Patrick gave an interview to The New Scientist when we announced last week. We're a little worried about Starfighter Fatigue and are definitely trying not to get too much onto the front page (Erin's diligently writing the game details post, which we want to be the next thing people see from us).
In the meantime though, if any of you have questions, we're watching and will answer.
What Thomas said. In particular, while we're really glad that many HNers like Starfighter, we were members of this community far before we were co-founders. We're keenly aware that a) HN is a community, not a marketing channel and b) the community frowns on saying the same thing over and over.
I'm CEO of a company and am pot committed to spending the next several years of my life repeating close variations of the same three things to anyone who will listen. ("How is that different from the last eight years, Patrick?" Answer: Previously I only had two things!) We would prefer people exercise their discretion in posting them to HN, to avoid boring anyone here.
Patrick, best of luck to you. With the success of Appointment Reminder (and consulting, and BCC . . .) I was very surprised to see you launching into something new. But your most recent "annual report" (which felt much more personal and vulnerable than usual) makes me glad you're trying to find something that makes you happy and motivated. I'm skeptical about Starfighter, with so many others trying to gamify and "reinvent" hiring, but you folks have the talent and audience to succeed, like Joel Spolsky with Stack Overflow. Thank you for all you've contributed to us programmers (which I still call myself :-)---it has made a tremendous difference in my life and in many others. I hope Starfighter achieves everything you are dreaming for it.
Oh, we do; that's what I spend all my time on. But there must be some reason behind your comment, and I'd be happy to discuss it with you at hn@ycombinator.com if you'd like. (Just not here, though, as the topic at hand is game-driven recruiting.)
dang, "reply" is not available for your reply to A07bei3s - I suggest it should, and not to start that conversation (agree it should be done separately), but in case someone would like to reply to you.
With the speed that things pop on and off the front page these days some times it's nice to see frequent stories about the same thing at least when there is something new to tell. I hadn't heard about the microcorruption site until it was mentioned in the first post about starfighter.
Why does the story say "video game"? I wouldn't call Microcorruption a video game. Was this your word choice or something New Scientist did on their own? Or is Starfighter actually going to be something I would call a video game (instead of what I was expecting which was an experience akin to Microcorruption)?
I would be because for a large number of people, including the people writing the New Scientist story, games played primarily on or with dedicated or general purpose computers are all "video games".
because of that choice, my brain spam filtering just gave up on the story. If not for the comments here, I didn't know that it was about microcorruption.Maybe I should give it a chance.
What will be the official policy for sharing game footage on websites such as Twitch and YouTube? Will you officially support content creators who might want to make something like a Let's Play of Starfighter?
I'm only asking because if it's actually a fun standalone game (independent from the job hunt aspect) this is going happen.
Will you officially support content creators who might want to make something like a Let's Play of Starfighter?
Heck yes. Existence of many Let's Plays is necessary (but not sufficient) to prove we're actually doing our job.
I'm peripherally aware of the mechanics of what I have to do to get Big Daddy G to be OK with this, and it's on my checklist. (Not asked, but the answer to "Are you OK with people monetizing those videos?" is "We have no objection to it but if you are both capable of programming and motivated by Youtube CPMs please let Patrick buy you dinner and explain how many astoundingly better options there are in life.")
Thanks for the reply. The unasked part is exactly where I was going with my question. I should have just stated it explicitly.
> if you are both capable of programming and motivated by Youtube CPMs please let Patrick buy you dinner and explain how many astoundingly better options there are in life.
Ha. I understand your sentiment, but what you've said is pretty condescending. I would agree that creating content for the sole purpose of making "YouTube money" is silly. I would argue that, to the vast majority of YouTubers, it's simply an enjoyable hobby with RPMs as a potential bonus. The RPMs are intrinsically motivating. There is nothing wrong with being honest with yourself and admitting that they play a factor in the motivation needed to create vs. do something else.
Anyways, my point is, if a capable programmer enjoys making monetized videos on the side, the fact that there are "astoundingly better options in life" is irrelevant.
"¿Por que no los dos?" I think the spirit of Patrick's response is "Knock yourself out." And I get the feeling, having followed what he's had to say about how programmers typically negotiate salaries and consulting opportunities, that he's encountered MANY who truthfully don't know the market value of their skills and are subsequently under-paid. I suspect the condescending tone you're picking up is really just him just trying to shed light on a place he's seen members of the community routinely overlook. There's definitely nothing wrong with earning money making videos. But there exists a group of programmers who honestly don't know how much more they could be making, and I think all he's trying to do is let them know.
