> Valve employed just 79 people for Steam, which is one of the most influential gaming storefronts on the planet.
It might just be me, but that actually seems pretty reasonable.
What I've seen a lot is companies, during growth, accepting less-than-ideal candidates, maybe without a good onboarding, which then end up as deadweight in the worst case, and underperformers in the best case.
This, combined with a tendency to keep people around who have been there a long time, and the people themselves knowing they're not going to be let go, could result in massive companies and hundreds of people per product.
Yes steam is big, but it's not a product so complex you'd need over 100 people to maintain it. I assume other companies need so many people because tasks don't get done, and the logical step is to add people, not remove people.
Staying lean means staying agile (in some definition of the word), and that can speed you up.
> steam is big, but it's not a product so complex you'd need over 100 people to maintain it
I dunno, I think Steam is pretty complex:
- The steam store (credit card processing, optimizing search, review bombs, gift buying, features around various ways to hide games you've purchased, marketplace)
According to the article, Valve is in fact currently maintaining all those features with a team of ~80 people. So this is a good opportunity to improve your understanding of the number of people it takes to maintain software.
One thing Valve seems to understand better than most big tech companies is that once a product/feature is done, you don't need the same size team to maintain it. A lot of parts of Steam have been functionally unchanged for the last few years and just... keep working. The people who worked on Cloud Saves have presumably moved on to something else, instead of entering a Google-esque cycle of pointless Cloud Saves feature launches and deprecations.
Oh, at my day job we are about 5 engineers maintaining and adding features to support thousands of users, I'm aware of how few people are actually needed to maintain and build software.
My objection was to the phrase "Steam is not complex", which I guess wasn't the posters point.
I've been joking that Steam on macOS is so bad that there must be no one at Valve who has used it because there's no way you ship such brazenly slow software knowingly, but with 79 people this is actually probably true.
I've never seen these companies (that are supposedly everywhere) that have all this "deadweight". Every company I've worked at has had 4X to 10X more work than enough staff to do the work. We beg for headcount to hire more so we can actually do what we set out to do, but we never have enough. I guess these places exist, but I've never really seen it myself.
Compare it to Uber, which employs 30k people and they "just" mediate transport contracts and still thousands of those are engineers. Or Airbnb which is 5k+ employees I believe.
Steam is another data point that convinces me that public ownership of companies leads to quick and ruthless enshitification and that growth is only a good metric in specific contexts.
Ride share and delivery app companies could take some notes on how to stay lean and profitable while operating an online marketplace. I know ride share apps are more complicated, but I don't think we're talking two orders of magnitudes more complicated (300 employees vs. 30,000 at Uber). The bloat at Uber/Lyft/Doordash is insane.
The whole idea is to attract market share to show investors that big things are happening. I don't think being pointlessly unprofitable is attractive to investors...
One involves delivering software to license holders.
The other has to deal with the possibility of humans using their app as an assault or kidnapping mechanism; from either end (rider attacking driver; driver kidnapping passenger, etc.)
Very different levels of responsibility. This can be seen, for example, by the size of Lyft and Uber’s security teams.
that doesn't even cover the salary of all the obama admin people uber hired. let alone the actual lobbying going on. that's just what fall under some specific tax code.
If they're breaking things down as is usual then "Admin" includes, among other jobs, the entire customer support team, which I can vouch for as exceptionally good by the standards of internet platforms.
Redaction is not simple.
What should the software do if the user's black box is drawn over an image, some text and part of a spline on a page? Modify the image and recompress it even if it's lossy?, find the text run and remove the obscured characters (what about a half covered letter?), somehow modify spline and possibly replace it with several smaller ones if the redaction is in the middle of it? All while taking into account any transforms that have been applied to the page. Also what if the black box is over part of a pdf form? Is it just the appearance that gets redacted or the form data (which is separate). Same questions if the black box is drawn over annotations. And what about other page content like embedded videos or 3d content, good luck redacting those. But if you only redact some things the user might then justifiably expect it to redact everything.
I can empathize with the feeling but this is the case in many many contexts. Reading the intent of a user without a clear mental model of software is difficult, sometimes impossible.
I know no os that changes the format of an image when a user changes the file extension, for example, which is a pretty similar situation.
