Hackathon Be Gone
brianchang.info
brianchang.info
We have some simple rules:
* you can't start anything new;
* you have to celebrate something being done - usually with a round of applause;
* there are no prizes - only the satisfaction of being done.
While there is plenty of coding, we've had designers finish their portfolios, screenwriters finish pilots, an artist finish his display, and even saw a sewing project completed.
And to top it off, we charge a nominal $20 for the weekend - with 4 meals, a great deal - and sell out every time.
We're putting together a howto doc to franchise the idea.. just need to finish it. ;)
I could really use something like this, though I feel like 20$ of your food is a lot better than 20$ of our food (I'm in Montreal).
(Already spoke with the organizers about this in another thread. I think they need someone who's organized an event of similar scale. That's not me, but I'm very willing to help out anyone who can take a lead role)
They make people hypercompetitive and unwilling to collaborate or help just so they can get their `sponsor_api` mixed with `sponsor_api` app done. I'm making a big generalization but IMO sponsored hackathons cultivate some of the worst aspects of our culture.
the sale pitch have more value, than the code written to build the app
Can you point to some empirical evidence to the contrary?
Edit: I had some more thoughts about this, specifically that there is a survivorship bias to legacy code. Not all projects survive, but when they do that code can live a long life of maintenance. If you are involved in the startup world, it may make more sense to get it done quick & dirty in order to get funding than to write "good" code. That might explain the disparity of our experiences.
My experience is that efforts by previous developers are often counterproductive (they designed for adaptability, but in the wrong direction), and certainly that designing for adaptability in general is less efficient than designing for specific adaptations for those specific adaptations (and therefore it's more efficient to defer work to the point where you know exactly which adaptations you want to make).
> Can you point to some empirical evidence to the contrary?
No - just my own experience (I was using empirical in a negative sense - contingent is perhaps closer to what I meant).
> Edit: I had some more thoughts about this, specifically that there is a survivorship bias to legacy code. Not all projects survive, but when they do that code can live a long life of maintenance. If you are involved in the startup world, it may make more sense to get it done quick & dirty in order to get funding than to write "good" code. That might explain the disparity of our experiences.
Indeed. My experience is that probably >50% of code dies before ever being used productively, and so any designing for maintenance before the code has been demonstrated to provide business value is premature.
Mhacks lists the 50/50 gender divide on their site followed by the reasoning: "because it’s about time for a little change in the tech world"
Disclosure: I am a CS-Eng student at this university. I did not plan to attend or care about this hackathon.
I understand that there is a lot of cultural baggage to the various categories but I can't count how many times I've been genuinely surprised by Americans identifying as black or Latina/o. The most striking ones I think were Halle Berry and Obama -- it took me a while to process that their recognition was genuinely noteworthy in the US because of their ethnicity. You could have told me Obama were Moroccan or Halle Berry were Indian and I wouldn't have been any more surprised.
For context: I'm German, grew up in the 90s in a city with ~1M residents and am literally colorblind (partially green blind). I'm not sure what part of that makes me unable to properly racially categorize people but I'm not convinced that's a bad thing.
You can tell from a single comment?
Blind meritocracies are rarely blind enough, and end up being an old boys club.
If I did the above, and someone made a stink, I'd hit 'em with "Are you denying my lived experience?" and then reaffirm the validity of my self identity (which is what matters) over what I actually am to observers (which is immaterial and doesn't matter).
- "Well, but they are for girls only."
- "Do you have something against cross-dressing?"
- "Nope. Sure, take one!"
It's neither inherently good nor bad, it's a chance to create with some new individuals over a short time period. Yes, it gets co-opted by corporate marketing, yes you eat pizza for a day or so. But these are minor issues if you find other benefit like learning of new processes, tools, technologies, etc.
More than anything, it just seems inherently "uncool" at this point, a proverbial pet rock for developers. Which is a shame, there is value there if you're willing to separate the concept from the aesthetic that surrounds the events themselves.
It can be great fun as an informal community gathering of developers to work on random pet projects, but the reality of the corporate hackathon trends toward being company sponsored and promoted technical debt creation.
It often seems like the corporate hackathon is the clearest distillation of the "business" people not understanding how to create a mature, healthy culture and continuous process around engineering autonomy, ownership, and innovation. Instead relying on campy caricatures of basement dwelling Mtn. Dew & pizza fueled "hackers".
