Every time you see the word "algorithm" just replace it with "policy" or "procedure".

It's just a structured list of rules at the end of the day, that some person felt was suited for the task at hand, and should be judged accordingly.
Algorithms can be obscure and uninterrogatable but the same goes for policies.
Ultimately machine learning is outsourcing the output to a giant pile of math soup, but we're still responsible for what comes out the other end.
One thing I've noticed is the more transparent the "rules" are the easier it is to get people to buy into the notion of "procedural" justice.

The more the "rules" are obscure and obfuscated, the more people obsess over outcome disparities alone.
I mean we always obsess over outcome disparities, and for good reason, but when you can't transparently appeal to a shared trust in a well codified due process then everyone just focus on what the thing poops out

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Lars "Sweet Leaf" Doucet

Lars

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @larsiusprime

19 Dec
Calico seems to be climbing the Nintendo Switch sales charts lately

nintendo.com/games/detail/c…

Pat yourself on the back if you predicted this
(I did not predict this because I was not yet paying attention)
Here's Calico's movement up the North American Switch charts.

(Keep in mind this is a sliding two-week window and I haven't matched this to discounts, which matters a lot because games leap onto the charts when they're on discount then decay slowly)

Still pretty impressive.
Read 4 tweets
19 Dec
The whole "Games are still a young medium" thing is a bit played out now.

How long did it take film to mature?

1st silent film - 1888
Cabinet of Dr Caligari - 1920
Citizen Kane - 1941

~50 years spans earliest tech demos to silent-era classics to titles in the modern canon
Video games:

Tennis for Two - 1958
Pong - 1972
Super Mario Bros - 1985
Minecraft - 2011

53 years from the first twiddlings on an oscilloscope to the best selling video game of all time
It's been 62 years since Tennis for Two.

If we count up from 1888, Video Games are now where Film was in 1950.

1950 is when Sunset Boulevard came out [1]

Was film still a "young" medium still finding its feet? Or was it an established art & powerhouse of culture & industry?
Read 5 tweets
17 Dec
Mobile game market question:
I always hear mobile market is "ruthlessly metric driven", ie, that it's all a user acquisition arbitrage game (spend big on ads/cross promotion, price out your competition, make a thin profit margin, then scale it all up)

1) Is this true?
...
2) To whatever extent it is true (b/c whether or not it's true for everything in mobile it seems to me to be true for certain segments of the market at least)... is there anything fundamental about mobile that means it HAS to be this way, or is just an artifact of store design?
So basically, is this meme actually true, if it is true, to what extent and what limitations, and for any segments for which it holds true, is there an easily imaginable alternate universe where it's not true, and what is different in that universe.
Read 5 tweets
1 Jul
Kongregate is no longer accepting new games, man. End of an Era.

kongregate.com/forums/1-kongr…
Kongregate and Newgrounds was the now-nearly-vanished "Minor League" of game development, a middle step between "indie nobody" and "massive success".

You could make a mediocre game and expect a mediocre return, which is where future careers can blossom from.
This is certainly where I got my start. If it wasn't for Kong (and Newgrounds) I know for a fact I wouldn't have sold many copies of DQ1, and then I wouldn't have gotten on Steam in 2012, and I'd have picked a different career.
Read 11 tweets
18 Nov 19
"If you could use Stadia as an extension of your current gaming platforms... that would be pretty cool. Stadia is not that."

But, XCloud *is*. Google put all their eggs in this one basket, Microsoft positions it as a value-add.

forbes.com/sites/paultass…
I've read several reviews now and the consensus seems to be:

1) Reduces loading times, often by a lot!
2) Under optimal conditions, has more lag and lower image quality
3) Under merely good conditions is intermittently frustrating
4) Under average conditions is unplayable
And that's without considering the fundamental flaws in the business model (subscribe THEN buy full priced games) and the lack of guaranteed, no-extra-effort-for-the-devs, cross-save with your existing game platform that you also own the game on.
Read 12 tweets
20 Sep 17
I've noticed 3 major viewpoints (w/ some overlap) on reviews:

1. Reviews are Hit Points
2. Reviews are information
3. Reviews are noise
Both the bombers and the anti-bombers largely subscribe to "Reviews are Hit Points."

Established AAA big boys seem to gravitate towards 3
Unpacking that a bit -- small/indie developers (including myself) get scared of review bombs b/c you feel your existence is marginal already
Read 9 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!