programming and human factors
by Jeff Atwood
In April I donated $5,000 of the ad revenue from this website to an open source .NET project. It was exciting to be able to inject some of the energy from this blog into the often-neglected .NET open source ecosystem.
As I mentioned at the time, I used a very hands off approach. While I did have some up-front criteria for the award (open source license, public source control, accepts outside source contributions) it's basically a no-strings grant.
The real money is being sent via wire transfer to Dario Solera, the ScrewTurn Wiki project coordinator. What's Dario going to do with this money? You'll have to ask him. That's not for me to decide. There are no strings attached to this money of any kind. I trust the judgment of a fellow programmer to run their project as they see fit.
When I said the project could do whatever they saw fit with the money, I meant it. Buy liquor and cigarettes, throw a huge party, play it on the ponies. I'm not kidding. As long as the project team believes it's a valid way to move their project forward, whatever they say goes. It's their project, and their grant.
I hadn't heard anything from Dario, and I was curious, so I followed up with him via email. He sent back this response:
The grant money is still untouched. It's not easy to use it. Website hosting fees are fully covered by ads and donations, and there are no other direct expenses to cover. I thought it would be cool to launch a small contest with prizes for the best plugins and/or themes, but that is not easy because of some laws we have here in Italy that render the handling of a contest quite complex.What would you suggest?
I was crushingly disappointed to find out the $5,000 in grant money has been sitting in the bank for the last four months, totally unused. That's painful to hear, possibly the most painful of all outcomes. Why did we bother doing this if nothing changes?
My friend Jon Galloway warned me this might happen. I didn't believe him. But what other conclusion can I draw at this point? He was right:
Open Source is to Traditional Software as Terror Cells are to Large Standing Armies – if you gave a terrorist group a fighter jet, they wouldn't know what to do with it. Open source teams, and culture, have been developed such that they're almost money-agnostic. Open source projects run on time, not money. So, the way to convert that currency is through bounties and funded internships. Unfortunately, setting those up takes time, and since that's the element that's in short supply, we're back to square one.
I had hoped that $5,000 grant money would be converted into something that furthered an open source project -- perhaps something involving the community and garnering more code contributions. But apparently that's more difficult than anyone realized.
Jon offered these ideas:
I must admit I'm at a bit of a loss here. Do you have any ideas for how the Screwturn Wiki project can use their $5,000 open source grant effectively? If so, please share them in the comments here, or on the ScrewTurn forum -- in the Suggestions and Feature Requests area.
Even I'm not naive enough to suggest that money can solve every open source software problem. But I don't have a lot of time to contribute; I only have advertising revenue. I'm absolutely dumbfounded to learn that contributing money isn't an effective way to advance an open source project. Surely money can't be totally useless to open source projects... can it?
- Specify a well-defined task to accomplish
- Ask for a coder to do the task for money on http://www.rentacoder.com/
Money in open source projects may end up just being:
http://en.wikipedia.org/wiki/Apple_of_Discord
The real question is how do you spend money in an open source project? The last thing you want to do is cause team infighting...
If the project lead took a vacation to devote to his project, how would the other contributors react?
brian on July 29, 2008 2:11 AMI think you drew too much of a conclusion of what Open Source is. And, it appears that many of the people who replied have similar ideas.
There is an idea that Open Source software is written by a bunch of software hippies who are anti-capitalistic and would believe in free love if they could ever figure out how to get close enough to a girl. Open Source is merely a philosophy that you can benefit by opening up your software to all coders.
Some Open Source projects are very large, need lots of hardware and actually pay people to do such things as documentation, testing, and even (gasp!) development.
Most Open Source developers are actually paid for their work. IBM, Apple, Sun, Google, and maybe even Microsoft pay developers to work on various Open Source projects.
Some Open Source projects (like Joomla) are run by the very people who use that software in their own jobs. Some like MySQL are created by companies that decided the best way to make money was to get their product used by millions of people and then make money through training and support.
All Open Source means is that the development team decided to make the code freely available to the world, and there is a wide variety of reasons for doing that. Sometimes it is purely done for philosophical reasons, but most of the time, it is for strong business reasons.
I don't know much about the Windows Open Source community, but in the Unix world, you couldn't even start coding without Open Source projects. At my work, our development team depends upon Linux, Spring Framework, Subversion, Eclipse, Apache, Tomcat, Hudson, and probably another dozen or more Open Source packages.
David W. on July 29, 2008 2:23 AMIsn't this what Bill G found with his foundation - money without strings attached is wasted?
Syd on July 29, 2008 2:25 AMForget about the money. You said no strings. That means no strings. No following-up either. Let it go.
Another +1 to Dave.
Not that it wasn't expected though. I mean, this is Jeff.
Christopher Galpin on July 29, 2008 2:32 AMMy fingers are so used to typing 'codinghorror.com' I just put it as my website. Spooky.
Christopher Galpin on July 29, 2008 2:37 AMWell now, how to spend money. hmm some ideas:
1) Google summer of code now fall and winter of code too
2) PAID student coding projects (learning based) We can only cover so much in a classroom
3) Scholarships in programming (not the CS light programming , but the real calculus, physics based stuff) at your local College.