When I was younger I used to work through the night with friends on stuff we wanted to do. It was fun and social. There was no commercial motivation or value. We just had a lot of time on our hands being geeks with no social life and we enjoyed building things. In a corporate setting this is maybe similar to Google's 20% time which as far as I hear isn't really working any more either. The closest thing to this in my day job world is people taking some time to just do whatever they want to do without giving a damn what management thinks. If you can do a little of that in your workplace then you're basically getting the "true" hackathon effect ...
First it has to have an end goal where, even if you don't come in first place, you still leave with something that will be useful. From an incredible learning experience to something that could make you money.
Second it shouldn't cost you anything directly except for time. Working extra hours for your own company? That's costing you more than time because it's becoming a requirement ("Look at Johnny over there, he's gone to all our hackathons and now look at him in a more senior position. Do you not want to move ahead?").
For instance I just finished participating in Launch's Hackathon this weekend. Other than the costs I paid for my travel I paid nothing but time (they provided all food, drinks and not all of it was unhealthy) and in the end, even though I didn't win a prize, I left with the start of a really cool product (if I do say so myself) where, realistically, I think I'm about a week from delivering a decent, beta experience. I even created a Twitter account just so I could post my fun experience as I went: @heysimex
TL;DR: hackathons that cost money, that don't provide you with SOMETHING useful even if you don't win or that are run by your own company are rarely good experiences (unless you win). Hackathons that cost you nothing but time and you have something at the end of the thing? Awesome. I love it!
Otherwise, I agree with his points. Not sleeping for 24+ hours really takes its toll, especially if you need to drive back to where you came from. Food is mostly junk, but this one other hackathon I went to had really well-made food, so there's that. They have been unrewarded since I always lose steam before I get to make an MVP and they do seem to be very commercialized and they're really trying to get groundbreaking stuff from us within a weekend, like the last hackathon had very specific prizes relating to using certain company's API as well has performing "health innovations".
I also like Brian's suggestions to provide more specific hackathons so that I know exactly what to expect from them. These events take up a huge portion of my weekends theI should be using on school, so it seems like a good idea to pick those which seem most useful at the time.
We actually just held our Monday-Friday Healthy Hack Week last week (http://healthyhackathon.khanacademy.org). One of my favorite weeks ever, particularly because we encourage "hacks" of all kinds that don't have to include coding, and the entire company dives in.
They've been internal only so far. Maybe some day we'll open it up.
After a week of hackathon'ing intensely I'll walk in tomorrow more energized than ever (and eager to figure out how to use some of the team's work in upcoming projects). The commitment to staying healthy while hacking feels like it's paying off for us.
Plus it demonstrates our commitment to long-term investment in teammates.
It's ironic that obviously very few Khan Academy employees have children of their own, but that's probably because it's only a few years since they were children themselves...
We do not require that anybody stays at our hackathon until 11:45, but we do require that people get sleep. These decisions -- and the decision to not only host over a weekend -- come from our family-aware mindset.
> We do not require that anybody stays at our hackathon until 11:45, but we do require that people get sleep.
Parents of young children don't need to be told by their employers to get sleep.
So I read that your employees actually do have family (which makes the thing about staying late worse), that you don't "require" that people stay until midnight (as if that was a thing you could require), and patronizingly insist that people get sleep.
Maybe I just didn't read that correctly, language barrier and all that but... I don't know, these extra hours, are those paid overtime or is that a further erosion of the 40 hour work week?
I think you're being overly negative.
Nobody is forcing anyone to go to these things. If you have other stuff to do, fine. If, like some people, you actually enjoy it, you can go. Maybe you even have an understanding husband/wife who understands that once in a while you might be out as late as 11:45pm
Many did not and made amazing things thanks to the commitment by KA to use company time.
That's just having more energy than you know what to do with.
The tragedy is young people have lots of energy and are scatterbrained and old people are sometimes wise but too tired.
The future is having older wise folk write out proposals to ambitious projects, and have young people fill in the necessary parts.
Too bad it's just not 'sexy', ambitious projects are long, boring and require upsetting other people who are old, boring and don't want to listen.
Modernizing the medicine field is one simple example of such an ambitious problem.
That's generally the past and present as well. With some notable exceptions I think that's how the world has pretty much worked through most of human history.
