary is at fault in omission? The fact that the English word "to remember" corresponding to the one given meaning also has a second (different) meaning (more like "remind oneself to do X, and subsequently do so") in an idiomatic combination in English does not necessarily imply that the Esperanto word must have the same additional meaning and idiomatic combination, I'd think.
Yeah, the problem is that it just gives random examples. What it needs to give are
structures. The examples for memori don't give any clue about the structure other than that it can be used with the accusative or with pri. There's nothing there to show how to use it with verbs.