OpenStreetMap: Indispensable People?

Should anyone in OpenStreetMap be indispensable?

I think firmly not. There are whole graveyards full of indispensable people.

A project as broad and important as OpenStreetMap needs the systems in place to withstand the loss of anyone. Should someone important get run over by a bus or simply decide to move on, as many have, the project should at most be slowed a little.

Today that isn’t the case. We have points of control which are fully owned by single individuals. The people around them readily acknowledge that they don’t have a clue what we would do if they quit.

I say that’s terrible. I say they should quit and we should find out what we have to do. By keeping the de-facto in place all we are doing is kicking the problem down the road, for they have to leave at some point in the future. Let’s find out how we distribute the workload now while we only have 900,000 registered accounts instead of 9 million.

Those people should, if anything, build that process themselves. Where that doesn’t happen we should gently ask them to work elsewhere in the project on positive things. For all of the amazing work they’ve done in the past, for all the time they put in, for all of the sheer good they have absolutely contributed to the project, there should not be anyone who’s indispensable.

9 Responses to OpenStreetMap: Indispensable People?

  1. Richard Fairhurst November 27, 2012 at 9:06 am #

    This is arrant nonsense. You need to give actual examples – you know, “facts” – because from here, it looks like you’re still fighting the battles of three years ago.

    The osm.org codebase? I don’t think so: that’s just undergone the massive change to Leaflet courtesy of John Firebaugh, a bunch of UI tweaks courtesy of Saman, there’s more on the way, Pawel is bringing OWL closer to deployment and integration, Kai has got Notes (aka “report a problem”) into a state where it’s almost ready to go, and so on.

    The online editor? No, not that either: the latest Potlatch commit log looks pretty healthy; and if you’ve not noticed, there’s an entirely new online editor coming along with new contributors and a whole lot of enthusiasm.

    The hardware? Given that the top hardware guy is currently 8,300 miles from the servers, which isn’t that different in practice from being run over by a bus, yet OSM still miraculously continues to function, that can’t be too much of a single point of failure either. And then there’s the fact that tiles are now being served from numerous non-London locations.

    The Foundation? Surely not; two of the three Board members elected this year are entirely new to the Board, and for the first time in six years we have a new Chairman.

    So what are you talking about here?

    I genuinely can’t think of any single point of failure within OSM right now. Yes, there is a lot of work to be done. Yes, we need more volunteers, more contributors – of course we do. But the way to do that is by showing people that OSM values its contributors, not by repeatedly badmouthing and manuring on them for their commitment.

    • Steve November 27, 2012 at 9:12 am #

      It’s interesting that you don’t actually say anything about the point, do you think that people should be irreplaceable or not?

      I thought it best, given yours and others feedback in the past, to talk about the principle not the specific individuals. Now having done so you want me to call out individuals. It’s very difficult to have a reasonable and honest conversation about what many see as a problem when you’re so understandably emotional about it.

      • Richard Fairhurst November 27, 2012 at 9:51 am #

        It would be an interesting think-piece if divorced from a particular project and its contributors, certainly. But in that case, your first sentence would be “Should anyone in a collaborative community project be indispensable?”; the direct references such as “We [my emphasis] have points of control which are fully owned by single individuals” would be excised; and you certainly wouldn’t say that there are real people who you think should quit, using words like, well, “I say they should quit”.

        You are not so naïve as to think that you can put a post up like this and expect people not to ask “who’s he referring to?”. Of course they will, and by specifically referring to OSM rather than generic community/open source projects, you are inviting this. I’d love to read a post on “Broadening involvement: What to learn from successful open source communities”, or something like that. Say what’s been done well in Rails, or Drupal, or Linux, or whatever. Say what other projects can learn from them. This post wasn’t it.

        Personally, no, I’m not hugely interested in “a reasonable and honest conversation” for the sake of it. Others might be, and that’s lovely, but I figure OSM has never been short of conversation.

        I’m more interested in actually fixing the real-world problems that may exist. I am genuinely delighted that the OSM developer community is broadening out and that there are some really top-notch guys getting involved, some of whom I cited above. I am also really pleased that, though we may be a motley bunch of personalities (what project isn’t?), the old-timers are learning to work with the newcomers and vice versa. There is still much to do, but I repeat, we don’t help the situation by denigrating our existing contributors – whether directly or via veiled allusions.