Me, I past 30 a few years ago and I (often) find it useful to give myself deadlines and force myself to get things done.
Things/words that have helped me:
* Pomodoro technique / time boxing
* Get it working, get it right, get it fast
* Cult of Done
That said, yes, I am also skeptic to the kind of "hackaton culture" that is being mentioned here.
Absolutely not. The most brilliant of minds made their marks on history at 30 or younger. Science has shown that, over time, learning becomes more difficult and slower. The whole idea of an older person being more "wise" is more of an old wive's tale more than anything else.
Young people and old people alive can have great ideas. Inversely young and old can be great at simply cranking away at building things. The future mixes the young and old more and more making the young younger for longer and the old far more capable than ever before.
Take politics for instance - can anyone under 30 accomplish anything ambitious in politics? Probably not.
The brilliant minds who've contributed to math/science I imagine oftentimes had great teachers along the way. Not always.
The ambitious projects I have in mind go beyond solving a previously unsolved math problem - I am thinking of something more calculated and long-term like wikipedia.
If we can give a lot of people in the world the best educational materials by the best teachers - that's a massive win.
If we can secure funding for people who show promise and let them affect the real world, not just 'do research' - that's a massive win.
In other words - the problems we're facing are not really science problems, they're philosophical and societal problems that require a willingness of the current ruling class to say ok, this isn't working, let us try something new here.
That'd be nearly unprecedented - another smart kid solving another science riddle, that's cool too :)
> Take politics for instance - can anyone under 30 accomplish anything ambitious in politics? Probably not.
Maybe? It's hard to say considering we have laws preventing the young from getting into politics. President / Vice President have to be 35, 30 to be a senator and 25 for the house of representatives. Most states also have age restrictions for many other political offices that I'm too lazy to look up (some are similar to the federal ones, some are simply 21).
Making it not possible for them to do it doesn't mean they couldn't if given the ability. Many politicians are career politicians. That may not be good but the young folks can't be career politicians, at least not right away.
I would be very curious to see the data if we removed all agent restrictions then let 1-2 generations go by and see if the demographic changes much / some / none.
I'd say moreso than the laws, is just the hierarchies entrenched in every sphere of life are the real mechanisms preventing any real change.
On one hand, stability is good, but at what cost, right?
Something as small as the apple app store - everyone knows it's bad, nothing has been done about it for years.
If Apple tomorrow, came to HackerNews and said hey, we know the app store sucks, we are going to hire a 20 person team of stars to re-build it. We will give you full autonomy, high paycheques, lots of publicity and a full year to do it.
Wouldn't that be an overwhelmingly positive move? Even if a year later, the team failed to deliver, at the very least it'd generate a ton of enthusiasm and new ideas.
This is a small example of big companies not willing to do anything outside the stale old box, even when there is absolutely no downside. It's those old, bored farts earning a paycheque that are the problem every time.
Apple has the money, just not the wisdom to go outside the box every now and again. Politics is 1000x worse I imagine.
This can easily be abused to keep talented young people 'in their place' which elders get all the power.
Have anyone write the proposal, and any critique it - just ensure both roles are considered equally valuable.
"This should easily be abused to keep untalented old people 'in their place' which the youngsters imagine they get all the power. We're still walking away with the profits.
Have everyone ignore the proposal, then force all to critique it - just ensure both roles are considered equally invaluable. That way no one will get a raise during performance review. And we're still walking away with the profits."
But I think companies that try to wedge "hackathons" into their employees' schedules as some cool twist on getting more free (or very cheap) work out of them should burn in hell.
Seriously, I've seen what some people cobbled together in a week long hackfest and been completely unimpressed, having built something similar, and more robust in a day and a half- I did not consider my efforts (or theirs) worthwhile. To be clear, all efforts that you learn from are worthwhile, but I am meaning (and I think what you mean) worthwhile from the standpoint of, getting something production ready or having completed an idea or concept to the point that we can figure out what the heck to do with it.
I think the best things to come out of any hackathon setting are the ideas. The execution that actually occurs should be more of a fleshing out the feasibility and sparking iterative ideas. Anyone walking out of a hackathon thinking they've got actual usable code is just wrong.
That said, the group I am part of, we regularly engage in focused "hackfests", where there is a clear goal, people work together, and a lot gets done in 3-5 days.