4) Create a game contest(s) - by the new old 8(16?)bit boards from think geeks(?) and give out 30 of them to schools for a battle of the side scrollers contest with a $ or thing award.
5) open source library of books (say the O' books) for the first 100 subscribers? Books are expensive!
How's that for getting code writing out to where it needs to be?
So many of my really bright students fight to find the small funding they need to go to school, and anything that helps that is good. Additionally anything that gets the students excited enough to crack open a compiler and code is a win - especially if I can get real programmers to give feedback (with the understanding that they are raw programmers, and feedback good critiques!)
Just some rambling from a college professor.
I'd think a Safari subscription or conference trip for a key developer(s) would pay off nicely.
Other options would be to pay for somethings that they might normally do without like higher quality hosting or service contracts. Would a TechNet Subscription help? What about a dedicated server? A MYSQL or SQL server service contract or VMWare Workstation licenses aren't free but can be big help.
Andy Fundinger on July 29, 2008 3:13 AMMost volunteer efforts I've been involved in go to crap once money is introduced. I'm not talking Open Source i mean ANY organization of volunteers doing anything.
There is a reason many charities do canned food drives and not money donations. Who decides What and How the money is spent? Are we now an oligarchy, representative democracy or dictatorship, who's treasurer, did they get elected, can they be trusted? If ANY of the other volunteers disapprove, now you have drama and major problems. It can turn a bunch of volunteers having fun into a squabbling bunch of children.
I've seen it in charity groups, live action role playing troupes, movie clubs .. you name it. In a lot of organizations i've taken part in, we go out of our way to avoid money. NO cash, donate a book, some supplies, your TIME that is fine but NO cash.
brian on July 29, 2008 3:21 AM3) Scholarships in programming (not the CS light programming , but the real calculus, physics based stuff) at your local College.
Just some rambling from a college professor.
I think I smell some bias!
Adam on July 29, 2008 3:23 AMJeff, I share your astonisment at the lack of imagination on the part of the developers. As a user of Screwturn Wiki I have 2 ideas on what they could spend the money on:
1. Hire a web designer to create some themes, the themes Screwturn comes with are appalling.
2. Hire a technical writer to create some useful documentation. The current docs are a shambles and make installing and working with screwturn anything but easy.
Aha! So open source is the way to a Star Trek (-like?) world where money no longer has value!
Seriously though, just about any way I can think of to make money useful for a project has the risk of making that project *depend* on that money, which might effectively kill the project. I.e. money might not just be useless to an open source project, it might be dangerous as well. Basically the thing you might be causing, and need to look out for is money addiction (project wide).
Giel on July 29, 2008 3:58 AMGive it to the Mono people so .NET won't be stuck in the Windows ghetto forever.
Rich on July 29, 2008 4:02 AM 1. Hire a web designer to create some themes, the themes Screwturn comes with are appalling.
2. Hire a technical writer to create some useful documentation. The current docs are a shambles and make installing and working with screwturn anything but easy.
Actually, the problem with hiring people is that it requires time (legal stuff among others). There's also the add more people to a late software project and it will be later principle, i.e. it will initially cost more time than it will yield. Thus the question is, will the money have run out before any additional spent time is gained?
As that's the real problem here: finding ways to convert money into time.
Giel on July 29, 2008 4:07 AMNext time around, maybe you should consider giving money to projects that have a clear idea of what to do with money, and how much is needed.
Paulus on July 29, 2008 4:10 AMThere are some great ideas in the comments here. I favor these, personally:
That's exactly why paying for transport (flights and hotels) is a better fit for an open source project - people are already donating their time, but the travel costs often make the difference between going to an event and staying at home.
Dario, spend it to attend to PDC in Los Angeles!!! Attending is half of what you have, the other half can cover the plane ticket and some lodging. This way it's something in the middle between a vacation and something useful!
Hackathon. Pay for tickets and a place to sit for the core team if they are geographically dispersed.
Pay for travel, fees and other expenses to go to a .NET conference or two. He could improve his skills and give a talk to drum up interest in the project, possibly leading to new contributors to the project.
It's up to Dario, of course, but I'm particularly fond of the offsite hackathon with key contributors. Putting a bunch of geeks in a room for a few days and letting them hack together on the codebase would be fun, wouldn't it?
Jeff Atwood on July 29, 2008 4:21 AMJeff, I really think it depends on the maturity of the projects. Some projects are just not... capable of handling money. What to do with the money? is their question. Or they are just happy with the few donations that they get to survive (which is not bad, at all).
I'm backup up a previous comment on Blender (www.blender.org). It has been an open source app for several years now. And it is constantly fed by money! They even have set up a Blender Institute which - in a few words - serves as a office/studio that aims supporting projects made on Blender (for instance the newly release Big Buck Bunny) and all expenses are covered by contributions or money made by the Institute. The purpose of it is that by making things easier for artists/developers, the whole Blender community gets benefited by awesome new features... for free!
