The Valve employee handbook [0] is an interesting read. Very unique company structure, even considering the relative penchant for anarchy in the tech field. The motivating factor seems to be preserving independent thinking and creativity. It's also worth noting that Valve is not publically traded, it seems to resemble a worker cooperative.
Not sure why you were downvoted. The handbook is an interesting read indeed.
You could maybe interpret it as shameless advertising on the part of the company (which it is!), and you could question their famous flat structure (classical point of critique towards these kinds of structures is: if there is no formal power structure, informal power structures emerge; and they can sometimes be more ominous.)
A lot of the time employee handbooks preach one thing, but reality is often quite different. Another good example of this is "Netflix's Culture" slides. They are often lauded as cutting edge ways of thinking about company culture, but are actually a load of bull that no one at Netflix really adheres to, and the benefits touted do not even apply to all employees.
I do believe her section of Valve was probably shut down based on the success of Oculus, at the time this article was posted on HN this did not seem obvious to most people. I'm sure it was a difficult decision for everyone involved.
From Ken Birdwell - the author - linkedin summary [1]:
Valve is the place to go as the culmination of a career. It's filled with 90% senior people, people who have run their own companies, not just projects or products. It's not uncommon to meet people here who have been here "only" 10 years. We lose fewer than 1 person every 2 years, and that's been true over the entire history of the company. Once you're here, you'll stay forever, there's no reason to ever leave. We're also a family oriented company. You're expected to work 40 hours or so a week. Maybe a 3-4 weeks of crunch time a year, but it's all voluntary. We'll actually send you home if we think you're spending too much time away from your real life. Vacation time isn't tracked. If you need to take time off, take time off. We don't care, it's up to you. Your colleagues are keeping track of your contribution anyway, we don't need some stupid accounting trick to know if you're valuable. On top of that, our schedule is our own. We have no outside investors. We own all of our IP. We have no "publisher". We do work with external folks to make and ship retail boxes for us, and they do a great job, but they have zero influence over what we do or when it's done. It's totally up to us. We're in control of our own success or failure. It's ours to make. Why anyone would want to work anywhere else is a mystery to me.
"Nothing is quite so humbling as being forced to watch in silence as some poor play-tester stumbles around your level for 20 minutes, unable to figure out the "obvious" answer that you now realize is completely arbitrary and impossible to figure out."
If you want to build software that is easy to use, you need to watch people use your software.
It's really interesting to compare and contrast the Cabal method and the idea of design by committee[1]. This must be one of the very few stories I've heard of where a committee with a rigidly defined structure actually succeeds at pulling off something creative.
The bits that really struck me were this:
During Cabal sessions, everyone contributed but we found that not everyone contributed everyday. The meetings were grueling, and we came to almost expect that about half of the group would find themselves sitting through two or three meetings with no ideas at all, then suddenly see a direction that no one else saw and be the main contributor for the remainder of the week.
and this:
The Cabal consisted only of people that had actual shipping components in the game; there were no dedicated designers. Every member of the Cabal was someone with the responsibility of actually doing the work that their design specified, or at least had the ability to do it if need be.
So, focusing on the quality rather than quantity of contributions makes sure that people don't just chip in ideas for the sake of chipping in. Also, making sure that the people on the committee are those who personally have to implement these ideas makes sure that everything is actually achievable.
Player can interact with the game world, destroy things (many small items), the environment remembers your action (e.g. opposite-sex bathrooms in Deus Ex 1 and No One Lives Forever 1), some funny moments, ...
IMHO bad examples are Deus Ex 3 with its very static game world, Assassin's Creed with its limited interaction options, many role playing games like Gothic series, Elder Scroll series (Morrowind, Oblivion, etc) were you can steal from all NPCs and their chests and no one asks/follows you, Call of Duty series with their unlimited AI enemies(if you stay and wait), etc.
One important point highlighted in the article is that there's very little if at all "group creativity". A single good idea is like an atom, it's concieved induvidually, from someone's previous experience, from a single person's background.
What this remarkable structure does is create a way of consistently filtering ideas, and generating many of them consistently -- all while integrating them towards a bigger picture. The group seems to be just large enough so that someone is being creative all the time while not too many are generating noise.