        • admin November 27, 2012 at 9:57 am #

          Come on Richard, where am I “denigrating our existing contributors” ?

          In the last paragraph I write “all of the amazing work they’ve done in the past, for all the time they put in, for all of the sheer good they have absolutely contributed to the project” for example. Is that denigrating them?

          You have to start to accept that there are people – lots of them – who’re frustrated by how some things have turned out, how there are people who hold some keys to the castle that aren’t particularly friendly and don’t give up control as I did, and do.

          If you keep shutting down any discussion by labeling it as “denigrating” you just shush those people. And they have been. Are we allowed to allude to any problems, is everything really so perfect?

          • Richard Fairhurst November 27, 2012 at 10:23 am #

            I never had you down as such a glass-half-empty person. :)

            This is what puzzles me. Three years ago you could have made the point about “distributing the workload”. There’d have been a lot of truth to it then. No, I’m not going to pretend that I’d have reacted with great enthusiasm if you had; probably quite the opposite. (You can probably find occasions on the lists where exactly this happened!) But back then, yes, we were at the place where we were starting to appreciate that we needed more people involved, they weren’t coming, no-one had any magic bullet to fix that, and as a result, we were in a bit of a pickle.

            That’s not the case any more. There are a whole bunch of smart people involved in pretty much every aspect of the project. I’ll take the bit I know about as an example. Three years ago we had Potlatch 1 which was pretty much just me. Since then we’ve had Potlatch 2 which has a wider range of contributors. Now we’re building iD in JavaScript in which, to be honest, I’m only a bit-part player. It’s a sign of the community moving on.

            Where once we were behind the curve, we’re now starting to be ahead of it. In recent weeks we’ve seen the community identify one point of failure and self-organise to fix it. It’s really encouraging.

            So I do, genuinely, want to know where you see the problem – because I don’t see it any more. I think we’re in a better place than we’ve ever been, and we’re continuing to move in the right direction. I believe it’s by encouraging an atmosphere of trust and consideration among (and for) the developers that this flourishes, and I hope that continues.

  2. !i! November 27, 2012 at 9:17 am #

    I agree with Richard, please give examples Steve.

    Basically I agree that we need to encourage people to support in “management of OSM”, so stuff beside contributing as mapper only. This covers IMHO basically communication (inward, outward), documentation (newbies, contributors, devs, languages…), organising events, promotion…. All that stuff should be made more open to be managed by a crowd, instead of single people or small teams (as OSMF for example).

  3. Harry Wood November 27, 2012 at 10:58 am #

    Personally I kind of agree with Steve on this and I’d also say that this blog post is actually showing some attempt at being tactful (unusual from Steve :-P )

    I agree that we should aim to avoid being too dependant on any individual. I think we should work to avoid such dependencies, and reassure ourselves that we have enough collective skills/knowledge to replace any missing cog in the machine.

    Sysadmin and core component development roles have been the biggest bottlenecks in the past. Meaning they’re important roles carried out by just a few individuals. There are very good reasons for that. Having development leaders who grumpily fend off enhancement requests and poor quality patches, is just part of the open source development process as far as I can see. It can make these people look like megalomaniacs at times, but it’s necessary.

    A larger team of engineers is the key to building in some redundancy. And basically we’re seeing that happen in the various ways Richard points out. Compared to this time last year, there’s more developers and sysadmins getting involved in core components, and becoming more trusted. It’s moving in the right direction, and it’s moving gradually, which is a good thing, because the team of trusted core developers is only growing gradually.

  4. FakeSteveC November 28, 2012 at 7:47 am #

    I’m indispensable! You cannot dispense with me! No! Never.

  5. Pieren November 29, 2012 at 4:36 am #

    I fully agree with the principle. It would be very scary if, for instance, a single person would know the root password on the main server. This is something that any organization shouldn’t accept in any way. But as Richard said, once the principle is declared, it would be interesting to know if such indispensable individual still exists today in the project. Or not.

Powered by WordPress. Designed by Woo Themes