That's putting money into good use.
I'm by no means advertising Blender (or am I?). I'm just using it as an example that some Open Source projects do know what to do with money they receive.
Federico Cceres on July 29, 2008 4:23 AMI would break it up at least part of the money and send it to the projects whose software I've used. I could also use another server as an offsite mirror (gotta love living in New Orleans).
Niels Olson on July 29, 2008 4:28 AMI might be a bit far down the page here, but $5000 will get you a student developer for an entire summer here in New Zealand.
Jamie Penney on July 29, 2008 4:41 AMI'm one of the primary developers on the Bugzilla Project, and we have a donations system. Here's some things that we have done with the donations:
1) Fund a contractor to do a survey of the major Bugzilla-using organizations in the world, to find out what we should be doing in the future.
2) Pay for registration at a convention for one of the developers. (OSCON, which has a particularly pricey registration if you want to attend the sessions.) The Mozilla Foundation covered the travel expenses, our donations covered the registration fee.
I think the problem isn't what to do with money, the question is what to do with *small amounts* of money. If I had $100,000, clearly one or two developers could take a year off work and just work on the project. If we made enough in donations to cover even a single person's work year, that would immensely enhance the project, where currently *all* developers are volunteers.
Also, I personally have a contracting company, so sometimes the donations can fund my business for developing features that the Project wants. This is handy because my company has more resources available than just me, and it also allows me to keep eating while I'm contributing to the Project.
You could also use it to buy some t-shirts and give them out to the best community members each month, or just give away *something* to the most active community members (perhaps computer hardware?).
Also, sometimes it's handy to use the money to buy a new computer for one of the contributors. The Mozilla Corporation helped me buy my machine (even though I don't work for them), and that really improved my productivity.
You have to get a bit clever with money in an open-source project, because it's not a traditional type of organization. However, if you have a decently-sized project and also some business experience, it's possible to think up various items like the ones above that can really help.
-Max
Max Kanat-Alexander on July 29, 2008 5:04 AMForget about the money. You said no strings. That means no strings. No following-up either. Let it go.
I think that makes +3 to Dave now? It sounds like Dario's got the situation well in hand: He made something, you liked it, you gave him some money, he put it in the bank, end of story.
I don't see why you are so eager for him to waste it, as if it was a check from Grandma on his birthday. (Come on, spoil yourself! Buy a third monitor! Have all your friends over for a party! Go on a weeklong vacation to, um, Italy!) He's running an open-source project, not a shopping spree. Leave him alone.
Anonymous Cowherd on July 29, 2008 5:30 AMAt the Platypus project (http://platypus.pz.org), we are about to hire someone to review our design and do a full code review. I think this is an excellent way of spending a few shekels, such as part of the sum you provided generously to the wiki project.
And one thing that I didn't emphasize in the above comment is that you can use money to *get* time. That is, some developers on the project can't spend much time on the project because they have a day job or contracting duties. But if they're being paid, they can spend time on it. I've funded some people who normally don't do much work on Bugzilla to help out with development, and that does make a difference.
-Max
Max Kanat-Alexander on July 29, 2008 5:39 AMNot all open-source projects have so little expending need.
I monitor FirebirdSQL project and, if I have US$5000 to help it,
I would donor it. (Actually, it's 3 months of salary of mine)
Database systems are a example of very complex software which need
a very rich collection of skills to get go. Let alone document it.
This type of development, where many developers deposit their professional reputation on the result, is always needing of real money.
I believe that exist other software types of open-source software that have enough complexity that the people behind of it could really benefit of some money.
Fascinating question Jeff. I think one of the areas open source needs to develop in is the financial aspect. Specifically, we need to see more open source projects utilize full-time developers. This is really where your money could come handy. Have a fundraising drive to raise x to take x developer on the project full-time. In so doing one escalates the progress of the project dramatically. This isn't the sort of thing that can simply be done with your donation, but a significant donation such as this could be used to start rolling a fundraising drive for a full-time developer.
David Mackey on July 29, 2008 6:19 AMI really like the idea of paying a 3-7 days offline hackathon.
Hiring a contractor for an OSS project is uneffective, as it will usually be someone without the flair for the projects, and without knowledge of codebase, and thus the money would not be effectively invested.
I recommend hackathon, because I know how email/irc communication slows down things - you have that awesome idea, but it takes an A4 with an image to explain over mail, and before you write the text, motivation is over. In person, you tell it in 30 seconds with pencil and paper, and immediately see reactions - and youu have just saved half an hour in a minute (I loved doing school projects with my roommate, i felt soo effective).
Making a hackathon is saving time on communication misunderstandings, and I believe that few guys in room could do 1-2 months of after-hours voluntary work in 4 days (and without time/attention tax of multitask-switching).
Keff on July 29, 2008 6:33 AMAnd simple version of argument is: you give time to people who are most interested in project, and let them have fun while doing it(coding something really fast *is* fun :)).
Keff on July 29, 2008 6:35 AMI am not sure about the laws here in the US. But what about having someone here (in the US) run the contest for best Screw Turn plugins?