Please also consider pushing Twitch "Let's Plays" as well. We have a lot of programmers watching!
I agree with what you're saying about video advertising CPMs, though it is quite possible to make a living on Twitch once you add in subscriptions at $5/mo.
How much of the game will be about "learning new skills"? Patrick mentioned at some point that it will be good for that, but I'm not sure how much of a focus that will have vs. testing out skills.
More specifically, are you planning to use this to eventually teach very specific skills (i.e. you'd use Starfighter to learn a specific new programming language?)
P.s. Incredibly excited about this. The last time two heroes of the programming community built something together, we got StackOverflow!
In my opinion you should have kept your mouth shut about the project until you were ready to launch, even if only in beta. Now you get all the marketing, but people can't play the game and have to guess how it works.
I'm not impressed that your submarine PR effort lead with a marketing line / hope, presented as established fact.
That's crossing the line from self-promotion to skeeziness, especially in a publication that calls itself "Scientist", and attracts readers who don't know it doesn't respect the culture of scienctific communication.
You probably ought to direct that concern to The New Scientist, who reached out to Patrick out of the blue. We're not soliciting press coverage. It's just a slow news week, I guess.
This is awesome. I'll definitely become an avid player, right after I contribute meaningfully to several open source projects, have an active GitHub profile, while keeping my Stack Overflow reputation high, and having a fleshed out CodeCombat profile, and of course making sure my LinkedIn page is up to snuff.
I hope when I get this awesome job my employer doesn't mind me keeping all my profiles updated on company hours!
I especially like your point there that this will highlight people who don't have a spectacular Github profile. Lots of people complain that they have a great track record writing proprietary code and the recent focus on Github profiles is unhelpful to them. Hopefully a Starfighter score, compared to a Github profile, is more fun to develop, less intimidating to share, more meaningful when still "unfinished", more digestible by hiring staff, less entangled with employer IP, and perhaps most importantly requires less "vision" to get started.
We're also not an information source for employers; they can't "query" us for all their candidates to get scores. Anything we ever share is opt-in. We expect most of our participants not to opt-in.
Seriously, this might be relevant to fresh out of college grads with no industry experience. But I have nothing to prove and see this kind of stuff as a waste of time. Why waste my efforts on contrived competitions when I can put them into projects that actually interest me.
For me this is definitely entertainment. Having completed μC in lieu of plenty of sleep I'm honestly a little anxious about how this release will affect my productivity. I'm not looking, but if they offer to replace my interesting, very gainful employment with similar then I promise not to hate them whatever I decide.
The idea here is that this will remove the need for pretty much all of that. My read on it is that it's meant to be an entirely new replacement funnel, not an additional one.
They're going to have a tough road ahead of them selling everyone on that idea, both employers and potential employees. But if they get some high profile success stories, it's possible. Hiring is faddish, there's no reason these guys can't leverage that to their benefit.
Personally, I think it's really cool they're trying this. And that the two HN posters I disagree with the most often (maybe because they post so often) are the ones doing a startup I think is really interesting. They've chosen a domain area that can have huge leverage effects on the industry as a whole (and thus the economy as a whole) if it works. I dunno if they're thinking of it like that, or would admit it if they are, but it's one of those frighteningly ambitious ideas.
"There are too many competing standards for x, so we are proud to announce a new standard, which will supersede all existing x standards and unify everything." -The founders of every competing standard ever introduced for anything in the history of things.
I agree with you (and really am looking forward to Starfighter and don't share the GP's), but I'd like to point out that your answer is sort of incorrect. I can launch a stupid idea aimed at, say, world peace. People can reasonably mock my stupid idea, while not being able to find a solution. E.g if I propose stacking rocks in order to reach Mars, a dismissive, snarky, comment may be warranted. Honestly, if it was Monster or Oracle behind this instead of well-known awesome people, I'd be as skeptical as the GP.
This is an especially weird snark to offer on this particular thread, as I have a single semester of liberal arts, and Erin is an audio engineer; only Patrick has a CS degree, and it isn't from Stanford.