Another inexcusable sharp edge of computing that nobody cares about. I propose we recognize it for what it is - a sharp edge, and don’t act like people are stupid for doing what seems intuitive and logical.
I know at least Preview on MacOS has this feature now: if you try to put a black bar over text it tells you this isn’t secure and offers to redact for you using a method that both visually replaces all characters and removes the embedded text underneath It is non-destructive until saved meaning you have a window in which to make changes before the data is gone.
It's not just the people designing the hardware but also those maintaining it. I work in an "Infrastructure Engineer" position (I hate that since it also correlates with construction but mine is strictly expensive servers in data centers) and I make less money than the people running the virtual infrastructure even though that all runs on top of the machines I manage, lol.
I once had an internal customer open an on call event to ask why one of their machines was running so slowly. I said "it's because one of the DIMMs has thrown about 30,000 correctable errors within the past month". I was able to correlate that by mapping the EDAC label for the DIMM recorded in /var/log/messages and some gzipped archives of the aforementioned log file.
Of course I deal with CPU, memory, motherboards, GPUs, add-on NICs (OCP or PCIe), storage controllers (HBAs mostly, some RAID controllers), BMCs, and of course I also have to evaluate the link width of PCIe bridges interconnecting all the PCIe devices.
Steam might have a very disproportionate market share, but no one complains a lot because the competition is downright laughable. Internet companies are very low maintenance
It has a very low barrier to entry for devs and a permissive DRM compared to other stores. A powerful review system and a return policy better than anyone else I've seen so far. Very community friendly and has an entire feature dedicated to hosting user mods and creations. Right now they're wasting money on VR just because they can
Is it a monopoly when no one else is able to compete even on the technical side?
While Steam did have a rocky launch, the competitor stores since launched are laughably bad. Seemingly no attention to detail in making a product people would want to use. Epic has a huge war chest from its successes, but the application is still not at feature parity with Steam.
Is there a reason why game developers can't just sell a key on their website using Shopify/WooCommerce/etc. and distribute the game files publicly via a CDN?
Assuming they can get their game discovered via some sort of non-store channel (e.g. Reddit, social media, YouTube, Twitch, ads, etc.), the user can simply search the web for the game, find the website, click "buy" and proceed.
Probably a bit more work than publishing on Steam but not that much. Seems worth it for large budget games or games that are very popular in a specific niche discoverable outside Steam.
>Is there a reason why game developers can't just sell a key on the web using Shopify/WooCommerce/etc. and distribute the game files publicly via a CDN?
It's not that easy; Steam brings down dramatically complexity of self distributing and it offers a plenty of Web 2.0 features e.g. reviews, discussions forum, connecting with friends, playing with friends, chatting with friends etc.
It's like asking why Amazon exists when sellers can have their own store page and distribution channel....
Some games that I know that are popular and not on Steam are for example Starsector and Escape from Tarkov. Game devs choose not to publish their game on Steam either because their game is not finished and they do not want to get negative reviewed to the ground and/or they think their game's brand is so strong that they do not want to pay 30% tax to Steam. For example Minecraft most likely wasn't on Steam because of aforementioned reasons.
Some devs do essentially do that in addition to Steam. You're missing a million and one features on top of simply running the game itself if you restrict yourself to that, though.
Well, since you mentioned DRM and return policy, let's not forget that GoG has no DRM, and unlike GoG's 30 day return policy Steam didn't even have one until Australia threatened to ban them.
And the Workshop and Steam MP might be very convenient for some, but are walled garden features that ought to be boycotted.
Microsoft now has several games available to download for Windows, a far cry from the size of Steam's library, but it exists. If you have a Game Pass subscription, many of those games are available to install on Windows as well.
Microsoft also supports streaming, both cloud streaming and locally from an Xbox.
Yeah, I agree that Games for Windows Live was a joke. But Microsoft today is certainly capable of bringing competition, with the right leadership.
Where does all the money go if it isn't being reinvested into the business? Are the massive profits immediately syphoned out by the execs and owners?
It's interesting there isn't a well-known real competitor to Steam (the Blizzard and Ubisoft stores, etc, don't count, because so few external parties publish there).
My gut says the lack of reinvestment could be someone else's opportunity, especially considering many of the longstanding VAC and other quality issues which have persisted on Steam's platform for more than 15 years. However, because of all the necessary catch-up, such an enterprise would be tricky to bootstrap and get to critical mass. Even then, where is the moat?