Is it always complete, no, but for solving spot problems, short term deep dives can be great.
Generalizing that, as hacktahons try to, there will be some interesting successes, but that won't be the norm.
Grandstanding and networking sound like b-school entrepreneurship competitions. Who can make the best slide deck. Plenty of companies got funding with nothing more than a few presentation slides and some good talk so I agree it's a potentially valuable thing. It just doesn't reward technical prowess so why feature it at all?
Hackathons are a blend of all of this and it doesn't really do any of them well.
My main complaint about hackathons is that they are a poorer recruiting tool than most people want to believe. Corporate sponsors think they'll be able to spot top talent easier. And some participants think hackathons improve their chances of getting hired. In my brief experience, the success rate is very low on both sides. At best, I think a hackathon is a fun, creative, and social collegiate experience for the participants and another avenue for companies to market their brands to impressionable young minds.
When all was said and done, the judges—all non technical—took about 30 minutes to decide that it was a 3-way tie. If you're going to make something a competition, just pick a winner.
the judges—all non technical—took about 30 minutes to decide that it was a 3-way tie.
This is not a bad thing. I've won hackathons partly–but not strictly–because we had a team that could present well in front of judges and an audience.
Public speaking, sales, and presenting to others is an important skill in business and entrepreneurship. The author would be wise to convey this to his students. A hacker who can build a product of value AND sell it well is a force to be reckoned with.
Sorry, but I love hackathons as long as they're run somewhat well. If I don't like the food, I'll go eat somewhere else. I'm too old these days to work through the night so I usually get a hotel room and get a good sleep. But college students are already used to studying through the night, so I don't see what the big deal is.
Beyond the intersection of making <-> presenting (see above), I also love that hackathons force you to work within constraints–most notably time.
Its a great way to network–especially if there are corporate sponsors. I've met some great friends at hackathons and I've received job offers from companies I've met. I've even been offered investment in a project.
Yeah, people cheat. Yeah, some hackathons are poorly run. The pros still outweigh the cons for me though.
Not so much with writing software. Communicating with others yes, but not necessarily in a public setting.
I imagine they'll continue to drag along, like most conferences, like most people, who just don't know what else to do, so they do something that someone else said was a good idea and isn't intimidating.
(I haven't tried Star Fighter yet, but I imagine it's similar too.)
It's like working in a hackspace for a night (which I often do when on the road) combined with a friendly competition.
I'd avoid a for-profit hackathon like the plague.
https://medium.com/@folz/selling-out-and-the-death-of-hacker...
Previous discussion:
It's a great way to have a fun weekend, get job offers from local sponsors and learn new technologies in an environment where you can ask get help from experienced people. It's not about writing realistic, maintainable code - but that isn't really the aim.
On the health side of things: Going to an event and having unhealthy food isn't terrible. It's a short event - having pizza for a day isn't going to kill you, and at these hackathons healthier food (sandwiches, cereal, burritos) was also provided. If you want to sleep, you can, but there's nothing wrong with occasionally working through the night.
[1] - https://hack.strathtech.co.uk/ [2] - http://gutechsoc.com/hackathon
I wonder if this can be approached with rules, such as:
* Provide datasets, interesting hardware, or other things that the participants do not have in advance. Release them only at the start of the hackathon. In the judging criteria, have heavy weighting for how well the teams make use of the provided resources.
* Forbid, on honor code, to bring your startup (or any other project you have started before) to the hackathon.
* Prohibit, on honor code, use of any code written prior to the hackathon unless said code has been under an open source license for 3 months, advertised and available for the public to download, AND your team does not represent the majority of the contributions to said code.
* Define one or more problems to solve, within the scope of the hackathon. Don't release the problems until the start of the hackathon. Require all teams to work on a solution to any of the stated problems.
Many hackathons are judged or attended by major investors, so the incentive to cheat is extraordinary. The last hackathon I went to there were several teams that clearly had everything lined up ahead of time and were there to pitch, and I doubt you can get rid of this without getting rid of the investors.
More than that, many hackathons pride themselves on being interesting for investors, so they're incentivized to look the other way so that the presentations are as impressive/fundable as possible.