No CodeJam victories or at the top of the TopCoder leaderboard? Extensive Coursera certificates? No-hire. Plus I'm beginning to doubt that you've even solved several Project Euler problems in a variety of functional programming languages.
Such a cool idea, and I enjoyed Microcorruption, but ended up getting stuck in the second or third level and couldn't even figure out the right questions to ask to get help. (I stepped into the irc channel at some point and asked some simple question, but got an answer that I couldn't even parse. So I realized I was out of my league and needed to learn more basics first.)
My one concern is about something that was mentioned during the announcement a little while back. The post mentioned something like "employers will be able to see exactly what you did when you had to create a REST service" (or something like that). Maybe I'm just learning this concept, maybe I'm only playing around! Will I have to have one "play" account, and one "real" account? If I knew that I'd be judged by not only whether I complete a level, but every single thing I tried.... Yeah, no thanks.
I'm with you. There's no way I'd play a game where my every move was fair game for employers to scrutinize. That's too much stress and it would suck the fun out of everything. The thought of a game like that irritates me to no end.
This is one reason I pay for a private github. I hate that if I'm learning a new technology, have "too many" commits on something simple, or am generally messing around it looks "bad" to employers.
We agree. We don't think many people would play a game like that. What we're doing doesn't work that way. Employers can't "query" us to get scores for players. If you share something, it's because you want to.
I'd argue that if an employer cares about public repo commits to personal repositories, one is probably better off not working for them (at least from my perspective of an ideal employer). If an employer thinks all their developers are infallible and instantly experts in everything they're learning, that sounds like a pretty awful culture. I know I've made silly/trivial commits to my github and it has yet to hurt my employability. I'd much rather work for a company that knows its employees aren't perfect and development can be a trial/error process at times (these kind of places do exist and I live and work in the Mid Western US).
I do agree somewhat about game scrutinization, since it's active process by an employer verses employers passively looking at your github profile.
>If they care about X, it's better if your don't work there.
Not speaking about private repos per se, but whether it comes down to a close call between you and another candidate, or one person on the hiring team is looking to make a quick decision, it's just smarter to avoid meaningless points of contention.
Sure, it's likely you'll never be hurt by this, but it would be sad to miss out on a great opportunity because of some flotsam... and you'll never know why.
I see this argument a lot and it does have some merit, but it ignores the fact there might be an idiot hiring manager standing between you and the job you want. Just because a company might hire a few bad apples doesn't mean they aren't worth working for, for pretty much the same reason a person can still be worth hiring, even if they make a lot of commits for something simple, etc.
Perhaps, but how many hiring managers are able to scrutinize commits in that great of detail? I'd figure anyone that could, would be in a more technical role and likely a developer/developer manager versus HR.
> have "too many" commits on something simple ... it looks "bad" to employers
Really??
I have frequently, in the line of my consulting work, advised my clients (those self-same employers) to worry when they stop seeing a healthy flow of commits from a developer. (Not necessarily to github, rather to whatever VCS they're using.) One friend - like myself an experience dev. - wired ctrl-S in his IDE to his VCS commit command!
Mistakes, meandering explorations and wandering up entirely wrong paths is all part of software development, and is precisely what your VCS is there to help you manage. The notion that some (potential) employer might consider using it to the hilt as a "bad" thing sounds profoundly worrying to me. Probably not someone I'd want to work for.
I was worried about this too, but when I think about how many times I've shared my Github profile/projects and not had people look at it, I doubt it's a real concern. At most people will look at your score, not at the details. That doesn't mean it won't be in the back of my mind adding anxiety while I'm playing the game though.
I am lulzing a bit that the top Google result for Starfighter starts "Official website for Starfighter, an erotic sci-fi m/m webcomic" Kind of like when Django Unchained came out and it got a lot harder to debug one of my projects...
I like the idea, although personally I wouldn't be excited about being "interviewed" like this. I am supportive of its existence however because it has definite upsides (i'm sure there are more, just enumerating a few):
1) eliminates some cultural/gender/language/fashion/appearance/disability biases
2) its not subjective
3) no difference between local and out-of-town candidates
I'm glad that the toolbox for candidate search is expanding. I hope one day employers might be able to ask candidates which form of "interview" they prefer to have so that candidates can best demonstrate what they excel at in a way that is most suitable to them.
