I have reason to believe that we no longer are utilizing the Merge Policy the way it is intended. The other day I asked @JayneG a message about two articles that had interchangable titles, and it seems like it’s fallen in deep, dark times when merges are now slipping into dealing with what the content has because title’s popularity is at-stake.

When Anna was here, we were all good - merges could be performed and there wouldn’t be so much back-and-forth with messages only ending in disaster. When titles are synonymous, it now seems like we keep both articles - because their readership is high! That’s not right!

I think we should check on the validity of our Merge Policy ( https://www.wikihow.com/wikiHow:Merge-Policy ) so it best updates our current viewpoint.

The sample I gave Jayne was the interchangability between https://www.wikihow.com/Fold-Paper-Airplanes and https://www.wikihow.com/Make-a-Paper-Airplane (about the only way to make something out of paper is to fold it - right - especially when it’s dealing with paper, you can’t very much cut a paper airplane out of paper without folding to make it flyable. I couldn’t merge it because that silly Summary Video, and got nipped when she told me that they are both highly-viewed on Alexa reports.

All because of #readership #geez !

I’d like to counter one point in your post before addressing the main point: readership is important. This change likely has nothing to do with Jayne or staff being Awful Corporate Who Don’t Listen to Us and everything to do with keeping us on Google results. Something like this wouldn’t have happened a few years ago because a few years ago, Google was not so strict on their algorithms and wikiHow was still appearing fairly regularly in search results even when articles were stubby or lacked experts (even on medical topics). Now, we’re at the mercy of whatever Google does this time, and staff may have to make changes that we hate because if we disappear off of search results, we won’t have a website or staff to hate.

That being said… I don’t think some of the latest changes mesh all that well with community spirit and autonomy. The article merges seems to be an isolated case at the moment, but I’m also thinking about the recent title thing where readership will crash if we retitle something. These are changes made solely by staff, based on data we can’t see for ourselves, and it does feel a bit like we’re losing autonomy due to information that not everyone would necessarily know about or understand. I’d imagine it gets frustrating to volunteers to merge or retitle something only for staff to undo it, and I’m personally still confused on whether it’s safe to retitle any indexed pages without staff permission now. And like you brought up, Byankno, sometimes the staff changes don’t mesh with community policy, which can lead to confusion with volunteers and also isn’t that great if the only way we’re aware of why we aren’t sticking to a policy is one-to-one communication with staff. (If something like this was going on when I was on my NAB trial, I would have gotten very confused.)

@JayneG and @Chris_H - I know we can’t give volunteers access to all staff data, because otherwise the information could get out to competitors. I’m wondering, though, if maybe we could consider some ways to grant community members some more information so that we don’t end up making changes that you guys later have to undo, or feel like we have to ask staff for things we used to be able to do on our own. Some ideas I’m thinking of:

  • Giving boosters and/or admins an opt-in, very stripped down version of some staff data (like “Readership: None/Low/Medium/High” or “Alexa: No/Yes”), maybe on the Helpfulness widget, so that people who would typically be handling title changes or merges can tell whether it’s safe to handle on their own or if they need to ask staff beforehand
  • Placing a warning on high-traffic pages when someone attempts to redirect the page (I’d imagine this is as simple as adding a filter - we have filters that prompt an “are you sure?”-type message when someone removes a FA tag, for instance)
  • Opening a discussion with community on the state of the title and merge policies, so we can discuss whether they need to be amended to comply with current-day data and search engine algorithms

Some staff-and-community discussion would probably benefit us all here, and help us find a resolution.

4 Likes

@EpcotMagic

Hi Chris,

Jayne and I discussed this merge, and after checking, realized that this article was one that we provided to the Amazon team as part of our collaboration with them to provide answers to popular questions on Alexa devices. We want to keep those articles live unless there’s a really good reason not to do so. The decision not to go ahead with the merge was mine, and the result would have been the same regardless of whether I was working with Jayne or Anna when communicating it.

And with respect to our merge/title policy, I can make the argument that these titles shouldn’t have been merged in the first place. There are versions of paper airplanes that are made with different types of paper, and glue . So “How to Make a Paper Airplane” isn’t necessarily interchangeable in all circumstances with “How to Fold a Paper Airplane.”

I’d also like to make a final point about policies on wikiHow, and how we’ve always differed in how we regard them than some wiki sites, most notably, Wikipedia. Policies on wikiHow have always been used as guidelines that help us make decisions in ways that best serve our mission, which is helping people learn new skills. They’ve never been applied in a black and white/all-or-nothing fashion. It hurts nothing to have two articles on closely related topics, if both of those articles are helpful and reach readers. It does hurt our mission if we make redirects and change titles in a way that causes us to reach fewer people. I’ve researched the matter extensively, and it’s clear that overly aggressive application of our title guidelines can hurt our mission of reaching the most people we can with helpful instructions. When making changes to articles, I’d encourage everyone to consider both our guidelines and the greater mission of reaching out and helping people—and when there is tension between the two, to give greater weight to the mission.

5 Likes

@anon74718567 I have some analysis from Chris H that I can share to give you an idea of how title changes are affecting readership. This graph here shows capitalization changes in titles from ROBLOX to Roblox last week. Small to no lasting changes with these ones. This graph here shows a larger change in a title; from Cut a T-Shirt Cute to Cut a T-Shirt to Make it Look Cute . It didn’t remove any words, simply added others in, and there was a pretty significant drop in views that didn’t recover, so Chris changed that one back. Hopefully it will start seeing readers come back as well.

So in this regard, making changes to hyphenation and capitalization is fine, but anything other than that should really be well considered on articles with lots of readers. Of course, as you mentioned, it is hard to do this if you don’t have access to any data - which is why we suggested coming to myself or Chris for advice for high traffic pages. However, I really like your idea of creating a widget with basic data that can help guide these decisions for Boosters and Admins. I’m going to see if this is something we could implement to help restore some autonomy in this case (no promises, but I will advocate for it).

I think a good focus here would be to make any title changes that are needed before an article goes live. Although there may be some out there that aren’t ideal, I think Boosters working in NAB at the moment are really on top of making sure titles are correct, and that’s where people who enjoy fixing titles can help. Additionally, changing titles on articles with lower readership (those that don’t have the warnings) can often help them find more readers.

5 Likes