Maybe you can get around the laws in Italy by using the laws in the US and running it from here.
Adam Lerman on July 29, 2008 7:06 AMThat money IS technically collecting interest, if it's in a bank account, so it's generating even MORE money they're not using!
I think they should just hire a freelance dev to fix the most grueling bugs nobody else wants to do for free.
Matt S on July 29, 2008 7:11 AMYou could hire a contractor or two, for the short term at least, to help with coding/design/etc or spend it on advertising to spread the word about the app.
Training, new software, new hardware for developers, a wild Vegas weekend... I can think of lots of stuff, certainly if I had $5k to use, it wouldn't be sitting untouched.
Kris on July 29, 2008 7:12 AMI think that is just awful. I'm sure there are tons of open source projects that would love to hire a graphic artist to give their website or app a better look. Or to buy a new server to run their site or host their app. I would just make sure that next time you give away money to an open source project, you should ask them if they want it first.
Justin Etheredge on July 29, 2008 7:13 AMWow, $5000 - pretty sweet. I think the problem is simply that these folks don't want to spend it frivolously. Sure they could have a party, but they'd feel bad about it. After all, someone sent them money to work on the project, not have fun with it (did the message from you to do whatever with it actually reach them?).
Maybe a hardware upgrade on their computer to make their compiles faster? How about a computer that they could have a local student(s) use to work on the project? You've really touched on an interesting issue here.
Bill on July 29, 2008 7:14 AMI say hire a designer to revamp their homepage and add more themes to their product.
David on July 29, 2008 7:16 AMThe spirit of Open Source Software dictates that money isn't the most important factor. Money doesn't bring people to care about a project more, therefore contributing better code. Money doesn't find talented programmers, willing to work for free because they like it. The whole point of open source software, is that the people writing it *like* doing it enough to do it for free.
The only thing I can see money doing for an OSS project, is paying for hosting space and paying for advertising to gain interest. Since the hosting is already taken care of, and $5000 isn't really enough to pay for the advertising, I suggest that they place it into a private account to save for when something DOES need paying for. Maybe they want to buy a server to host in-house? Maybe the lead developer will need a new development machine soon? I suggest that they keep it available until it's needed, and use it then.
My $0.02
Alex
Surely they could purchase some hardware, no? Otherwise, I'd go with lots of peanut butter cups, single-malt scotch, and video games. Just kidding! Well, not really.
Kenneth on July 29, 2008 7:19 AMCould Dario use the money to further his knowledge which could then be fed back into the project? A coding bootcamp, [insert subject here] training course, etc.?
Maybe a holiday to recharge his batteries!? :-)
Alastair Smith on July 29, 2008 7:20 AMI agree with the above commenter: if they want to run a contest, run it from the US - I'm sure this blog could play a part in that. Also, if the money is in an Italian bank account, it's most likely losing money because of Italy's shitty banks.
David N. Welton on July 29, 2008 7:20 AMI think some projects are more able to use money than others. I also think that amounts of money make a difference. 5k is great, but 60k could pay for a full time developer for a while. In this case, I would say what are the most critical goals? If they want more users, ad money and paying for documentation would be a great use of 5k. If they want plugins, don't do a contest, just pay someone to make some plugins. If they don't have enough users to know what a good plugin would be, see my first suggestion.
Russ on July 29, 2008 7:21 AMJust a few things off the top of my head, I really like Jon's technical writer idea by the way.
1. Outside UI help, perhaps with focus on usability.
2. Hackathon. Pay for tickets and a place to sit for the core team if they are geographically dispersed.
3. Paid time off work for one or more of the lead developers. A paid sprint if you will. Buy time.
Sometimes it amuses me how badly people misunderstand Free Software development.
What got them there was not money. The incentives for Free Software tend to be either fun, reputation, or (in corporate cases) getting some needed work done. If you want to reward Free Software developers, you give them more of those things.
I bet if you followed up, you'd find that the positive mention on your blog was worth *way* more than the money to that project.
Exactly. _Every_ opensource project could use:
1. A good artist, to make the product more appealing
2. A good web designer, to make web-site of the product look good (hint: check my page, I love its design)
3. A good SEO to make sure the word about the project is spread well - thus getting even more people to work on it.
As well, they could announce a hacking contest - the one who manages to hack the wiki server gets a hefty sum of money - thus heavily boosting security of their project - which for projects like Wiki are quite important - imagine PR9 Wikipedia linking from each page to enlargement-selling sites.
And they could finally remove that ugly blinking 'ScrewTurn wiki compatible ASP.NET web hosting' advertising from their main page at a href=http://www.screwturn.eu/http://www.screwturn.eu//a">http://www.screwturn.eu//a">http://www.screwturn.eu/http://www.screwturn.eu//a
And of course - throw a party! Things like this _really_ boost the enthusiasm in open source communities.
Gary Schubert on July 29, 2008 7:24 AMPerhaps some kind of bounty system where rewards are set for certain features to encourage more contributions. I've not looked at the project itself so I've no idea whether or not they have enough contributors involved already or whether the project itself isn't well suited to this.
