noahtheduke

take data in change it push it out

cis - straight - white - 30 or 40 years old
clojure programmer
living in the shadow of grief


nothing remains forever empty


Profile pic commissioned from @ICELEVEL


kylelabriola
@kylelabriola

I occasionally see people say things like “Why did Valve do X” or “Why doesn’t Steam work like Y?” And while I think more and more of us are aware of the answers to those questions by this point, as Valve’s strategy with Steam becomes more clear over the years, it could still be worth saying out loud.

I figured I’d throw my two cents into why Steam is the way it is and why it will always work better for certain types of games than other types of games.


stu
@stu

but there's a few things rattling around for me in the ole dome

  • valve has had some formative experiences around curation where peoole were very upset with them for picking winners and losers by choosing who goes in the store at all. so, floodgates open and minimal curation.
  • i think their third party curation is also because they didnt want to get burned by picking winners
  • in contrast in tabletop land someone
    complaining that one of the biggest storefronts, target, doesn't do enough for the little guy would sound silly. buddy, target doesn't stock the little guy
  • i really think that more than managing user trust or believing in markets or anything else, valve doesn't want to do curation because they hate the idea of paying people to just do a job day in and day out - primarily because they dont want to pay for it and it and goes against their whole cross-functional ethos from that old employee manual. valve makes a remarkable amount of money for their headcount
  • i dont think nintendo would put a game with no visual content in a direct either. unfortunately not every game that is successful as art has a straightforward path to commercial success
  • all of this curation really feels like the role of game journalism which sadly isnt in great shape at the moment
  • i'm not saying it wouldnt be nice for valve to do more, but i do suspect anything human curated would lead to a whole lot more yelling at valve

You must log in to comment.

in reply to @kylelabriola's post:

Valve has always struck me as a very odd true-believer-libertarian kind of company. Standard ethos in that regard: everyone technically has the opportunity to get preferential treatment, no bias whatsoever.

(It did make me indescribably furious the first time I found out that big devs aren't getting charged the full cut though.)

Yeah...

Yeah the cut situation is weird. I think I could defend it as an idea if the thresholds and qualifications were much more generous. Like maybe if it was just a way to punish "spammy" or "illegitimate games" and then everyone who made decently-well-liked game got the other cut...maybe it would work? But personally if I was running a store I'd just do it the opposite way, give the generous cut to the small indies.

Good post.

Personally, to me, all of Steam clicked into place once I learned Gabe Newell is a libertarian. They will never do curation because the company was founded and still operates on those principles: that the market will pretty much regulate itself.

Which in real life tends to end up meaning "the big get bigger"

I agree with most of the above but outside of the vaguely libertarian ethos at work, the main other motivating factor at valve (and where their ethos falls flattest on its face) is in enforcement.

Being able to view the struggles of adult game makers and visual novel makers on steam leads me to believe that if valve is delusional if it genuinely thinks it's a meritocracy with no one putting their finger on the scale.

Meaning because they decide to put their thumb on the scale to disadvantage adult games and visual novels intentionally?

Yeah, I agree with that. That's why I feel like there is no point in pretending that it's an unbiased system. They clearly make some decisions from the top.

Essentially yes. The general capriciousness encountered by people trying to deal with Valve when attempting to sell visual novels and adult games of a certain style puts the lie to that notion imo.

FWIW I do agree that for all the problems, they've gotten a lot right, but it's cold comfort when they sometimes seem like the only game in town and the things they get wrong leaves those people out in the cold.

Tangentially related but I've always wanted to look more into the curators on steam and found it frustrating that it's not easier to find ones that will surface games I'm into. It would be nice if there was a way to pick a list of obscure games I like and say "show me a curator who has recommended all of these"

Absolutely. I feel like the curator system had potential but the execution was so botched. Since I've been in this headspace lately, I've actually been looking into well-liked curators.

Might make a separate post about strategies for finding hidden gems on Steam.

I'll keep an eye out for that!! So far the only thing I really know is their little interactive game recommender which is okay. But I can't help but feel that there's more out there that I'm missing

I feel like Perfect Tides is the perfect type of game for human curation/testimonial. It's fresh and personal and it's the kind of thing you might not KNOW you want until you see it, or hear someone's feelings about it. I don't think I've ever seen Perfect Tides pop up in my Steam "algo" on its own.

Yeah, my only thought after reading this is maybe curators need some improvements, because I don't see what other solution there would be. I don't want a single person employed by valve to be curating, I think many people would agree on that point. There is no data-driven model for "exposing indie gems" because what is a valuable game experience is human subjective and not even sales or reviews accurately tell me if I'm going to like a game, they're only the roughest approximation.

I'd also just hazard that if the problem of the market is "exposing more viability for smaller niche products beyond the capacity for any human market that has ever existed in history," we're in the realm of nitpicking for utopian solutions. I'm not saying there isn't improvement to be made here, and you've written a more reasonable assessment of the issue than I see elsewhere, but the plain reality is the current Steam market surfaces to me more individual creative art objects that I can pay someone to interact with than has literally ever existed, and it's already by default currently better than any other model I'm aware of.

Etsy is worse, Spotify is at least slightly worse, all streaming platforms are worse, twitch and YouTube are slightly worse. Most forms of niche media require "marketing" of some form -- social media, word of mouth, etc. Steam meanwhile has curators, a discovery queue that you can click through endless games related to your interests (but few people ever seem to bother using this for some reason), massive sales that consistently surface an enormous range of games, massive demo events that surface even more. You can infinitely scroll down the front page for more and more and more game recommendations. Everyone is out there fighting on Twitter and other social platforms to get wishlists to get attention to get sales, but like, what is Steam supposed to do to step in beyond that? At a certain point, interest requires a customer who cares to put in the effort.

