Topic on Talk:GlobalProfile/design

This initiative seems predicated on the idea that other editors are targeting editors with redlinked userpages as probably editing in badfaith. An alternative theory is that the truly sure mark of a newbie is a redlinked talkpage, so the best way to protect goodfaith new editors is to welcome them once you've checked their first edit and found it to be in goodfaith. In my understanding Huggle etc divides users into four broad groups:

1. whitelisted editors
2. editors with bluelinked talkpages and no recent warnings
3. editors with redlinked talkpages
4. editors with recent warnings.

Editors with bluelinked userpages and redlinked talkpages are currently an anomalous subset of group 3 that includes spammers and myspacers.

Efficiently moving users more quickly from group 3 to group 2 and group 2 to group 1 would reduce the load for Hugglers and other patrollers, thus making recent changes more efficient, less stressed and hopefully less likely to miscategorise a goodfaith newbie as a vandal.

Possible ways to do this include:

A user preference for those who have hotcat installed to "welcome unwelcomed new editors with a personalised welcome based on the category I've just added to their new article". This would require a bit of code using the wikiproject or more likely category to tailor the welcome to include a Wikiproject or Wikiprojects that they might find interesting - we already have w:en:User:DASHBot/Wikiprojects/Templates the table for this, we just need the code. Newpage patrol would I believe be transformed if there was an option to auto welcome newbies whose articles you had patrolled. This could just be done by creating an extra button for patrollers where the article is by a Newbie - so [Mark this page as patrolled and welcome author] would appear as well as [Mark this page as patrolled]. T

Linking Huggle and the other vandalism reversion tools and manual vandalfighters so that white lists can be generated more efficiently would speed up the transition from the second group to the first. I revert vandalism manually and welcome good faith users using Twinkle, but I'm pretty sure that when I check an edit and do nothing that does not currently contribute to getting that editor whitelisted.

I believe that there are currently a couple other experiments being run by the Account Creation Project team that involve automatically creating User and User_talk pages - specifically with the goal of avoiding the "redlink" problem described above. In some ways, the impetus for this project is driven from that - if we remove the redlink notifications, then page patroller difficulty increases radically so we need to look into ways to avoid that.

I am currently of the mind that providing default User and User_talk pages for new users is a good idea, if only so that we can show them that these things exist at all. The sooner new users understand that discussion pages exist is better, in my opinion.

As far as templates are concerned, I am very leery of them. We have some research that tells us that "manually saying mean things" to new users is far, far preferable to "dropping a template on their talk page". Templates are rather forbidding, generally, and overall poorly designed. Most are blocks of text which instantly cause users to drop into "scan" mode.

It might be better to connect this with WikiLove to help show appreciation.

That's really interesting re dropping a template on the talkpage. The only research I've seen so far shows that welcomed users are more likely to stay, despite a high proportion of welcomed users getting a "welcome I've tagged your article for deletion" welcome, and as we know those users rarely stay, so the logical assumption is that welcoming really works. If there has been other research on this I'd love to see it (I know there been some focus group sort of stuff as to what people say they like, and templated welcomes do badly there. But if the actuality contradicts the focus group I'd be inclined to go with what works in practice even if it shouldn't work in theory).