On 11.10.17 21:29, Andreas Tille wrote: > On Wed, Oct 11, 2017 at 06:20:59PM +0200, Steffen Möller wrote: >>> One question regarding the 'NA' values. I'd consider it more readable >>> if we would just print nothing if there is no entry available. Or am I >>> missing something? >> I put them down internally to record that one of us had looked for it. Our >> task list is likely the source for the maintainers of the respective >> registry to get the missing bit curated. I would therefore like to >> show that. And frankly, it is a bit of effort to show that there is no >> such entry (or that the cognate entry refers to a different tool) and we >> should take some pride in that. >> >> How about putting the "NA" (or should it better read "N/A" since we >> are not in R ?) in a hyperlink of a query URL of the respective target >> registry? We would then be userfriendly again. > Well, as far as I know we have no content to point the user to. We know the query URL. > So why > pointing anywhere? I keep on wondering whether I'm missing the point. It saves some time for the user and you did not feel good about just showing the "N/A". > If a user want to do a query why should we fake to hold a hand while we > are clueless ourselves?
After some time there should not be much N/As to show at all. Showing the NAs is a bit of an embarrassment for the respective database, so I expect that showing N/As will help to get them fixed. And if we show the NA but the user follows the link and finds something, then there is more of a chance to get a bug report suggesting an update of our then outdated reference. Which is what we want. But just as I said - there should not be many NAs after some time, no need to talk much about it. I am happy for every pointer that surfaces. Steffen