Sean on July 29, 2008 7:26 AMThey should at least, buy new hardware... It's wise to take some time to think but, 4 months? Come on
Ignacio on July 29, 2008 7:26 AMI don't think that's the worst possible result: you've given the money to someone who's apparently pretty conscientious. If it sits, it sits. That falls under whatever they saw fit. My suggestion would be to spend the money on a vacation to re-charge. That or use it to get all the people on the project together in one place for beers, assuming there's more than one person and they're geographically distant.
Tom Clancy on July 29, 2008 7:26 AMI understand Italy's laws are too oppressive to hold a contest, but bounty-hunting shouldn't be such a problem. or how about improving that ugly site they have? perhaps turning into something that showcases many examples of beautiful themes used by different websites that use ScrewTurn for their wikis...they could hire a graphic artist for 5k that'll make for them the site of their dreams.
Brian on July 29, 2008 7:27 AMOne more thing Jeff. Your 'no HTML' feature is bad, face it.
It wasn't me who screwed the link to screwturn website - it was screwed by your blogging software when it replaced the url with html and then refused to post it because I have mentioned v-i-a-g-r-a on the page.
Just a friendly bug report. Don't be offended - I still think you are a good, um, web developer.
Gary Schubert on July 29, 2008 7:29 AMZombie defense.
With all the zombie movies that came out of Italy it must be a real threat.
N on July 29, 2008 7:31 AMWhen confronted by a gift of money intended for an open source projects benefit, as the maintainer, there can be a huge element of guilt involved when dipping into the money, a mental tax of sorts.
To some projects, a gift of $5000 would be more of a burden than a help.
Tom J Nowell on July 29, 2008 7:32 AMDario, spend it to attend to PDC in Los Angeles!!!
Attending is half of what you have, the other half can cover the plane ticket and some lodging.
This way it's something in the middle between a vacation and something useful!
Ciao
Massimo
Take part of it, design some cool polo shirts/t-shirts. Then offer them up freely to code contributors... as a marketing technique to recruit code contributors, not necessarily users.
Unfortunately, the shipping costs for something like that outweigh the costs of the shirts themselves... so you end up giving your money out to Fedex.
philibert on July 29, 2008 7:41 AMThe donation should be split between all the developers, to be treated as plain income. It's like a bonus for a job well done. They can use it to buy new toys, or pay rent.
Since many of these open source projects depend on hosting companies and other supporting infrastructure, why not donate the money to an organization that specifically supports open source projects. In that way, you are suppporting numerous OS projects by buying a server that hosts numerous groups.
Edgar on July 29, 2008 7:42 AMMaybe it would be worth buying something practical - for instance, the Top 100 Best Software Engineering Books, Ever (http://www.noop.nl/2008/06/top-100-best-software-engineering-books-ever.html). The coders may already have some/many of them, but there's bound to be something new to pick up...!
Alistair Collins on July 29, 2008 7:43 AMAn Open Source project can progress in two ways: Either there are many many users, a small % of them apply modifications, and a % of that % actually submit their changes, OR developers are actually PAID by a higher corporation to work on that open project (e.g. Novell).
That 5000$ should have been spent on developers, the rest is always cheap enough to survive. +1 for a PDC trip for Dario!
Martin Plante on July 29, 2008 7:47 AMHiring a Graphic Design would be my way to go.
Myles Braithwaite on July 29, 2008 7:47 AMWasted your money on YET ANOTHER FUCKING WIKI. As if there wasn't enough of those fucking things floating around.
Even if they had used it would it have made a difference? Nah we'd just have a slightly better YET ANOTHER WIKI
Trev on July 29, 2008 7:48 AMGiving money to a Microsoft Technology Open Source Project is like giving a barbecue and veal steaks to a religious Indian family so they eat it in their Sunday gettogether where they pray to Allah and play Nintendo Wii: just a plain ridiculous mixup of shit altogether.
Plain and simple: if you love open source, you use open source tools, not .NET, because using .NET is widening M$'s niche and giving them more developers. The more developers you have who know how to work M$ technology, the more M$ technology projects you have, and the less you support real Open Source industry.
So, basically, what I'm saying is: serves you right
Lacrymology on July 29, 2008 7:49 AMI second the notion that they should use it for graphic design. Software developers can write fantastic software all they want, but there is no substitute to a very good graphic design and branding.
Perhaps they could hire the guy that did the branding for Firefox, John Hicks: http://www.hicksdesign.co.uk/
James on July 29, 2008 7:50 AMWhat about buying hardware? I know hardware is much cheaper than it used to be and devs probably have enough hardware already, but it could be the best investment (and the easiest to make).
Basu on July 29, 2008 7:55 AMGary Schubert
There's nothing on the form that says it accepts HTML, why would you think it does? If you had just posted the URL you'd have been fine.
a on July 29, 2008 7:56 AMPay yourself to submit a patch.
