Talk:Main Page

Jump to navigation Jump to search

Use of Level 1 section headings

I just had a first stab at some editing an existing entry, and creating a new one. One small thing I noticed: all of the pages I've seen use Level 1 headings for the main sections, but this is discouraged by the MediaWiki documentation (esp. since Level 1 is already used for the main heading of each entry). See e.g. here:

And also:

So maybe it would be better to change those to level 2 (also in all cases I've seen the child sections use level 3 headings)? Not a big deal of course but it's a bit ugly and it's probably easier to fix this now than postponing it to a later stage (also existing entries will most likely be used as a template for creating new ones, so the problem will get progressively worse if left as it is). I suppose this is also something that might mess up attempts at automated text extraction. johanvanderknijff (talk) 14:58, 20 November 2013 (UTC)

Good point - I didn't know about this recommendation. It should be possible to automated this transformation using the Pywikibot framework and the mwparserfromhell. We could change the tool template now and use a bot to modify the rest? Andy Jackson (talk) 15:15, 20 November 2013 (UTC)
Great! It would also help to mention the template on the main page (I don't have edit rights there), as I didn't even know there was any! Anyway, I just updated the template myself.johanvanderknijff (talk) 16:07, 20 November 2013 (UTC)
I updated the main page to include an explicit link to the template that should be used when creating new Tool entries. Andy Jackson (talk)

Purpose field in infobox

Another thing I noticed is that for most entries the purpose field in the infobox more or less repeats what's already in the Description section. So maybe it's better to leave it out altogether (on the other hand it might be useful for automated discovery/analysis)? johanvanderknijff (talk) 14:58, 20 November 2013 (UTC)

The intention was the other way around - that the 'purpose' field was a brief description that might be published and re-used in other forms (due to being in the infobox), and the Description section should be more detailed. Andy Jackson (talk) 15:15, 20 November 2013 (UTC)
Yes, that's what I thought as well. For some existing entries both are identical though, so it wasn't immediately clear. johanvanderknijff (talk) 15:56, 20 November 2013 (UTC)
I would say the goal is to improve the body Description section for those items. Is the Description in these cases exactly identical to the purpose? If so, I might be able to check for that automatically. Andy Jackson (talk) 21:18, 20 November 2013 (UTC)

Perhaps rename Imaging category to Disk Imaging?

Another thing: people might end up mixing up the Image and Imaging categories. Maybe rename the latter to Disk Imaging to avoid this? johanvanderknijff (talk) 16:15, 20 November 2013 (UTC)

Good idea. Unfortunately, I think that due to the way MediaWiki works, this means editing all the items tagged with that category (unless there's a trick I'm unaware of). However, perhaps I should tweak COPTR Bot to do this too. Andy Jackson (talk) 21:22, 20 November 2013 (UTC)

Problems with login procedure (VeriSign)

Having created an OpenID with Verisign, signing in to COPTR is a bit hit and miss. It always takes multiple attempts before I can successfully log in, and I first have to work myself through multiple authorization errors. Not clear to me whether the error is caused by COPTR or VeriSign, and results seem to be a bit random. Also after some time of inactivity I seem to get logged out automatically. Confusingly, the COPTR login status at the top right then shows I'm still logged in (depending on which page I'm on, it seems), and I have to go through the whole login procedure (including failures) again. johanvanderknijff (talk) 11:16, 21 November 2013 (UTC)

This StackOverflow answer outlines some of the possible causes of problems with OpenID logins. I guess we could perhaps enable non-OpenID logins, although I wanted to avoid that for fear of making it even easier for the robots/spammers. Using Google as an OpenID provider is working fine for me. Andy Jackson (talk) 11:40, 21 November 2013 (UTC)