OpenStreetMap

Thoughts on the how and where of the OSMF starting to hand out money in the OSM community

Posted by imagico on 31 July 2020 in English (English)

In the last board meeting the OSMF board has revealed specifics of their plans for paying people for software development work. These plans have apparently been in the making for some time - but so far the board had in public been very quiet about the details - it was first publicly mentioned in the June board meeting but just vaguely and without any specifics. That was the same meeting in which the microgrants selection has been approved.

The details were now made available just minutes before the meeting started (deja vue) - the plans amount to about half the volume of the microgrants program - but there are apparently further plans to also get financially involved in iD development - this has been hinted at in the meeting but details of this have not been revealed.

I want to share a few thoughts on the plans that came to light now - both on the procedural/policy side of them based on the perspective of for many years having critically followed OSMF work and also on the practical side and the likely implications this has on volunteer work and volunteer projects in the OpenStreetMap world based on more than five years of volunteer work on one of OpenStreetMap’s core projects - the OSM-Carto map style - and on more than 10 years experience as an entrepeneur in the cartography and geodata world.

On the procedural aspects

My critical view of the OSMF board moving more and more to making and deliberating decisions outside public scrutiny is well known and obviously applies here as well. But even if i accept this as a given and as how the OSMF works these days no matter how sharply this contrasts with OSM community values there is still significant peculiarity in the whole thing. This centers on how the three projects the board has selected now were chosen. There is no substantial communication about this but from context and from what has been communicated i see mainly the following options:

  • the projects were chosen from the microgrants applications and were invited to submit a new proposal privately to the board based on that. That seems logical for the osm2pgsql and Potlatch projects for which there were microgrant submissions that got not accepted by the commitee. This however cannot explain the inclusion of the Nominatim project (which is financially the biggest of the three). If this is the case that bears the questions (a) if the board was dissatistied with the selection of the committee - to a committee member this would be a logical conclusion and (b) that willingness to apply to the microgrants program was apparently either a prerequisite or at least a strong plus for being chosen to receive this kind of project support from the OSMF board later. Given the specific constraints of the microgrants program that made it certainly unattractive for quite a lot of people in principle in need for financial support for OSM projects that would be an odd choice. In any case this would have huge implication on any future microgrants programs in the OSMF.

  • the projects were invited based on the board members’ personal selection which projects they consider deserving and/or in need for support like this and/or essential for OSM. That would certainly be the board’s prerogative - it is the OSMF’s money and they are the board of the OSMF. But still it would be remarkable that the considerations of the board based on which this choice was made are completely outside broader scrutiny.

  • in a similar fashion - and the comments in the June board meeting point in that direction - the selection of these projects might have been meant as a personal stipend/grant to people considered valuable/deserving by the board rather than financing of projects independent of the people involved. This possibility comes with the same issue as the previous one - while the board certainly has the right to do this the lack of public documentation of the criteria and the considerations of selection would be remarkable.

  • these projects were not actively chosen by the OSMF board but had actually proactively asked for money and were granted this request. Even if that did not involve a selection by the board that would still bear the question what criteria were used to approve these requests (or if you look at it from a different side: would the board just give money to anyone asking for it?).

That leads me to one other central procedural question: Were there any other projects considered or invited to submit proposals and not chosen?

To be clear - these critical considerations do not mean i find the outcome of the process - however it took place - necessarily questionable. I know Sarah and Jochen and think they are doing good work and although i never used Potlatch i have met Richard and find his views and perspectives on things very valuable. My critical questions are on the procedure and if at all i am concerned that questionable procedure and the possibility of inproprieties involved in that might stain Sarah’s, Jochen’s and Richard’s good work. And i think a well defensible outcome does not justify a questionable procedure to get there. In the same way it does not matter how the board might retroactively justify their decision now that it has been made public - what would have been important to know is how the selection and decision process actually took place (another deja vue).

On the practical effects of financing select projects in the OSM community

What the board has publicly communicated in the meetings is that - independent on the procedural questions discussed above - the justification for supporting these three project is their significance for OpenStreetMap. I think this could have pretty far reaching consequences for volunteer work and the social dynamics around it in the OSM community in the future and this is what i want to discuss in the second part of this post. I try to keep this compact and not go deeply into the overall dynamics of the interaction between paid and unpaid, economically and not economically motivated work in the OSM community. So the following will necessarily be somewhat sketchy.