There are plenty of competing stores, all of which pales in comparison. I don't know what their secret sauce is, but for me personally what makes a difference and why I keep preferring them is their dedication to Linux as a platform and a general feeling that they are not just driving me towards rent seeking. I know that last bit is slightly absurd, and I can't quite explain, but I think the fact that they are private means their priorities don't just exists as quarterly goals.
> but for me personally what makes a difference and why I keep preferring them is their dedication to Linux as a platform and a general feeling that they are not just driving me towards rent seeking
For what it's worth, I have the exact same feelings. They provide so much value on-top of just being a store, and it does feel like they take the money and re-invest it into their ecosystem to make it better.
They made gaming on Linux as easy as gaming on windows. They improved the controller situation in games. They do the whole remote-play stuff so you can play local-coop games online. They do the family sharing stuff. They do the steam link stuff. And they don't lock any of it down to their first party offerings. e.g.
* They could have easily made steamlink only work with their set-top box, but they just made it an android app.
* They could have limited the controller bindings and tech to only their controller, but they made it work for all controllers.
* They could have forced you to use their VR headset with steam, but they let you use any headset.
I can't think of any other company off the top of my head that has done something similar.
Steam is actually made with their users in mind. So much that they managed to beat piracy. It is no easy feat, you can't beat piracy on price, and it is hard to beat in convenience when all you have to do is download a torrent and run an exe file. Piracy is like the final boss of software distribution platforms, but they beat it.
To play a game, select it on the store, pay (with all sorts of payment options), download (from damn fast servers), and press play, that's all. The game sucks or crash, click a button and you are refunded, no need to be always online just for DRM check (eventually you'll need internet, but you have enough leeway so that it is rarely a problem), ads are very unobtrusive and maybe even desirable (suggesting games you may actually like or announcing sales), they don't usually force an update just before you want to play a game (a much too common occurrence), dark patterns kept to a minimum (or at least subtle enough), generally good software quality (in terms of bugs, broken UI/UX, etc...). You bought a game, it stays in your account, Steam has 20 years of history of not screwing you, and to play it, just install Steam on any PC (it it doesn't already have it), log in to your account, download and play, it is that easy. You can even share games (with some reasonable limitations).
Others just have some annoyance here and there. Some always-on DRM, bugs, dubious UI choices, annoying ads, etc... Only Valve seems to understand the importance of not getting in the way between gamers and their games and do the appropriate amount of effort (which is a lot). Not even in the name of profit, because they understand that it is not worth it, if gamers are satisfied, profits will come. Remember, they have piracy as a competitor, and it is hard to compromise against such a powerful competitor.
My guess is that the secret sauce is network effect. People have all their games in their steam libraries, so they want the next game in their steam library too.
They perceive the Steam launcher as something positive that provides them convenience. The network effect also makes it hard for game companies not not be on Steam, allowing Valve to enforce some customer-friendly policies like refunds, further making the platform nice to use.
Epic is trying to break through this by giving away endless free games. All other stores are generally forced upon the player through exclusives (you buy the game on Steam, but you still have to install their shitty launcher-store because the parent company owning the game wants some of the store-platform cake), breeding resentment and further cementing Steam's position.
(GOG being an exception, but they don't have enough major/interesting games to get players to use them as their primary platform.)
The problem is that Steam for all its faults is DRM, anti-cheat, and a well functioning store rolled all in one. Most of the alternate stores are bad in one way or another; Epic launched without a cart, and it still doesn’t have user text reviews. Itch had to redo its whole UI after that one bundle they did that had thousands of games in it.
Notably Valve doesn’t really use its position to make competing games. This is an unknown for most publishers, and Epic pretty blatantly pivoted Fortnite to eat PUBG’s lunch after working with them.
If anything the leading platform for competition as an indie store is the Nintendo eShop. Which is probably why the Steam Deck exists.
It is one of the most serious, in that it exists, and probably has one of the biggest catalogs next to Steam, which is not true of say whatever Ubisoft or EA is doing. at the very least there is substantial enough catalog overlap that one can price compare.
There's also Nutaku that's supposedly as big as Steam (they are focused on NSFW games).
Imo Epic is the only competitor to Steam that tries to offer a similar service to Steam. It's just popular to hate on it among pc gamers.