Matt on July 29, 2008 7:57 AMI've been working on a small project for TWO YEARS now. I COULD finish it in less than a month if I had a month to dedicate myself to it EXCLUSIVELY (meaning: taking time off from my current job).
Of course, my financial needs (being married with two kids) are such that I can not take an unpaid leave of one month, and vacation time isn't feasible either. So, if someone donated me the amount of money corresponding to a month's wage, I could go to my boss and tell him I want a month off., and I'd probably get it.
I bet that there's a lot of other programmers in the same situation
right now... you're right in saying that open source projects run on time, not so much on money... but money DOES buy time.
I also agree with Bill: part of the problem is that the donated money HAS to be spent in a responsible manner, even if Jeff has stated explicitly he doesn't care how it's used up.
Hire somebody to do Quality Assurance. This is the area most often sorely lacking in open source.
It's also the area that few people what to do in their free time to contribute to projects.
Practicality on July 29, 2008 7:57 AMHire somebody to do Quality Assurance. This is the area most often sorely lacking in open source.
It's also the area that few people what to do in their free time to contribute to projects.
Practicality on July 29, 2008 7:58 AMAnother thing they could do, if the main developer is, as most Italian devs are, a contractor, stop working for a month on his daytime job, and work for a month only on his project.
Or, at least, that's what I'd do if I was a contractor.
@Trev: this is an ASP.Net wiki, and the .NET ecosystem isn't that rich compared to the dozens of wikis, forums, portals and CMSes in PHP. For an intranet, it would be pretty neat.
@Lacrymology : Spelling it M$ doesn't really add to your credibility.
Rob Janssen on July 29, 2008 8:00 AMGive it to contributing developer who wants to spend a month away from their day job.
Or hire a contractor for a little bit, but not for coding--perhaps art, documentation, or web design.
Casey Rodarmor on July 29, 2008 8:03 AMRegarding comments about paying developers:
1. Existing contributors don't need to get paid - they're already doing it for other reasons
2. $5000, whilst a generous donation, wouldn't pay a contractor long enough to make a difference to the project
3. I personally wouldn't let a CS student loose on my code :-), but even if Dario wanted to, an internship is (a) mostly a US thing, except for medical doctors, AFAICT, (b) kind of difficult to provide when you are not a company
Spending it on equipment or training or professionals that don't require in depth project knowledge (which leaves out tech writers) seems more sensible. Or just lots and lots of beer.
Jim Cooper on July 29, 2008 8:03 AMA statistical sample of size one is impossible to derive meaningful data from. Also, if the individual is employed full time then there is perhaps no direct mechanism to translate money into time -- a contractor would probably be able to use it between contracts for a couple of weeks of full time development.
Richard on July 29, 2008 8:03 AMI don't think it's the case that money is useless to open source contributors, it's that such a big amount it useless. A lump-sum of $5,000 isn't enough for a developer to live off, so it's not enough to justify the developer leaving their job to devote more time. Lots of frequent donations which amount to a steady amount would be more useful, as it would allow the developer to perhaps cut down on the hours they work and be confident in being able to still live of that amount of income.
It's more of a teach a man to fish situation. The developer can't do much with the $5,000 except indulge themselves, which isn't in the open-source developers' nature. But if you encourage more people to donate and slowly-but-surely bring the amount of donations to that person up, then you can have much more of an effect.
Or you could buy him another monitor, those always save time!
Maybe they can donate it to another open source project that might be able to make use of it. ;-)
Scott Marlowe on July 29, 2008 8:07 AMItaly? Best way to use the money would be - 4000$ to bribe a local politician to make the software mandatory for a certain council office. 1000$ to bribe a local journalist to write about the software. Then get government money to open a local software house, grab the fundings, move to the US, give back 6000$ and open a serious business there without all the hassles (taxes, laws, corruption, friendships, etc) involved in anything resembling decent work we have in Italy.
whops, did I say it out loud? Of course it was a joke.
The better way to probably spend that money like Jon suggested is probably a bounty for features. These software projects always have a long list of things that need done, and a little money is enough to spur a college student to work on the project. Probably as little as $50.00 a week would be enough to get the ball rolling and get some real features out the door.
Mono has had some really good success with the Google Summer of Code internships. Maybe that is the way get things done.
By the way on a different note, if anybody has a good idea what I can do with the domain name buggd.com I am all ears. Seems ripe for software developer website or messaging website.
Such as your code is full of _bugs_. Or stop _bugging_ me with your messages. Might even consider doing something like I mentioned above with it, if I had time to plan out what is needed.
If anybody has any good ideas, please contact me at http://www.coderjournal.com.
Nick Berardi on July 29, 2008 8:11 AMHaving looked at his website, Dario is a contractor, 5000$ is 3300€ more or less, which is a bit less than what a contractor gets in Italy for 15 days of fulltime job.
He could, between contracts, spend 15 days on his project.
Pay for travel, fees and other expenses to go to a .NET conference or two. He could improve his skills and give a talk to drum up interest in the project, possibly leading to new contributors to the project.
Or new hardware:
* an Aeron chair :)
* extra/new larger monitors
* his dream keyboard
...
