Thursday, October 17, 2013

On the subject of my recent blog about funding Eclipse committers with taxes...

On the subject of my recent blog about funding Eclipse committers with taxes...

Some people have commented saying it's the worst idea they've ever heard.


That was kind of the point :-)

I was trying to be subtle (maybe too subtle), but the whole thing was really intended as an exercise in why I think the notion of taxation isn't going to work.

The best hope you have of getting taxation to work is to do exactly the opposite of what IMHO you should be trying to do.  I.e. in order to try to get the tax to succeed, you end up shooting yourself in the foot by making it harder to contribute, and hence you probably lose more help than you would ever gain.

My point as well was that in order to fund developers at the Eclipse Foundation, it's going to take a lot of money.  I don't think some people have been appreciating the scale of it, so the lousy idea of taxing committers hopefully illustrates that the amount of money required for even a token effort is not going to be easy to get, regardless of the source.  You can't just wave a magic wand and say "we'll ask people to donate it!"  Half a million dollars is a LOT of money.

Is there a viable way to raise the money on a recurring basis?  I honestly don't know.  I'm open to suggestions.

Wednesday, October 9, 2013

The Tragedy of The Eclipse Commons and Taxation

I've seen it proposed recently that in order to undo some of the "Tragedy of The Commons" happening with the Eclipse platform, that we ought to somehow enact some sort of tax whereby the Eclipse Foundation would be funded by the community and would itself hire developers to work on the Platform.

It's an interesting and tempting idea.  I'm not entirely sure how it would work though.  What would you tax?  How would you tax it?  How much would you tax it?

When it comes to taxation in our "real" lives, it comes in various forms.  Some taxes are assessed on things we do, such as income taxes.  Others are assessed based on things we own, such as property taxes.  Some taxes are on things we purchase, such as sales taxes, gasoline taxes, etc.

The Foundation does not have the power of law behind it the way our governments do, so they cannot declare by fiat that the community must all start giving them money, just because they say so.  They must somehow convince us that we should give that money voluntarily.  We are all already free to donate money to the Foundation, and this hasn't happened, so I think we need to be practical and assume that people aren't just going to open up the chequebook all of a sudden out of altruism.  The only plausible solution would be that the Foundation would somehow tax some service that they provide to the community, and if you don't pay, you don't play.

Some have suggested that seats on architecture councils and whatnot are the solution.  Pay the tax, and you get a seat.  We've already basically had that for years (various levels of corporate membership got you seats on various councils), and it hasn't really worked, because as others have pointed out, in a largely autonomous meritocracy like we have at Eclipse, these councils can't really force anyone to do anything.  The architecture council might tell project X that it ought to be doing Y at point Z in time, but when it comes down to it, it's almost entirely the committers on project X and/or their employers that decide what they work on and what direction their project takes.

I don't really call purchase of council seats a solution, because really this is part of how we have gotten here in the first place.  By and large, if I'm a company paying people to be committers on project X, it is likely that I'm going to tell my committers to work on the things that I care about.  Where the needs of others intersect with that, that's great, and if I'm a really generous company, I may even devote resources from time to time towards the common good, but the bottom line is that if push comes to shove, I am paying those people to service my needs first and all other concerns are secondary.  This is just a simple fact of life and no one can be faulted for this.

When it comes down to it, I think they only thing you really could tax is the one thing that really gives one real power on Eclipse projects, and that's being a committer on the project in question.  Committer status and the rights and privileges that go with it are the single most valuable thing the Foundation has to offer.  Without committer status, your power to effect change is very limited, but with it, you can not only commit changes to the code itself, but you get a say in project votes and strategic direction (the latter by virtue of the "every committer has a veto on code changes" principle).  The pros are very tangible.  Pay the tax, and you are clearly empowered to do things;  don't pay, and your ability to do things is definitely diminished in comparison.  Note that we speak not of simply purchasing commit status outright;  you would still have to earn your commit rights the old fashioned way, but in order to have your committer account in working order and the associated committer rights and privileges, you would have to pay the tax periodically.

Such a tax though would be a barrier to entry for those that have the desire to help out, but not the financial means.  For example, students and small companies might not be able to afford the tax, and companies might simply fork the code and keep their contributions private if they thought the tax was too high.  Some might not pay no matter what the price.  So is this tax a good idea?

In order to start trying to put things in perspective, let's try putting actual numbers to the scenario.  To keep things simple, let's say you decided that individual committers (i.e. that aren't associated with a member company) were exempt from the tax, but all other committers had to pay.  Looking at the dashboard, if you take out the individual committers, there are 465 active commiters left that might, in our simplified scenario, pay the tax.

If your goal is to hire say, 4 developers in the Ottawa area (where the Foundation offices are), your burdened cost for them including benefits and whatnot is going to be somewhere in the neighbourhood of $140k each, or $560k total for all four of them.  Whether or not 4 developers is even sufficient is up for debate of course, but let's use the number just for the sake of the thought experiment, because we have to start somewhere.

Dividing that $560k up amongst 465 committers, you need to tax them a little over $1200 a piece to get the required funds.  (Maybe they would even need to pay more than that, I'm not sure how donations to the Foundation would or would not be taxed by the government.)

So the big question is whether companies would be willing to pay that.  Some companies already pay hundreds of thousands of dollars to be "members" of the Foundation... would they be willing to pay this tax on top of that?  How many committer would you lose that either couldn't or wouldn't pay the tax?

I do not purport to have the answers here, but it's something interesting to think about.

What do you think?