As i already hinted above when mentioning my own volunteer background in OSM-Carto there are a lot of volunteers involved in projects in the OSM community that were not selected (and probably not even considered) by the OSMF board for receiving financial support and they will all take note of that. For OSM-Carto i can say while i would not have applied for receiving money from the OSMF for OSM-Carto development even when invited to do so there might have been others who might have. The same applies probably for a lot of other projects - think of JOSM or many smaller under-the hood software projects that are not very prominent in public awareness but that none the less play key roles and are often severely underdeveloped.

In projects like OSM-Carto or JOSM of course the problem for the OSMF in financing specific design or development work would have been that they could not be sure that it would actually be accepted by the project maintainers. That is not a problem for the projects that are now supported by the OSMF where decisions are made by the same people who receive the support. Naturally for a centralized organization like the OSMF projects with more centralized control and decision making are more compatible. But that just as a side note.

What will be visible to any voluteers in projects in the OSM community - and to be clear: I am not only talking about software development here - is that the OSMF board now expresses how they value certain activities for the OpenStreetMap project by giving money to them. Many volunteers in the OSM community of course could not care less if their work is valued by the OSMF board but still it is a statement and the OSMF is sitting on quite a lot of money to make this statement with a lot of emphasis - both in where they give money to and where they don’t. Where i think we can already see this clearly foreshadowing is volunteer work directly in the OSMF - the working groups are desperately looking for people to contribute but people naturally will think if they don’t volunteer someone might be able to earn some money for doing the same work. And if not - well, then that work probably was not that valuable after all. No one wants to volunteer for something that is ultimately considered superfluous. That is the situation for hobby contributors - of course for corporate employees that is a bit different since they are not practically eligible for receiving financial support from the OSMF anyway. At the same time volunteering and the potential influence this might result in could be encouraged by their employer. And of course some professional contributors might consider working for the OSMF as a possible alternative to their current job and start viewing their volunteer work as something like an unpaid internship and a possible lead-up to paid employment. All of this massively changes the social dynamics between volunteers in the projects.

Self employed people like me are somewhere in the middle here. I tend to separate my volunteer work on OSM from my professional work quite strictly. And hence as i said i would not be very interested in receiving money from the OSMF for my OSM-Carto work. Like many hobby contributors in OSM i value that in contrast to my day job in my OSM-Carto work i don’t have to make decisions based on economic needs and the wishes of my employer/client. And even if i was willing to make an exception i would not give a discount to the OSMF like Jochen and Sarah do. Because that would mean communicating to my other customers that they are coming up for the balance. Not that the current OSMF board would be willing to give me any money at all even if their life depended on it. ;-)

But if i’d look at it from the entrepreneurial perspective the logical reaction to the board’s decision to start financing projects they deem significant for OSM would be to see the OSMF as a potential customer - which means not providing work for free unless it promises a clear chance for a timely return of investment. And i am not a software developer and the board’s choice to support three software development projects indicates clear priorities here. And of course that would also mean i’d focus my energy on work where i can make myself indispensable and to put emphasis on controlling key parts of projects. Getting down to the nitty-gritty details and working on those in a cooperative fashion, discussing decisions to achieve consensus - all the stuff that traditionally earns you respect and influence in volunteer projects - are of little benefit if your goal is to as an entrepreneur get the OSMF to pay you for OSM related work.

A large number of people will start viewing the OSMF more as a commercial actor with commercial interests now and less as an organization representing the interests of OSM as a social project and acting primarily out of social responsibility. This has the potential to massively change the social mechanisms of volunteerism in the OSM community. Since we don’t know any details about the considerations behind the decision it is hard to tell how much the board is aware of that and how much conscious strategy is behind this.

Comment from SimonPoole on 31 July 2020 at 12:38

I think it is a bit confusing to conflate this with the microgrants.

One of the rules communicated around the microgrants was that it wasn’t intended as a vehicle to finance core OSM infrastructure, and there is no doubt that iD, osm2pgsql and Nominatim belong in that category (actually this is about the only rule that was actually upheld in the microgrant project selection so we should be really grateful for that).