I'm surprised and disappointed by this as well. For me, my development time has a potential monetary value, so unless that time is already being compensated for via other avenues, it seems like that's kind of an obvious compensation.
Other possibilities like design, marketing, docs work seem like good ones too. I'm surprised this hadn't occurred to Dario.
I have put hundreds of hours into my own open-source projects, and donations are minimal. I did receive a grant from OWASP to work on a particular OSS project I proposed, and the money meant I could dedicate my free time to working on that project rather than other pay-for work.
Ed Finkler on July 29, 2008 8:14 AMWe had a long discussion about this in the Tikiwiki project. Currently, we have no ads on our websites. No donations or whatsoever. Hosting is covered by some contributors, quite a few of which (including myself) make a living out of consulting around the project. When someone hires a developer to add a feature, fix a problem or anything, it's a targeted decision. They choose what they want, they choose who will do it. If you give money to the project, the community has to decide what to do with it. It's not that the projects don't need the money, it's just that they often don't have the decision making structures to decide what to do with it.
It's just a balance problem between major and minor contributors, and the conflicts that may occur simply because major contributors are pretty much the only ones who can take these decisions.
We found a few things we could do if we ended up having money:
* Pay for bounties on open bugs voted by the wide community that no one volunteered to fix.
* Cover travel expenses for developers to meet.
The last one does feel like the best way, but then we need to decide who to invite. Gatherings in the project are fairly common and they have a huge impact on community building. However, not all contributors have the same resources and currency exchange just makes matters worst.
Seriously, you should only pay for something specific. This decision making process you skip by just telling the community to do whatever they want just pushes the problem away. Instead of being alone in the process, you ask a group of people to do it. O(1) vs O(n^2) in communication paths.
Next time:
* Ask for communities what they would do with the money when they apply so the money does not sit for 4 months, or
* Invest in something that matters to you and handle the decision making yourself.
Save your charity for curing diseases or feeding the hungry.
Needy programmers should get a fucking job.
Angus Glashier on July 29, 2008 8:29 AMHire a developer for them. Do the work to track down someone with the right skillset, then pay them. It's the gift of time. Let them decide what to do with the developer's time, but a decent contractor should be able to give you 50-100 hours? And maybe, you end up creating a contributor after his paid time is up.
Marc on July 29, 2008 8:39 AMI've got a little Firefox extension I develop for free under the MIT OS License (Status-bar Calculator) and as much as I'd love to say that $5k would spur further development on it, I'm not sure it actually would.
Probably the only thing it would do would be to convince my wife that the hours and hours I've put into it so far were actually worth it. Which I suppose would go towards enabling me to convince her I should spend even more time improving it...
The problem is (I think) that even if a one time donation of $5k drops in your lap, that's not going to enable you to quit your day job and work on the project. Now, if $40k dropped in your lap, you could probably at that point consider finding a way to quit your job and spend several months focusing specifically on the project and getting it to a place where you could make money on it...
Chris on July 29, 2008 8:39 AMIf time is what you need, buy a flux capacitor.
Charles on July 29, 2008 8:39 AMSimple. Hire someone to work on it full time.
Bill K on July 29, 2008 8:39 AMYou said it yourself: you don't care what they do with the money. So why do you care?
Dario: but yourself a new Aeron chair, get 3 monitors, and hire me.
Josh Stodola on July 29, 2008 8:41 AMHere in Germany (and I guess everywhere else) students are always short on money and take a job besides their studies. Go to the local university search some promising higher semester cs student and hire him. 300 Euro/month should give you student with 30 hours per month for one year. Even better - instead of one for a year take 4 for 3 month. Give them for the start no critical task, for the introduction let them write some docs/howtos/faws/tutorials or fix some really easy tasks. Later they can do more critical stuff. An if you are lucky afterwards one or two of them are so interrested in the project that they will continue to work (for free) on the project.
Flolo on July 29, 2008 8:41 AM@a:
There's nothing on the form that says it accepts HTML, why would you think it does? If you had just posted the URL you'd have been fine.
Write a simple url. Without wrapping it into html. Like, a href=http://codinghorror.comhttp://codinghorror.com/a">http://codinghorror.com/a">http://codinghorror.comhttp://codinghorror.com/a
Somewhere in a text of your message write a 'forbidden' word. I will settle on 'v-iagra' since I know it is forbidden for sure.
Type the verification word 'orange'.
Post the message. You will receive a message in red letters: Your comment could not be submitted due to questionable content: [here's the forbidden word]
Type the verification word 'orange' again.
Post the message. Enjoy the mess instead of the link.
They should spend the cash outside of their core competencies, which in this case is probably design/marketing.
Peter on July 29, 2008 8:46 AMWHILE perhaps disappointing it was not spent, GIVE the guy credit for being so honest and forthright!! He does not want to abuse the gift, or use it in some way that might be criticized.
Giving many people in the world money is not the total answer -- many need help in knowing how to spend it wisely.
I think you picked a good guy to give it to.
Steve on July 29, 2008 8:46 AMMotivation money could be used for...
