Online community management thread pt "numbers are meaningless"

Remember when I said early on that the user interface goes a long way in setting the initial tone of the site?

Go read this massive thread by @zensaiyuki about that. https://mastodon.social/@zensaiyuki/102683452946911475

Internet Community Management Thread continued: earlier in the thread I talked about tactics abusers use to abuse and how it's important to be aware of those tactics and to make your site members aware of them too, in order to build community immunity. Here's a Fediverse deep dive into DARVO courtesy of @hafnia:

https://wandering.shop/@hafnia/104997378933311199

The current Big Thing I'm working on in Improbable Island is an interface overhaul.

These are FRAUGHT. Bigger sites than mine or yours have collapsed because of getting new interfaces - either the new UI was too different to the old UI that people loved, or the new UI just plain sucked even outside of comparison to the old one.

Ravelry's ugly and migraine-inducing New Look is a famous recent example.

Ravelry's redesign is a series of mistakes which compounded one atop the other until things got really, really bad. The mistake train stretches even past the interface redesign and into their handling of the pushback of the redesign, and it's still chugging along.

This compounding of mistakes is actually pretty common. It's easy for us to see a series of bad decisions and go "Why didn't they correct things at X point?" but c'mon, we've all been there. A mistake early on leads to more.

Ravelry's first mistake was in doing an interface redesign in the first place.

Ideally you should never do a complete, ground-up redesign of an existing, well-loved site. There are reasons why people love the site!

Folks tend to get into complete redesigns because of reasons that don't actually matter.

Example 1 (Ravelry): You're doing a brand overhaul, changing your logo, and you want the site to match.

I promise you, you are the only person in the world who cares about brand cohesion.

Example 2 (metafilter & others): You're doing a new interface because google says you have to.

Again, your users don't care what google thinks. Why would they? Your search rank dropping is your problem, not theirs, and thrusting an unnecessary interface change on folks is just an attempt to take google's problem (which google made your problem) and make it into their problem.

And that's problematic.

Example 3: Your codebase is old and creaky and your site is still laid out in tables etc, and you just tried adding some new feature and it's a big pain in the bum because of all the creakiness, so you decided to recode it all and heck, why not give it a new coat of paint, since you're in there already.

Again, your users don't care! Who gives a heck if your site uses tables! That just means it'll probably still work in twenty years!

The only people who care about your HTML or will ever even look at it are other web designers, and as I'm sure you're aware by now, nobody cares what web designers think.

(except other web designers of course)

So these are all bad reasons to do a big ground-up site redesign. In general there aren't good reasons!

Example 4: you're redesigning your layout to make the header smaller, or take stuff from the top of the page and put it on the side instead, because computer monitors keep getting shorter.

This is actually legit. At least until we get the whole screen situation under control and start making computer monitors again instead of just small tellies, we're gonna have to deal with lousy displays, and putting stuff on the side instead of the top can help in that regard.

Anyway, Improbable Island's retheme falls into category 3, rewriting because the old one was a big pain in the bum.

Because this ain't my first rodeo, the very first thing I did was to recreate the old theme, as near as "to the pixel" as I could, in modern HTML/CSS.

The second thing, because Ravelry is fresh in my mind, was to LOUDLY and PUBLICLY and ENTHUSIASTICALLY and REPEATEDLY open the beta testing period to *every community member.*

Seriously, whenever I'm in chat I'm asking folks how they like the new theme, getting feedback, and converting people over to the new theme.

Ravelry only opened their beta theme test to a handful of people. Don't be like ravelry.

Next, I made some very small improvements and ran them past the people using them. Most were small quality-of-life improvements, adding spacing on specific elements to avoid finger mistakes on mobile, that kind of thing.

Then, I made one big dramatic change, but tied it to a new feature (backgrounds that change depending on what you're doing in the game) that would be impossible in the old interface.

And people love it and are excited about it.

You *can* do an interface change humanely!

To do a successful interface overhaul, you have to be intimately familiar with the reasons why people love your site, and the things they love your site in spite of. Fix the second and leave the first well alone.

The things in the middle, the things people are indifferent about, can retroactively become people's favourite things if you mess with things too much.

You want your members to feel that you're being slow and careful and considerate. To have them feel that way, first you have to actually BE slow and careful and considerate. Second you have to SHOW them that you're being slow and careful and considerate, by making changes gradually - testing locally and then uploading and asking for feedback on each one.

Let users be a part of the process and make it clear that this redesign is for THEM more than it is for you.

Barometers to see whether it's going well:

* Someone says "I just swapped back to the old skin and it looked WEIRD and BAD"
* People talk about interface *improvements* rather than interface *changes*
* People talk each other into trying the new skin
* People are curious about upcoming changes
* People ask you to fix other little rough edges while you're in there

Chucking this into the community management thread 'cause it reminds me of a thing that happened to me:

Our out-of-character chat area used to be called "Banter." We changed the name like five or six years ago. We had good reasons for doing it, not ones that I remember off the top of my head, but I do recall there was much discussion about it.

Players continued calling it Banter, and still do to this day. Everyone knows that it's called Banter.

So yeah posts are still gonna be called toots.

A thing I learned the hard way is that if you're gonna make a change, or even *talk* about making a change, do it late morning. Don't do it late at night just before you go to bed, because then when morning comes everyone's been talking about it all night.

That means that you spend half the day catching up with everything everyone said (and driving yourself bananas while you're at it), and it also means that the conversation has probably veered wildly into Speculation Town while you slept.

As an admin you've really gotta look after your own mental health very carefully, 'cause there's a lot of work involved that boils down to doing other people's emotional labour for them.

By that I mean calming down people who can't calm themselves down, helping people to remember that the admin is a person too, and where necessary removing people who've gone too far down paranoia rabbitholes to still interact with the community in a non-harmful way.

This is pretty tiring.

Expecting your admin to be good at his or her or their job is a BIG ASK, honestly.

Expecting them to be good at it in a way that benefits you, specifically, all the time, is just flat-out nuts.

An admin's job involves occasionally annoying or inconveniencing some portions of the userbase, for the benefit of others. You're gonna have to do that. It's gonna suck.

Online Community Management Thread Continued, a thing that happened today in which I wanted to talk about harmless misunderstandings.

In my game, you can make Mementos. These are items to which you assign a description and some text that appears when you "use" them. You can gift them to other players, show them off in your bio, attach stats to them, use/query/gift them in conjunction with little programs in player-owned buildings, that kind of thing.

Anyway, a long-time player in good standing made a memento that talks about someone called Kim Keller. The memento used the colour blue heavily. At the bottom, it said "Thank you LEO."

Being unfamiliar with the work, I DDG'd for Kim Keller, and turns out they're some republican arsewipe. This, plus the blue, plus the "thank you LEO" made me wonder if this was some sort of blue lives matter nonsense, but it seemed very out of character for this player, so I messaged them about it.

Turns out Kim Keller is a character in a series of graphic novels made by a guy who goes by the initials L.E.O. It had absolutely nothing to do with the police and it was all just a series of coincidences that led up to a harmless misunderstanding.

But if *I* had that misunderstanding, it's reasonable to assume that my other players might also have the same misunderstanding, and have cause to feel unwelcome or unsafe.

So we talked about it a little and we ended up erasing the memento, which I didn't feel good about (I would have preferred to just change it so that it didn't set off any potentially harmful misinterpretations).

The important part here is catching these things early, because if someone misinterprets this thing to mean blue lives matter, and talks about how that makes them uncomfortable, and the creator isn't around at the time, then that sort of convo can spiral.