Once that is out of the way the question is simply if the OSMF wants to keep the software projects it heavily relies on heathy (the list is quite short in any case add the rails-port, mod-tile/renderd and I suppose osmium and that’s about it), not that there isn’t more software in OSM space that has high use, just that it wouldn’t fall in to the category of core infrastructure.

As to financing the conversion of P2, yes that should have been a no-brainer for a microgrant, and it is unclear why it wasn’t selected. Probably it simply fell through the cracks between playing HOT 2.0 and trying to make everybody happy. I would see financing this more as correcting an obvious mistake than anything to do with the more strategic questions of financing core infrastructure.

Comment from imagico on 31 July 2020 at 13:31

I am not sure how you define core infrastructure software. You include iD, osm2pgsql, Nominatim, rails-port, mod-tile/renderd and possibly osmium. But if you include mod-tile/renderd why do you not include mapnik and carto (which both have been in a precarious situation for years since mapbox has lost any interest in them)?

Comment from SimonPoole on 31 July 2020 at 14:03

Simply because there is no hope that they could be revived? I was actually on the brink of leaving all tile server/map related stuff away as, it is overall unclear how “core” they are.

Comment from imagico on 31 July 2020 at 14:31

I was actually on the brink of leaving all tile server/map related stuff away as, it is overall unclear how “core” they are.

Sure - but why include Nominatim and osm2pgsql then? If you want to define core infrastructure narrowly you should only keep what is required for the API and for generating and distributing planet files and diffs. If you adopt a wider definition including geocoding but not including map rendering is kind of a weird choice.

Comment from SimonPoole on 31 July 2020 at 15:33

Well the difference is that the osm-carto layer is just one of 4 (all OSM based), with a substantial number of other ones available and technically replacing it is a one minute change (just as with the routing engines, which we don’t operate ourselves either). AFAIK there is no alternative to the Nominatim service that is only using suitably licensed data (not to mention that there was a board decision way back to operate such a service ourselves). osm2pgsql sits a bit in the middle as it is something used by a great number of sites and is required for Nominatim.

Comment from SimonPoole on 31 July 2020 at 16:08

To be clear the tl;dr version is I was being fairly arbitrary with my list, there are obviously a lot of things that could or could not be included in to “core infrastructure”.

What is true is that the initiative doesn’t seem to be that well thought out, for example just to touch on one aspect that affects me: the OSMF, FB and Apple are competing with me for mobile editor users. Is there a scenario in which continued development still makes sense? And even JOSMs niche is becoming smaller and user number are stagnating or even going down.

Comment from imagico on 31 July 2020 at 18:22

Yes, i think you highlight an important point - that is diversity in tools. Nominatim as you said has no real alternatives or in other words: it has more or less a monopoly. Map styles and map rendering toolchains seemingly are available in large numbers and different varieties but if you really get down to specifics (community maintained map styles suitable for broad mapper feedback, tools not under precarious control of corporations - see mapnik, carto or here) things look much more bleak. And editors see a massive market concentration towards iD and derivatives. The OSMF plans should also be seen in light of this - specifically we probably have a case of the Matthew effect.

Comment from Andy Allan on 5 August 2020 at 08:56

I think it’s worth distinguishing between software that is on the critical path for data (mapper -> editing -> API -> database -> planet and replication), and then everything else. If you remove something on the critical path then OSM stops working. But if you remove something else, such as tileservers or search or routing, then sure mappers will notice but they can still put data in and get data back out again.

That’s how I used to distinguish software and systems when I was on OWG. We had three tiers - the core infrastructure required for the editing API and data distribution (and everything else involved in that, like NFS and DNS and whatnot), then tier two was stuff that would be really impactful for mappers if it wasn’t working (like tileservers and nominatim and the wiki) and then the tier three stuff was things that most mappers wouldn’t even notice if it was missing. Obviously there’s room for debate about this!

Comment from imagico on 5 August 2020 at 09:10

I agree - Note that there is another fundamental difference between tier 1 and the other tiers: In tier 1 you naturally concentrate on exactly one toolchain to work reliably - replacing tools as they become outdated and incompatible but not usually developing alternatives for permanent use in parallel. In the other tiers however having the mentioned diversity in tools tends to be highly desirable for a healthy community, incentivizing innovation and avoiding abuse of monopolist power.