* Meet and greets - 5K is a lot of Pizza
* SWAG - I crack code for Project X T-Shirt?
- I also liked the bug reward proposal
Lars Fosdal on July 29, 2008 8:47 AMDario's a good guy, and i'm sure he'll do something cool with it once he's done punching out the next version of ScrewTurn (which WILL ROCK).
I'd totally have blown it on beer though...
Shog9 on July 29, 2008 8:49 AMOpen source projects run on time, not money.
Time *is* money! If he is giving his time for free, no doubt he is spending time doing other gainful employment. The money might enable him to afford some unpaid leave to develop or promote the product. Also many more people might use his product if they only knew how or could get some help. A thriving business in consultancy perhaps, the money could help start that up.
Clifford on July 29, 2008 8:50 AMAs you quoted - open source projects run on time, not on money. And it actually TAKES TIME to spend that money.
Jakub Anderwald on July 29, 2008 8:51 AMWrite me a cheque?
dnm on July 29, 2008 8:52 AMhookers and blow, duh!
randy on July 29, 2008 8:52 AMYou've touched on a very interesting point. The few open source projects I've worked on were driven by passion more then anything else. Some of them happened to be useful outside my context and I applaud that, but for me, I'd be just as likely to implement a new feature if someone gave me a question which started with, Would it be possible to... as giving me money for a feature. A completely open slate somehow seems daunting - most open source projects are driven from a specific need the original creator has. Anything which can't be tied back to a need is usually fluff to be ignored on an open source project... and maybe an open pool of money falls into that catagory.
I think an alternate way to give donations would be to organise a phone conversation with the important team members of the project. Make it outlined early that the conversation will likely be about directions for the project, nice to haves and questions about whether anything could be done better (tie everything back to a needs based approach). After that, if it sounds like the team have their interest piqued offer the money in a very non-commital way to see some of those features implemented. If it gets done, great, if not... well I'm sure they'll enjoy the beer and cigarettes.
I guess it all gets back to chunking... break it down into a few achievable things which would be nice to do and you'd see something done. Giving it and saying do anything with the money somehow seems daunting to someone who's not used to factoring in money into the equation.
David from Oz on July 29, 2008 8:52 AMThis one's actually pretty easy. The key weakness of nearly every open source application is its interface and branding, and ScrewTurn appears to be no different. Hire a graphic designer with some experience in usability to make a better icon and streamline the interface. (And don't, for god's sake, apply open source methodology to the process.)
Eric on July 29, 2008 8:55 AMRunning a bounty requires a good bit of time overhead, not to mention legal involvment. The same goes for hiring outside contractors/artists/etc. I'd bet money that most open source projects don't have the time or the expertise to do those sorts of things in any reasonable way, nor would they want to spend the time on it, it is rather likely a larger time-sink then the time/benefit to be gained.
Furthermore, running any sort of public bounty opens up the floodgates for people to submit poor-quality code. Normally poor-quality code would be indicated as such and ignored, but with a bounty you now have someone who was expecting to get paid for their time and effort and all-of-a-sudden are not. That's a recipe for ill-will, on all sides.
Etan Reisner on July 29, 2008 8:56 AMThey should just hang on to the money until they need new hardware or something.
Oorang on July 29, 2008 8:58 AMAlso, as to the repeated comments about hiring people, as was commented before I don't think $5000 (while a rather large sum of money to be donated in one piece) is likely to be able to buy anything remotely resembling a useful overhaul of the design of an application. That is, of course, unless graphic designers are wildly less well paid than I imagine they are. Or are willing to work at a smaller cost simply because of the project involved (but in that case I'd have hoped they would have offered their services to the project already).
Etan Reisner on July 29, 2008 8:58 AMWhile you may be right, I think it's a bit early to try to draw any conclusions about money and open source software in general. You've donated money one time to a single project. Maybe if you had this experience at least even 4 out of 5 times you could begin to see a pattern.
But otherwise, the only thing you actually seem to have discovered is that Dario Solera has not been able to determine how to use $5,000 to benefit the Screwturn wiki project.
Neil (SM) on July 29, 2008 8:58 AMI think a few people have hit on the right idea, but overall isn't the mistake yours?
Free money almost never helps. In fact, although I might be wrong, I can't think of a single situation ever where free money has helped.
Money helps when it's used as a tool and in this case, although I'm assuming it was an experiment of sorts, I fault the no strings attached aspect of the money. Throwing money at something never makes it better.
Instead an analysis of where the project is at, what it lacks, and most importantly, what it would need to better serve the benefactor should of been the first step, and then a conversation with the developers to make it clear what you were intending.
Money with strings would of been the way to go, perhaps some up front, and more upon completion, there's a few ways it could of been done, but without a goal for the money it doesn't surprise me that the money accomplished nothing.
Most people aren't self-organizing. We're just a pile of self-organized cells. :)
Micah on July 29, 2008 9:02 AMThe comments to this entry are closed.
Subscribe in a reader
Subscribe via email
|
|
Traffic Stats |