For games, I’m a pretty thorough pre-purchase researcher. But clothing? Nah. I’ll browse. I think that may be a disconnect between most of us on this board and many Steam shoppers. I get the sense that a lot of them use the front page as their shopping research, mainly due to the way the publicly available sales stats bear out that being on the front page (regardless of a discount) can increase sales by 20% or more.
So there’s also a glut of ebooks on Amazon, for example, many (most) of them also crap. Self published. This is just curiosity on my part, but does Amazon have any better curation via users? Is that also a problem for that market? Or is the perception of Amazon, as a seller of more or less everything under the sun, allow them to escape the criticism we find of Steam?
I’d seen it multiple times, and remember thinking every time that it was a unfortunate name. Didn’t realize it was from the Rogue Legacy team, but honestly I would have taken that as a negative signal anyway. (I might have been into the puzzle aspects, but have zero interest in what appears to be a forced multiplayer brawler game from the people who made one of the worst controlling platformers ever.)
They fucked up so badly with this game that it’d be impossible to draw any general conclusions from it. The strange part is that this article makes it appear as if they learned their lesson. But even months after they realized their marketing approach was bad, their Steam page still doesn’t mention the puzzles.
In my experience Amazon’s recommendation algorithms are light years ahead of Steam’s.
Maybe they should just start copying Netflix’s reco system on Steam, so we can get some awesome goofy subgenres of games.
“Games set in the 40s with robot protagonists that love dogs”
“Strategy titles where you play a vampire in high school”
“Arcade shooters with classical scores and trippy visuals”
And the ludicrously inapt “you might like Cannibal Death Fest 3 because you enjoyed Fluffy Bunnies Being Cute - The Movie” sourcing.
Unless it comes ti big ticket items you just bought.
“Oh you bought a monitor? Here’s a shitload of monitors.”
“No thanks I already have one. You know… from you last week.”
Well yes, it doesn’t seem to understand that some things you don’t need recommended after you’ve purchased one.
The best is when you buy a game from them and they insist on recommending the same game on other systems.
I still contend that the logic behind those recommendations is much more applicable than Steam recommending stuff solely because “it’s popular”, though.
I think anyone who has a clue about Machine Learning knows that Valve doesn’t even try.
Yeah, multiplayer is an additional gamble. Might luck into a big payout ala PUBG or Minecraft, but more likely to drive sales away for fear of empty lobbies.
A family member just got YouTube TV, so obviously I’m glomming off a family account. Anyway, they recently patched in the ability to sort and hide channels. I immediately hid every sports channel, as I hate sports. I have never watched sports. And yet Google, the machine intelligence AI company, continues to recommend sports. Because it’s popular. Sigh.
Well my position remains the same. It is the criticism of Steam and the number of games itself that is in error. There is no games glut, no less of the pie to go round, no indiepocalypse. These are just category mistakes. Games are even more healthy right now than books and books are in good shape. So I am a “problem? what problem?” guy.
Amazon has mildly better curation, but its really not a particularly complicated. But yes Steam should just copy it (it is as simple as “what have you searched for and what did other people who bought this item buy?”, hardly difficult to replicate).
I have no idea why they stick with their own out of date discovery algorithm. Amazon’s maybe mediocre but Steam’s is just poor.
I don’t think the industry is in trouble, but the question isn’t that, it’s how likely any given game is to be profitable, and in particular, profitable enough to let the creator(s) make them for a living. And I would say the number of games has a much higher impact on that than discoverability. Books are actually an excellent example because, sure, it’s healthy insofar as many books come out every year, and many copies are sold. But can you make a living as an author? Chances are, no.
Well thats a different question :) My answer is “for most artists probably not.” Just like painting, film, music, writing, poetry, acting, dance, sculpture etc.
Perhaps the answer to the whole question of this interesting discussion is we just dont firmly put games creators in the artist and entertainers category as we should.
I am reminded of Orson Wells who just laughed at the term “Film Industry”, I will see if I can dig up his quote…
Steam is chock full of absolute garbage on par with the App Store, so if that is a true average of everything available, that sounds pretty good. Median would be a better indicator than average, probably.
I wonder if this is also a problem with (made up word) genre-fication. Some mediums seem to more openly split into subgenres, others do not.
Film, for example, still generally fights with the limitations of Drama, Comedy, Action, Romance, Horror, Thriller, Period, etc. Pretty staid descriptors that are sometimes lacking, but (for whatever reason) still generally applicable across a breadth of releases. I think “Comic book film” may be a subgenre in the making, but that may be a passing thing.
Books and music, on the other hand, divide rapidly into dozens, nay hundreds of subcategories, allowing someone to more easily discover what they may like. Exploring the nuances between types of Metal music, for example, is extremely daunting to an outsider, but someone in that scene understands the breakdown fairly well, and is probably able to quickly explore apples to apples music selections. And in books, you can drill down fairly far - let’s say I like Fantasy Romance Books With Hardcore Magic Rulesystems (I don’t, in fact, like that, I’m just making an example). You can probably find something like that fairly quickly.
Game genre descriptors have been fairly limited - perhaps even moreso than films - probably due to the limitations of the medium until this point. But now, we’re seeing a lot more experimentation in the space due to more easily accessible dev tools, and more voices entering dev, with no easy way to categorize their creations. Steam maybe tried with their “tags”, but that system fuckin sucks, and isn’t really applied well to their storefront.
Yeah, I guess what I’m saying is - they need better discoverability, but maybe some of that needs to come with updating gamer vocabulary. And that’s a risky proposition, given how picky, asinine, and anal some gamers are. But someone better take that bullet.
I think thats a good point. Although genre tags can be limiting they can also help you find what you are looking for. So I can search for “best new SF books 2018” or “Bets new SF movies” and get a bunch of recommendations. For games it is harder as it primarily is focused on mechanics not setting.
Hmm. Interesting point.
Shit, I didn’t even think of the genres going beyond mechanics. My brain was still thinking “Well ‘walking simulator’ started as a derisive term, but maybe it’s a good new genre”, not about setting, etc.
LABELING GAMES = HARD.
Maybe games do need human curation of tags, and then a really smart system for searching\grouping via those. User generated tags are inconsistent. That’s why Netflix’s system - goofy as it is - may be generally somewhat better. All of it is run by humans, but they do choose from a pre-set list of tags. But I doubt Valve wants to hire, like, fifty or so people to play every single game that releases for their platform. But maybe they should.