Board 8 > Blizzard reveals their diversity tool

(+) Yeah! (+)

mainly you just endure the drama and toxicity longer than everybody else

but theoretically the teams fight each other and you vote for the winners while you wait for that to happen

masterplum posted...

I still don't understand how teams in mercs win and at this point I'm too afraid to ask

You decide which side is stronger given skills and battle conditions. So sora vs cloud who would win? Then you add more to each side. Then you take into account ability use like say sora getting a damage boost or cloud starting with a broken leg. Finally there is terrain.

Mostly simple once you have feats for everyone. Kinda

Also I had merc 4 broken before rage monster sadness hit me hard. I forgot who I replaced but I turned that team into a fun beast

Being able to care about winning for that long (that's what sabotaged me, and also most midgame team eliminations) and being able to keep a cool head about it (what sabotaged Chris and a few others) is a test of mental endurance to be sure.

You won't find anything else like it on B8 or anywhere else really. I'm glad I played but I would never do it again.

Look all I am saying is if a night club killing hadn't happened I win mercs 4 in a walk.

Which is my way of trying to make light of a horrible period in my life but yeah!

No argument here man. You seemed pretty on balance in Mercs 4 until that point (not so much Mercs 3 where you seemed high strung throughout most of it but yeah, you'd grown by Mercs 4 a good deal)

It is funny to think how long I have been on here and how much I have grown. From me when I was 16 to 34 its like geesh. I am at the age where I can look back at the person who I was when I first came here and have sympathy for that version of myself. He was just trying his best and he said a lot of stupid shit. But honestly that kid didn't think he would still be alive at 34 let alone with a lot of happiness in his life along the way to go with the sad times.

Lots of trips down memory lane in this place for me. Good times.

Tom Bombadil posted...

my "statute of limitations" is particularly short re:b8 because daggum we used to all be jerks back in the day, myself included

That's not true! Whenever we said "Everyone involved in Mercs is terrible", we always added "except Bolo".

Lopen posted...

This was me like 5 years ago with foolmo.

Yeah you, Maria, Ulti, Sephg, Nanis, Realo, etc have said the same

The problem really must be me

MariaTaylor posted...

Why aren't they defending me in this topic?

They are

foolm0r0n posted...

They are

please talk to your therapist about me. it will help you a lot more than...... whatever the hell it is you're doing in this topic.

foolm0r0n posted...

Yeah you, Maria, Ulti, Sephg, Nanis, Realo, etc have said the same

The problem really must be me

The irony is you're illustrating my point perfectly and you don't even realize it.

My therapist said the difference between "we" and "I" is in fact not pedantic

Your therapist sounds like a yes man

Understandable as you probably stopped seeing all the therapists that challenged your world view because they were terrible at their job

i think that a diversity tracking tool for an ensemble cast of 3d models is fine because it's an easy way to generate extra ideas and avoid accidentally getting homogenous, which is different from a tool that tracks the diversity of the employees and gets mad whenever you've hired too many white people, but that's just my opinion. what do you guys think?

excuse you we're talking about mercs and therapy

NeatoAnAccount posted...

i think that a diversity tracking tool for an ensemble cast of 3d models is fine because it's an easy way to generate extra ideas and avoid accidentally getting homogenous, which is different from a tool that tracks the diversity of the employees and gets mad whenever you've hired too many white people, but that's just my opinion. what do you guys think?

In the general sense it's not a bad thing, but this tool as it's been shown seems flawed. First because it assigns numerical values to different identities to rank them by which one is "most diverse" feels very close to phrenology & bioessentialism, and also could still lead to a lack of representation among identities that are ranked lower. Like the screenshots shown say that a white character scores 0 Race, but an Arab character scores 7. My understanding is that a Black character would be somewhere in between those two, so this tool would consider it "more diverse" to always make characters Arab and never Black.

Secondly, the way it's set up places greater emphasis on the highest scoring identity in a category, rather than a breakdown of how frequently a given identity appears in that category. Going back to the example of a white character getting 0 in Race and an Arab character getting 7, they could have all support characters in Overwatch be white, and then add one Arab character, and suddenly support characters go from 0 to 7. That's the definition of tokenism, which this tool is supposed to prevent.

It also isn't very useful to prevent homogeneity, as long as it's not homogenized on low-scoring identities. Like making all healers female would score higher than having a mix of genders, even though that's not actually diverse at all and in fact feeds into a stereotype.

A better tool would just be to track the facets of each character's identity & gameplay role, and then be able to view a breakdown of how different categories intersect, i.e., what's the breakdown of race among DPS characters, gender among elderly characters, body type among female characters, etc. The push to metricize everything is what makes this tool weird.

MrGreenonion posted...

A better tool would just be to track the facets of each character's identity & gameplay role, and then be able to view a breakdown of how different categories intersect, i.e., what's the breakdown of race among DPS characters, gender among elderly characters, body type among female characters, etc. The push to metricize everything is what makes this tool weird.

I'm pretty sure the simple solution here is to just have two pools of data. One that's aggregated across many different games, and one that's specifically drawing from what already exists in your game.

I think it would be useful for identifying gaps because there is a lot of unconscious bias that even a seemingly diverse team might not think about. But yes it'd be weird if this was the only deciding factor in what kind of character to make, and it definitely could lend itself to some tokenism if that was the case.

I don't think the tool is trying to maximize the numbers, but trying to have an even distribution. So if you have all 7s for race it will flag that as a bias.

Problem is that the numbers being picked are all subjective so if the tool says you need a 4 for race you can make a Jewish person or whatever and just call it a 4.

I just wanna know how it works- I was thinking of it as automatically assigning numbers based on the rest of the cast (or games in general?) so foolmo's token Jewish person would be automatically scored, probably as a 7. And then either yay, you got a 7, number goes up, or maybe the goal is to add more Jewish people until they're rated 4 and presumably don't standout as tokens anymore. Who really knows? AFAIK all we have to go off is that one graphic.