At the start of my day, the counter went down from yesterday's, but not to zero. Is it taking neither my midnight nor my schedule refresh time as the start of "today"? I'm OK with that. It would just help to know.
I find myself choosing (I’m not sure) more and more. What am I supposed to do when two or more senses fit?
I have the same experience, but maybe it will help if you post a few such dilemmas. Without that, "then choose (I don't know)" is a reasonable response, but perhaps not a very satisfying one.
I believe it is now set to the server's time, not your time. As to the "no sentence was shown" issue, that might be tricky to nail down (there is a lot of randomness intentionally built into the system), but I'll see what I can do.
If two senses fit, I'd choose the one earlier in the list.
The earlier sense will generally be the less specific one though.
It would be good if there was some way to request leniency for very similar meanings. Perhaps a forum thread to discuss such sentences and meanings.
Like I'm trying to find the most fitting meaning, but others could be simply picking the first. Then perhaps I'm marked wrong for my extra effort (and for not using game theory!) when in fact I feel I've picked the best meaning. Of course, I understand such efforts could feel like overkill.
I also prefer this, but on first glance my eyes skipped straight past "to sell" and landed on "to betray", so it would help if the highlighted definition was still highlighted green or in bold or something (or maybe the other definitions could be greyed out to deemphasize them?).
To be completely honest, we just do not have the community or activity level necessary to expect more than a rare discussion in another forum about the which sense a word in a single sentence (out of 200,000) is using. I can always tune the acceptance system to require more agreement before accepting a sense, but it's plenty hard enough (this is not a complaint, just the reality of it) getting volunteers to do this first step.
Also, small update going in today - after it chooses a word, it'll prefer sentences that have been marked by someone else, so there is a bit more of a hidden push to getting sentences rated enough so the info becomes visible.
It would be great if we could choose to review the last X number of submissions we make to change any mistakes we discover. Sometimes after making multiple submissions for the same word, I realize that I had overlooked the best choice for an earlier question. Even if this option is tucked away in an obscure corner, I would like it.
When I discover that I've made such a mistake, I may feel so guilty that I stop making more submissions, when in fact I've just become "more competent" at dealing with that word, so it could be argued that I should be making more submissions for it. In any case, in my opinion, there is no need to use "quiz logic" (ie, answers are final and unretractable) for such projects.
I might be able to do that at some point, but not in the meantime - just too busy, and about to release the grammar markup tool.
It sounds like an easy change, but that is a completely different set of data/layout to create, tools to adjust, go back, etc., so it is far from a trivial change. (as it's more of an "addition")