Comment from SimonPoole on 5 August 2020 at 09:46

I think there is no disagreement that having a selection of tools would be nice, but is it realistic? The economic realities would say no. Lets just restrict ourselves to the 2nd most popular editor, JOSM, for the purpose of the discussion (because other examples are even more extreme),

A rough estimate is that JOSM (development, maintenance and infrastructure) costs 20 Euro / head / year, iD on the other hand 2 Euro. With increasing market share of iD and stagnating, if not going down, number of JOSM users this will get worse over time.

Now currently JOSM users outstrip the iD users in actual edits made, but the trends there are clear too, and that will accelerate with more feature parity in iD (actual feature parity is not required, there are likely only a handful of missing features in iD that are used by more than a handful of JOSM users).

In any case so while right now a case could be made for the OSMF covering the costs for JOSM. as it intends to do with iD in the medium term (1-2 years), the case will be non-existent.

A different way to look at it from an end user pov: 2 Euro a year is something you could conceivably collect from a large number of iD users, 20 Euros not.

Comment from mmd on 5 August 2020 at 11:28

In tier 1 you naturally concentrate on exactly one toolchain to work reliably - replacing tools as they become outdated and incompatible but not usually developing alternatives for permanent use in parallel.

Just wanted to point out that your statement is inheritely incorrect in case of the Rails port and CGImap. Any development in CGImap needs to have its corresponding Rails development in place. Effectively, this doubles development efforts. We do this for policy reasons to have both worlds compatible at all all times, even though one of them isn’t used in production to some degree.

Comment from imagico on 5 August 2020 at 11:40

Just wanted to point out that your statement is inheritely incorrect in case of the Rails port and CGImap.

Note the operative word here is usually. There are exceptions and there are good reasons for exceptions at times. But i would be very surprised if in this case the duplication of work was not at times put into question.

Also - as you said - only one of the two implementations is operationally active. Therefore i would consider the double implementation less a case of permanent use in parallel and more a software development strategy - possibly more related to the plan to throw one away strategy.

Comment from mmd on 5 August 2020 at 11:45

But i would be very surprised if in this case the duplication of work was not at times put into question.

As I said, it’s a policy decision, so in theory you can put the double effort into question, but it doesn’t change anything.

possibly more related to the plan to throw one away strategy.

No, that isn’t correct either. Both worlds have their merits, it just depends on the use case you’re lookling at.

Comment from imagico on 5 August 2020 at 11:58

@SimonPoole - i would not be surprised if the average JOSM user was willing to pay ten times as much for their editor compared to the average iD user, ideological motives aside (the i don’t pay for open source software attitude). The vast majority of iD users are lurkers with just one or two edits and then loosing interest. The typical JOSM user is much more serious about OSM editing, making a conscious choice to use that editor and not the default. A 1:10 ratio on average in how much they value their choice of tool does not seem completely unrealistic.

@mmd:

No, that isn’t correct either. Both worlds have their merits, it just depends on the use case you’re lookling at.

Sorry - you lost me here. What different use cases are there? AFAIK these are developed for a single use on OSMF infrastructure. What other use cases do you have in mind? Things like opengeofiction?

In any case this whole part of the discussion is besides the point of course since my point was that tier 1 usually does not call for diversity in different independently designed and developed solutions while the other tiers do. If you have two entangled implementations of exactly the same functionality in tier 1 that is a different matter.

Comment from mmd on 5 August 2020 at 13:56

Sorry - you lost me here. What different use cases are there? AFAIK these are developed for a single use on OSMF infrastructure. What other use cases do you have in mind? Things like opengeofiction?

Good question, it’s not related to any external projects, we’re only looking at osm.org topics here. Andy always wants to have a development environment that is easily approachble for new devs, hence the need to have a Rails port offering the full functionality. For performance reasons, we’re switching to CGImap for production. Both solutions have been independently designed (and their design details differ quite a bit), and serve different purposes. They’re only as much entangled, as they’re using the same database tables, and share the same functional requirements.

In any case this whole part of the discussion is besides the point of course since my point was that tier 1 usually does not call for diversity in different independently designed and developed solutions while the other tiers do

Yes, that’s exactly the point I disagree with - even for tier 1, a “one size fits it all” approach doesn’t work all the time.

Comment from imagico on 5 August 2020 at 14:09

Thanks. I think i get your point now.