Many people cite a lack of features on Epic's store as the reason, eg no forum feature, but then turn around and decry Steam forums as the worst place on the internet. I don't think the problem is a lack of features (for players, devs have a point), but rather the Epic Games Store isn't cool enough.
Epic games seems to lack even basic functionality. I just added Win11 as triple boot on one computer, the kids have loads of games already installed from win10 in a tertiary drive ... on the website it says the solution to them not having a facility to tell their installer/launcher which folder to use is to reinstall all the games.
Now, once you open Epic you can ignore it, use Explorer open the game folders and launch the games and they run. So it looks like they literally need a way to change a config file to avoid re-downloading 100s of GB of games.
I find it literally crazy. They even scatter .egstore files around so actually scanning for and adding games to the library automatically should be easy.
Steam makes the same task easy, and why wouldn't it it's pretty rudimentary for an installer/launcher to have a easy way for you to tell it where your downloaded games are.
Oh, and the instructions for reinstalling Epic _launcher_ say to just re- download all your games. They're insane.
>Steam for example, allows user submitted ratings with explainer text. This is huge for online shopping decisions.
I don't know anyone who actually relies on Steam ratings/reviews to make purchasing decisions. I've always assumed that rating systems on the platform that is selling you the product are not trustworthy. But then again, I find reviews for media to generally be useless.
>Steam Workshop is easy to integrate mod distribution for games.
That's a developer feature rather than a player one though.
Steam reviews can be useful depending on their content and their general direction, particularly if they mention specific things they do or don’t like, and there are consistent things popping up. Game devs are not flush with money and I’ve never seen a flood of suspiciously positive reviews.
Steam also has the features to do social proof of reviews. The reviews contain metadata about if the person got the game for free, and if you click the profile associated with it, you can see all the games a user has played and their achievement progress, which can help signal if they are a real gamer, how far they've gotten in the game they reviewed, but also how far they get in other games they've played to check that they're not a sock puppet.
This has enough value that Epic displays a star rating that is allegedly user generated, but because it doesn't indicate who is doing the reviews, and there is no text, their star rating is a lot more suspect.
---
Modding support is also a player feature. Official, well supported modding can make or break a game, and Steam facilitates quite a lot in that regard; Nexus, as an alternative, is ad supported and caps download speeds for mods unless you shell out for a premium membership.
> Steam for example, allows user submitted ratings with explainer text. This is huge for online shopping decisions.
Arguably, it requires an account, and has things like "number of hours played" included in the review.
Far, far more useful than the easily-gamed spam reviews on places like Amazon. Doesn't mean spammy things can't happen, and the aggregate reviews on Steam usually mean what the 14-24 year old demo think
They've released five games in the past 4 years[0] and have another one in active closed testing[1], along with keeping DOTA2 running with major game changes and mini-games[2].
[2] DOTA's had massive patches recently, including adding a new attribute, expanding the map, and redoing every hero to have innate abilities and a choice of passive effects. They've also been adding some pretty big PvE/external content events between Aghanim's Labyrinth and the current Crownfall battle pass (which just added a surprisingly fleshed out fighting game mini-game).
Dota 2 is a prime example of enshifitication of modern games where content is added for the sake of content and regards for original style are completely ignored.
Ironically, Dota 2 would benefit as a game if they STOPPED developing it and focused just on stability of the game, but alas.
They basically have. Half Life has been Open Source for a while, Portal and Half Life 2 were both more-or-less open along with the Source SDK. Team Fortress 2 and CS:GO aren't technically open, but their source does exist in the wild and can even be built on modern machines.
That still leaves DOTA, Artifact and Half Life: Alyx without any proper source code, but the stuff people actually care about has mostly been freed.
It might just be me, but that actually seems pretty reasonable.
What I've seen a lot is companies, during growth, accepting less-than-ideal candidates, maybe without a good onboarding, which then end up as deadweight in the worst case, and underperformers in the best case.
This, combined with a tendency to keep people around who have been there a long time, and the people themselves knowing they're not going to be let go, could result in massive companies and hundreds of people per product.
Yes steam is big, but it's not a product so complex you'd need over 100 people to maintain it. I assume other companies need so many people because tasks don't get done, and the logical step is to add people, not remove people.
Staying lean means staying agile (in some definition of the word), and that can speed you up.