In other words, it highlights every creator not as a cog generating an enormous structure but it identifies individuals with pieces of work, with full responsibility. Every piece of art or tech is someone's pet and it gives people the opportunity to nurture them and see it succeed.
(Although I wish I worked at Valve to be able speak from experience...)
The Cabal met four days a week, six hours a day for five months straight, and then on and off until the end of the project. The meetings were only six hours a day, because after six hours everyone was emotionally and physically drained. The people involved weren’t really able to do any other work during that time, other than read e-mail and write up their daily notes.
I find this a staggering amount of work for the planning stage of a game. Is there anything like this amount of planning at games outside of Valve?
The bit about them deciding to take all the cool elements they liked from v1 and fit them coherently into v2 was interesting ("Oddly enough, when trying to match these artificial constants, we would often create our best work"). It's funny how constraints often inspire the greatest creativity.
"The more constraints one imposes, the more one frees oneself of the chains that shackle the spirit... the arbitrariness of the constraint only serves to obtain precision of execution."
- Igor Stravinsky
Nice quote; here's my attempt to put it in a less of a "hazy 19th century philosophy" kind of way: "The more constraints one imposes, the narrower the decision tree and smaller the design space, a focus which enables quick enumeration detailed search of such design space... making it therein less costly to obtain good results instead of aimless exploration of a space of "infinite" possibilities"
If you like this type of content, check out the Reddit subforum http://www.reddit.com/r/TheMakingOfGames/ , which covers behind the scenes footage and 'making of' material for video games. There's an alphabetical listing of all submissions that you can use to search for a particular title.
Press CTRL-F on that page and try searching for your favorite game. Searching for 'Valve' will lead you to the numerous developer commentaries that Valve has included in their games, similar to cast commentaries on DVDs.
Valve founder Gabe Newell gave a talk about productivity, economics and corporations that you might find of interest:
Kojima used the same sort of stuff to make metal gear solid 2, they gathered sketches around and debated what ideas were good or not.
I still strongly believe ideas should be gathered around and discussed. There was a voting website about gaming ideas once, it's gone now but this was exactly what made half life not only a very interesting gamedev platform, but a pillar of how successful games should be made.
Of course it's not easy to make people work in those conditions, because it's always asking people to make compromises, but if you can make it work it's obviously a recipe for success.
My only wish for a game company is to do that on a game with a long term cycle, like blizzard game usually are. That was a game can be constantly improved and worked on.
The article says at the end
> Engineers are really slow.
That's why I think that sometimes, a game engine need to freeze its hardware requirements for a while, and let developper improve the game as time pass. 3D engines are hard, and most of the time, when a 3D engine upgrade is planned, all previous work is dumped, and engineers start from scratch, again. It's tricky for a game business model, but I wish game were developed like google release new products in an endless beta stage.
You're essentially describing a console: a system with fixed hardware requirements that will last for several years. That's the ultimate advantage of being a console dev - you have no variation in hardware whatsoever* and if your game works in the studio, you can be 99% sure it will work on any identical console in the field.
And generally throughout the lifespan of the hardware, the games get progressively more advanced as the developers learn how to fiddle the hardware to do what they want.]
*not totally true, but ignoring things like disc space and peripheral availability (like the SNES coprocessors - even then they were in-cart!) all consoles can be considered equal.
There are a lot of reasons and I think it's time to write them up, but I think HL2 was actually over-polished. They didn't follow their own rules, either, leading to a lot of "Uncanny Valley" situations.
I wonder if we will ever see Valve recognize that HL2 was not so great, since it won the "best game of the millenia" awards all over the place when it was out (and it was way over-hyped even at the time).
What I'd really like to see is a sample of how they managed minutes for their brainstorming sessions, and the same for the final design document, because I always on the look for how people can capture this kind of complex information in a way that is also useful to share with others (that weren't present during the discussion of the features).
Anyone knows where/if this is available somewhere online?
[0]http://assets.sbnation.com/assets/1074301/Valve_Handbook_Low...