Last note I'd say is people really under value the discovery queue. I've scrolled through thousands of games at this point, bought and played and recommended many I never would've found otherwise. But there's no system in reality that could have put 3827 (as of now) games in front of me to consider without me putting in the effort of clicking through the list.

Yeah that makes sense. I regret maybe overselling the "Valve curator" aspect of my blog post, although I still don't think it would really be so bad (the Featured tab on the Nintendo eShop is pretty great imo.)

I do think there are data-driven solutions to finding hidden gems, personally. An algo as simple as "high review score, but review count between X and Y number of reviews" is pretty effective at pooling together good candidates.

Some people have taken their own try at things like that: https://steam250.com/hidden_gems

If there's a way to get something like that on the homepage of Steam, I haven't seen it. But I know they've dabbled with the idea. If you go to Your Store -> Interactive Recommender and set the Popularity gauge to "Niche", they're probably using a similar tactic to find hidden gems. So I think it would be an easy thing for Valve to pull off, but I don't think it would benefit their bottom line.

I'll take the way Valve does it over whatever results in the Nintendo Eshop or the PS Store, and I don't even like Steam.

Yeah it's rough, but following steam releases, I've also noticed a few games releasing with absolutely nothing and still getting noticed. This wouldn't happen on Switch or Playstation, and I really doubt these would be games highlighted by Valve employees.

Another thing is that when you do have stuff like Nintendo Direct and State of Play, then it's the only thing people look at. They're good to showcase already pretty well marketed games to people who look at nothing else, but they won't create success out of nothing by looking at releases and picking what they find interesting.

Of course it would be nice if the home had a spot for lesser known games, but then you're just moving the problem, who gets picked, what stops them from going for high potential for sells instead of quality etc. I get why they're staying away from this.

The one place I've seen Valve do this is for Steam Fest trailers, because they do pick games here, but yeah it's mostly about what looks good and varied for a trailer.

Personally I just want something halfway between the Steam storefront and the Nintendo/Playstation one hahaha.

Yeah, the console stores are horrible for discovery. On their end, they have to solve all the opposite problems and learn from what Steam does well.

If Valve had human editorial and it happened to suck, I would just write a blogpost about their picks sucking. But personally I'd prefer to be in that situation. Also all things considered, I think Nintendo/Sony/Xbox does a pretty decent job with their picks! Valve employees could easily be as good, or they could get guest contributors.

But this is all hypothetical bc it is directly in conflict with their philosophy.

Really good text, nothing to add really. Just that my general image of steam builded in all these years has fallen in this year in a relatively little time. Not that they did a big major bad decision, but it was just, so little things, one by one, that have created a mountain of disliking that I have to deal as a player and dev... A visual novel dev, without 100$. Glad Itch exists, not that it doesn't have problems, but it's good things make it really good. But yeah I really suggests start with Itch if one is a dev wanting to grow, specially if you want to do anime inspired visual novels.

Yeah, it's this false premise that the single independent variable is the quality of the game. But obviously it isn't true if you look at the varied circumstances under which the games are developed and distributed.

To me the idea of browsing Steam is like scrolling the front page of Amazon looking for ~something~ to buy. I only go there to check out a game by name, god knows what Valve's demographic actually is for their "recommendations" algo

I agree that curation is valuable, but I'm not convinced that Valve (or their storefront) are the best place to do it. This is what games journalism is for. It is better, I think, for the inherent bias of journalism to be something that readers opt into, rather than vertically integrating with the store page.

The one argument that I can see for this is that Valve can afford to do it, whereas most games journalism is either dead broke or in a tense relationship with their advertisers. Perhaps it would be ideal for Valve to offer something akin to Amazon's Affiliate Links; where if someone bought a game by following the link on your website, you'd get some small cut of the sales price? This might allow popular curators to earn a living doing game journalism based on their effectiveness, rather than just being paid to recommend something whether they like it or not. I kind of doubt that this would ever be done in an equitable way, though, given the power dynamics of the parties involved.

I don't know; I honestly have never gone to Valve's store page and decided to browse for a game. I know that some -- many! -- people must do this, but I don't have time to play the games that I've extensively researched. To me, the steam storefront is an obnoxious banner-ad that plays beneath the Search Bar. I get most of my game recommendations from either social media or streamers, who are often in a good position to spend a lot of their time researching less-popular, still-fun games.

I quite agree with you, i like human curation greatly, and so found it elsewhere than on steam
But i feel like the festivals they get (next fest, deck builders fest, farming fest etc.) and the indie editors, award events having some too (like the LudaNarraCon, the MAZE. Awards etc.) are good new way (i think it's quite recent but can't remember when it started) for them to have this curation. it's not perfect as it still suffers from the meritocracy you've talked about with better selling games being shown first, but i feel like it's making progress in a good way as i've been discovering a lot of these hidden gems.

in reply to @stu's post:

Valve has exactly two "lean on buttons", which is the one day banner takeover (which is always usually a big title or a huge famous indie game), and the publisher sales/daily deal (which is more "the top 40 publishers gets 1 week of the year").

That 100% does nothing for anyone without a publisher to vouch for them though. It's entirely based on "Is your game bought and played a lot? It goes up the list, with it going higher on people's list who buy this genre", which kinda works but really requires word of mouth.