User talk:Mikeblas/Archives/2018/September
Jump to navigation
Jump to search
 | This page is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Mentioning
Why do you mention me in your edit summaries? ~SML • TP 20:52, 1 September 2018 (UTC)
- Looks like I mentioned you in this edit summary because I was repairing a referencing error that you created in the article. I figured that you'd want to confirm that my correction got the article into the state you had originally intended. -- Mikeblas (talk) 23:04, 1 September 2018 (UTC)
Refname ":0"
I was wondering who else was fixing all the duplicate ref names - well done Mike, some of them really are unfixable without spending an inordinate amount of time on them. My pet hate is Refname ":0" or similar meaningless ref names, we should probably have a policy against those.:) CV9933 (talk) 16:29, 15 September 2018 (UTC)
- Hello! I was wondering how we'd find eachother. :) Do you know of anyone else who is working at it? Indeed, the ":0"-style names are a problem. Do you know what tool creates them? It's also quite difficult to track-down references named in transclusion from other articles, or in templates. I've been trying to update Category:Templates that generate named references as I go. -- Mikeblas (talk) 23:55, 15 September 2018 (UTC)
- It has been a while since I fixed this category of errors in earnest, it was one of your edit summaries that got my attention; those users probably don't realise that they are contributing to a pile that was well over 30,000 in 2015. I think the worse part is spending some time fixing a page, checking before saving and then discovering that the cite that you worked on is actually a dead link. I watched the number of errors reducing but wasn't aware who else was doing the fixing, although at some point I noticed that a whole swage of them disappeared in a chunk. (I can't imagine a script could do that with the amount of dilligence required.) Your right about transcluded template refs, they are often difficult to track down and when you do, there is still the dilema of relying on the transcluded refname or changing it. Only another 3,500 to go! CV9933 (talk) 14:23, 16 September 2018 (UTC)
- There are a few categories that can be swatted down. {{Canada census}} creates some named links, for example; lots (and lots!) of pages duplicate this reference. I completely agree that most users don't realize they've made an error in their edits. I wish Wikipedia was a bit better about checking syntax and semantics before allowing a page to be saved because I think it would help. Some pushback has come from users I've identified (see above) but far more "thanks" notifications ... at a ratio of about 100:1, wihch is reassuring.
- Are there more of us? How do we find eachother? -- Mikeblas (talk) 15:05, 16 September 2018 (UTC)
- Maybe there is only you working on them now, I don’t think I would have had the tenacity to stick at it for as long as you have. I just occasionally pick out the ones that interest me.CV9933 (talk) 16:02, 16 September 2018 (UTC)
- I’m don’t know what tool creates the ref names with :numbers but looking at this edit one might suppose it is a tool frequently used by med students. Perhaps user: Doc James might know? CV9933 (talk) 12:17, 17 September 2018 (UTC)
- I think it is visual editor. There have been requests that the team working on it to improve the naming but they have not yet.
- Looking at that edit and that is a disaster :-( Basically people are using visual editor to write content in their sandboxes and copy and pasting it into a single article. Ian (Wiki Ed) what can we do to prevent this issue?
- I have taken the first stab at fixing it[1] Not sure if more issues remain. Doc James (talk · contribs · email) 15:39, 17 September 2018 (UTC)
- Hi Doc James! Indeed, colon-number rereferences pasted from sections grown in sandboxes seem to cause lots of problems. The fixes in the diff you link look great to me. I think the trick in these cases is to find the change that first caused the duplicate ref-def error and figure out which references were which; which stay ":0-old", and which are ":0-new" and can be renamed ... if that makes sense. :) -- Mikeblas (talk) 12:23, 18 September 2018 (UTC)
- I’m don’t know what tool creates the ref names with :numbers but looking at this edit one might suppose it is a tool frequently used by med students. Perhaps user: Doc James might know? CV9933 (talk) 12:17, 17 September 2018 (UTC)
- Maybe there is only you working on them now, I don’t think I would have had the tenacity to stick at it for as long as you have. I just occasionally pick out the ones that interest me.CV9933 (talk) 16:02, 16 September 2018 (UTC)
- It has been a while since I fixed this category of errors in earnest, it was one of your edit summaries that got my attention; those users probably don't realise that they are contributing to a pile that was well over 30,000 in 2015. I think the worse part is spending some time fixing a page, checking before saving and then discovering that the cite that you worked on is actually a dead link. I watched the number of errors reducing but wasn't aware who else was doing the fixing, although at some point I noticed that a whole swage of them disappeared in a chunk. (I can't imagine a script could do that with the amount of dilligence required.) Your right about transcluded template refs, they are often difficult to track down and when you do, there is still the dilema of relying on the transcluded refname or changing it. Only another 3,500 to go! CV9933 (talk) 14:23, 16 September 2018 (UTC)
This edit is interestingbecause it has triggered the fact that there is a duplicate ref name ("johnson") which is getting transcluded from Template:Table of MRI sequences. I only know that because I worked on that table recently. The edit looks like the first pass of a series of edits, but I'm not sure why this edit was the trigger - no error displayed before this edit. CV9933 (talk) 14:33, 18 September 2018 (UTC)