I've addressed all the issues in the latest changeset, I hope. The start date is only for this part, the rest is much, much older. (I seem to have gotten the name wrong, oops.)
https://en.wikipedia.org/wiki/Grace%27s_Old_Castle

Alright, thanks. I've corrected it with the two petrol stations I remember mapping recently.

Fixed

The tactile paving is indicating where the traffic signals and crossings are located, though. I read the wiki before I started adding the tactile paving. I only added the tactile_paving=no to the traffic signals ON the street, so I could run an overpass looking where there is no information about tactile paving. Obviously there is no tactile paving for cars.

typo most likely, I'll fix it

oops, sry, I fixed it now

I might have taken mapillary, can't remember.

I don't think I can. I was walking the Barrow way that day, I'm not in the habit of doing that regularly. ;-)
Thanks for adding the house.

Thanks, would you mind added a shorter version under the video,if that's how you found it? I've fixed this one and will bear it in mind for the next cases.

I'm following up on it with the owner of the land.

I don't know, I haven't surveyed them in the past. There are no house numbers visible from the street anyway.

But there's a car parked in it.

Hi,
thanks for adding the ringfort, always good to see people interested in that. For better readability for AI, the correct tagging is
historic=archaeological_site
site_type=fortification
fortification_type=ringfort
And they are early medieval, not prehistoric. Hill forts would be prehistoric. Cheers, Anne

Hi, thanks for adding the ringfort, always good to see people doing that. However, the correct way to tag them is
historic=archaeological_site
site_type=fortification
fortification_type=ringfort
rather than putting "Ring Fort" as a name. Also, they are early medieval, not prehistoric.
Cheers, Anne

Oops, according to mapillary, it's the one further south.

There are a few issues with river relations (I think), but I can't make head nor tail of why they look broken in iD.

Hi, I downloaded the whole relation into JOSM and there was nothing missing, the members were just unsorted. If I was the cause earlier, it probably had to do with the DED creation, but of course not intentionally.

Hi,

I think I split them in the process of finding where the 1911 census boundary went and then probably didn't need the split after all. I wasn't aware that splitting lines had such an effect, because i didn't delete any members of relations.

Anyway, I'm not planning to add any more historic bondaries in Limerick.

Sorry about that.

the key is actually post_box:type, i'll fix them for you. Thanks for adding that information!

Well spotted. It doesn't seem like it.