The biggest misconception we've managed to foster about what we're doing is that Starfighter is a form of interview. It isn't. At all. A better way to think about it is that it's a sort of Venn intersection between "Github resume", "Coursera course", "Dwarf Fortress", and "recruiting firm".
Our crusade against the tech job interview is roughly analogous to Google's "organizing all of the world's information". It's a mission statement, but it isn't a description of our first offering.
That said: the technology we're working on ports beautifully to actual interview processes (it came from a successful process I built at Matasano). We're also very happy to talk to companies building hiring processes and evangelize and help. Companies with smart, modern hiring processes are especially fertile ground for what we're doing, so we'd like to help more of them come into being!
I thought I recognized this. I wanted to work for Matasano at one point but found the CTF environment incredibly difficult if you come from a Web Dev side of things. Hopefully starfighter is different.
That's too bad, because the Matasano hiring process is actually not at all CTF-focused, and is very web-dev centric. If you're ever looking again, you should reach out to them.
Will Starfighter be focusing mostly on security? That seems to be most of the examples that I find, and it's not a field I'm terribly interested in. I love the idea though, and would like to see it applied to other areas.
Security is the through-line of the game; it gives us structure, incentives, little dopamine hits, and makes it so that the systems we're building reward exploration. But the game is not about security; it's about programming and interacting with complex systems.
If you're a software developer with no security domain expertise, one benefit of goofing around with Starfighter is that you'll come out knowing a lot more about software security. But it's not a tool for getting people security jobs.
It'll make a little more sense when Erin posts the game details.
We will not primarily be a security company. It's a fun and interesting field(s), but it's a small subset of engineering skill, and we plan to be substantially broader in focus.
I am super excited for these gaming hiring funnels to take off due to your first reason. Dollars to donuts there's a ton of unidentified talent out there that is being filtered out due to those biases (I would also add in academic pedigree bias which is correlated with economic/cultural upbringing). Granted these types of funnels also select for those who have free time to play games so they are not perfect.
Knack, the other company mentioned in the article, seems to have profiles with pictures and names that I assume are visible to hiring managers and others. In my opinion that's a bad move. I want to see a completely objective view of how a candidate is performing without age/experience/appearance getting in the way. These gaming funnels aren't perfect solutions but they are certainly a step in the right direction.
Would this be a good way to teach my son technical skills? He's 6 years old, homeschooled and is interested in technology. He's already figured out how to bipass the parental controls on his Kindle. We're trying to figure out how to stimulate that desire when you can't read well yet.
Geek to geek and parent to parent: honestly, no. Can I recommend Besiege? It doesn't require literacy. It has the same sort of freeform expressiveness you're allowed to do in Minecraft. "Syntax" errors are inherently fun rather than demotivating. There's fun, rewarding visual feedback for success.
It doesn't teach programming qua programming, but it is almost perfect for a gateway drug into engineering. (You get to build catapults! And airplanes! And catapults which fire airplanes!)
I was going to suggest Robot Odyssey. That was my first introduction to digital logic, although I didn't realize it till years later, when I actually studied some digital logic.
There used to be a clone in java called Droid Quest that would run without an emulator, but I went looking for it a little while ago, and the web site appears to be empty. The only remains I found is a github repo with the source code, but I haven't tried to see if I can build and run it yet.
Having worked on an applicant testing and tracking system in the past, it was a decent business case to operate because a single company has a job type with 100's of openings and high attrition so it helped screen people. In that case it is self-serving and you know if someone is applying twice (SS Number or other identifier) to try and ace the test.
If Starfighter is open, there are people out there that will try to game the system ("game the game" doesn't sound right.) If someone wants to be a level 50 app_coder, they could have a pre-made sequence to follow to do that. Much like in a MMORPG where you have helpers level up a character or a "mule" for hand-me-downs.
Interviewing candidates last year showed no hesitation on their part to share info with each other or their recruiters. How will Starfighter prevent player "cheating" or knowing the route to leveling? How will content be updated or dynamic?
This is a promising thing for software geeks in general, can't wait to try it out. I started playing Microcorruption and if this is similar to that, it will be fun.
This is my major concern, but they've said that it's easier to learn the skills than cheat. I'm don't know how, but it's something they're obviously concerned about.
> Of 12224 players, just 182 passed the hardest level. The firm will get in touch with these elite players and help place them with one of their clients, who pay Starfighter a fee.