I don't think I'll be going to another hackathon for a good long while - the combination of people who seem to have no interest in the spirit of hacking on stuff, and the incredibly toxic culture I saw (getting smashed, pounding back energy drinks, flying drones around annoying people, nerf gun fights through the hallways, it was like a goddamn freshman dorm but sweatier and somehow more childish), makes me completely not interested at all.
The downregulation of dopamine is partly my fault, but damn the food is beyond just visually tempting.
[1] https://wiki.haskell.org/Hac_%CF%86
[2] (http://www.meetup.com/Boston-Haskell/events/223607923/)
In addition to all the provided stimulants, hackers are
encouraged to work through the night. Sure there may be
reminders that you should sleep when tired, but why a 24-
hour event instead of a 12-hour one then?
How does one without free nights and weekends (eg, has kids and a family) get the benefits (network, exposure, etc) without being able to participate?
That sounds entitled. Let's reverse it. How does a company find individuals who have better things to do than code for free all weekend? This is of course based on the assumption that such individuals are valuable.
The only answer I've come up with is coaching little league.
The angle was to bring new ideas to the table and build an MVP/POC to demo the idea. Overall was a positive experience other than the politics involved.
The event was voluntary, took place in the hotel with beer in the hotel bar and nice catered lunches.
>If this sounds like advocacy for mediocrity, instead of merit, it is not. Keep in mind the following facts
And then follows up with a bunch of excuses about how they are not measuring merit by any means because traditional presentations and different project timelines are to difficult to compare for merit.
"it sounds like we don't want reward merit, and that's because we don't have the competence to measure it"
It's like LAN parties, but programming instead of gaming.
I do like the FU Weekend idea, though.
The group that organizes them has recognized a few powerful factors:
- The organizers are diverse and have a wide array of skills
I didnt start the group that I have collaborated with, but I immediately tried to contribute to the best of my ability. We have people who are good at promoting, thinking about diversity, good a logistics, organizing speakers, etc.
- Intentionally diverse groups in gender/skill/background
We spend time to open applications, then manually select the participants who are accepted. From there, we keep a high touch relationship with 100-150 people, so that they are excited even before attending.
- Pre-crafted/well-rounded teams
Based on applications, we make pre-crafted teams that have balanced skill sets. We make these teams based on our understanding of the participants, and their own experiences.
- (most importantly) NO PRIZES
We make our events based on the valued gained from meeting and working with people who are fundamentally interesting. We build the concept of the event based around getting cool people together to work on cool problems.
For the past two years, we have welcomed well over 600+ people in 4 different cities.
- Partner with awesome organizations with strong brand names
We spend significant amount of time to raise sponsorship. We make a point to involve cool organizations in the space, so that the event attracts high quality participants. Even if a great organization that we like cant sponsor money, we try to involve them as a partner in someway, so we can involve their staff. This has always paid off.
---------------------------
Reference:
- http://hackingjournalism.com/mobile/ (Based on mobile journalism at MIT in Cambridge)
- http://hackingjournalism.com/video/ (Based on video production at Conde Nast in New York)
- http://hackingjournalism.com/ (Based on data science at Washington Post in DC)
- http://codexhackathon.com (Based on book publishing at SF and Boston)
- http://audiohackation.com (Based on podcasting with ThisAmericanLife in New York)
It's also possible that my view of hackathons is skewed by the fact that I'm a university student at a school with a very active hackathon culture.
or how to teach early to young new passionate developers that the "norm" is to work long hours for next to no pay, pizza? yeah good enough.
It also vehiculate the idea that code is a just commodity, that anyone can piss some code over few days and build full working apps supposed to impress people.
Well, understand those poor company, they are not sure you can be a good slave, so they need to test you over a week-end: does he/she complain for not being paid? not sleeping? being forced to use this API vs another?
howdy boy, "slap on the back", here your prize, by the way everything you have produced this week-end is owned by whoever organized the hackathon and they can do whatever they want with it while still not paying you, now go away you need to sleep and you stink.
I saw companies organizing hackathon for their own employee, not something open to other developers or involving a public API, no sir, a company week-end event.
They did those hackathon to allow their own employee to work on "what they like" or "what they did not have time to work on during working hours", but still have to be related to the company business, in exchange of beers and pizzas .. oh and if you don't show up we'll know you are a lazy bastard not inline with the company direction.
So yeah maybe not all hackathon are like that, but most of them has this feel of scam where at the end you want to scream "fuck you pay me".