I've played a few rounds of this and love it. As far as I've gotten, it's been refreshing to work on a machine where everything is "all there". Too often there are black boxes that contain too much magic, or just too many layers of abstraction to efficiently reason about something.
Companies which are satisfied with their ability to hire and employees who are satisfied with their ability to find jobs don't have to even care we exist, though they're certainly welcome to play our games.
Sadly, many companies are dissatisfied with their ability to hire and many qualified engineers are dissatisfied with their ability to secure desirable jobs. In economics, they call this a market failure. In CS, a search problem. In business, "a shedload of money available for the taking."
In Starfighter, I guess we categorize that as "the mid-boss."
More like a unicorn 10X Developer shortage. Companies will be pleased to find candidates who are eager and willing to jump through yet another proving ground, after having had so much fun with commit histories and stars on GitHub, SO points, and what have you[1]. They'll find smart candidates who are willing to work for them in exchange for being able to work on Interesting Problems and the occasional you-are-so-smart implicit approval.
[1] Meet the new standard; same as the old standard, and in addition to it.
Great idea. I loved Microcorruption. But please please don't call it video game! Average person reading this would think this is some sort of World Of Warcraft type game and stuff like that but this is more of a programming competition.
While the efforts of Matasano to make wargaming cool and hip again certainly is commendable, I think HR drones of the world also appreciate a game of "pretend to be sociable teamplayer/brogrammer".
Too often hackers don't get the job because of being overqualified and prima-donna, rather than lack of technical skill - a game teaching/testing the art of office politics could be applicable to wider market [outside of infosec].
I know this article is referring to a different kind of gaming, but back in the day, I used to give potential candidates a NES controller and Tetris when they were in the waiting room before the interview... You can tell a lot about how a person deals with time constraints, pressure and approaches problem solving by watching somebody play tetris for 4 minutes. Not that Tetris would be the only input into the decision, but it was a thing.
I've learned to be cautious whenever anybody says "You can tell a lot about a person by..."
You probably can't. You can probably justify a lot of preconceptions about a person. But unless you've done statistical studies that demonstrate high correlation between what you're testing, what you think you're observing, and what you're actually looking for, then no - no you can't tell a lot.
Young male candidate, lets a big stack build up on one side, then clears the lot down with a couple of well-placed four-long blocks down the edge. Shows an easy confidence, clearly worth considering for the role.
Female candidate, lets a big stack build up on one side, what is she doing? Does she even know how this game works? Then she drops a four-long down the side and clears some rows. Huh - lucky break.
"Not that Tetris would be the only input into the decision, but it was a thing."
My sci-fi scale projection/evolution of this concept: because the things we do at work aren't intrinsically unenjoyable to begin with, but the way we think about them has the potential to make them feel that way—slow fusion of games that resemble work and work itself produces games that do work, and the distinction between the two is forgotten, a relic from a cruder era.
Seems possible since the activity in games (for all the examples I can think of) has a pretty strong relation to 'work' activities to begin with: puzzle solving, mastering agility-dependent procedures, military stuff, social finesse, exploring etc. —though typically it's more 'primordial work' (exploring, hunting/gathering), which makes sense since atm we'd be putt off by something that resembled contemporary jobs too much. Dislike of one's work is not pervasive in human cultures/history.
>My sci-fi scale projection/evolution of this concept: because the things we do at work aren't intrinsically unenjoyable to begin with, but the way we think about them has the potential to make them feel that way—slow fusion of games that resemble work and work itself produces games that do work, and the distinction between the two is forgotten, a relic from a cruder era.
There was an episode of Black Mirror that hinted at this.
I'm not very familiar with this style of programming puzzle, but I wonder how this is less arbitrary than whiteboard coding topcoder or kaggle. I know people who were hired based on all of these, but the latter two are different from Starfighter in that they exist for their own reasons (competition for its own sake, and getting answers to data science questions) with providing a signal to employers being a side effect. I wonder if the selection bias in a system set up for the purpose of employment will be better or worse.
It also only works if you can get a lot of employers to go along with it.
There isn't much point spending a lot of time on something if it's only going to get you an interview or two (if even that). And if I'm looking for a game I still have a lot of stuff in my Steam account from bundles that I haven't had time to play.
In the meantime though, if any of you have questions, we're watching and will answer.