And the same argument about developer approachability applies to tier 2/3 as well of course - there need to be map styles and map rendering toolchains that are easy to get started with for new designers and there need to be options for sophisticated cartography with high performance as well

Comment from SimonPoole on 7 August 2020 at 16:11

I don’t really think that a singular quirk, which in the end is due to a disagreement between Andy and Matt on architectural issues, is anything else than an exception to the rule.

Comment from mmd on 8 August 2020 at 07:24

The only other bits in tier 1 that I’m aware of are Planet file generation and replication diffs, at least according to the OWG primary services definition. It really isn’t that huge a number of tools, as an unsuspecting reader might have guessed.

Minutely replication diffs still uses osmosis, which is soon to be replaced by osmdbt. For the hourly/daily diffs, we could either stay with osmosis or switch to osmium-tool. Depending on how that decision is going to be made, we might end up using libosmium (as part of osmdbt) and osmosis (for merging hourly/daily diffs), both of which again have a large overlap in relevant functionality and have been independently developed.

I find it a somewhat difficult to come up with a general rule for tier 1 here, except for the only rule is the exception, maybe?

Comment from ᚛ᚏᚒᚐᚔᚏᚔᚋ᚜ 🏳️‍🌈 on 9 August 2020 at 13:59

(Context: I’m on the OSMF Board). The OSMF Board has not yet decided to give the money yet, we are just asking the wider OSM(F) community. The process for making the decision will be a regular circular resolution. I support this funding proposal. The suggestion to fund these came from someone on the board.

I’m not really sure what your complaint is, I think it’s “The OSMF Board hasn’t written detailed policy decisions in advance before suggesting it” and yes you’re right. We’re making decisions on a case-by-case basis here. You want “public documentation of the criteria”? Here’s mine: “These software tools are very popular, and the people behind it have a track record of delivering things the OSM community benefits from”. Is that OK? (If not, please tell me what you are looking for).

Comment from imagico on 9 August 2020 at 18:15

Rory, please - i analyzed and commented on the board’s decisions and their likely effects from someone with an outside perspective without access to privileged information on the decision making process as i pointed out but with - as also explained - significant experience with the social dynamics in the OSM community. That is not a complaint (in the sense of an articulation of a negative opinion based on something going against my interest), that is an assessment (in the sense of an evaluation based on logic, ethics, arguments and reason).

If you think that analysis is wrong because i lack important information not available to me so far i would welcome any additional information that could lead me to revise my assessment. If you think my arguments and reasoning are flawed i would welcome counter-arguments and reasoning why that is the case.

What you present here - the assessment of the chosen projects/people being popular and having a track record of actions beneficial for the OSM community - that is a domain i deliberately did not comment on and which i don’t feel competent to assess. Popularity obviously not because i have way too little exposure to non-German/non-English speaking parts of the OSM community (which is the majority) to assess popularity among them. Track record of delivering things the OSM community benefits from - it seems to me that beneficial is a subjective characterization that depends on the goals you have and the time horizon you look at. I am not saying i would necessarily disagree with those characterizations (i would from my subjective perspective indeed consider what Sarah, Jochen and Richard do predominantly beneficial for OSM - though i am sure as far as Potlatch is concerned there are also quite a few people who would see that differently) but i don’t feel qualified to make an objective assessment here - hence i left this out of my analysis and i don’t think my arguments hinge on my opinion in that regard.

As the board you are obviously free to make decisions on whatever basis you deem appropriate. But i am also free to and in fact i consider it my moral obligation to critically look at those decisions and their likely consequences as i become aware of them. That is what i am doing here. I think it would be highly beneficial if this process could be based on (a) complete information being available on the decision making process so i could spend less time on assessing things i don’t know from the past and could spend more time looking at likely consequences in the future and (b) generic auditable rules, in particular in case of money spending decisions with an impact on the social dynamics in the community so the analysis and discussion could happen largely before concrete decisions are being worked on. This would lead to easier work, higher quality reviews of board decisions and ultimately higher quality decisions more beneficial for the OSM community. Not to mention more predictable decisions of the OSMF and as a result more trust from the community. OTOH - in case some board members are contemplating that - not doing that, making decisions always at hoc, not considering binding principles for decision making and not disclosing any more details on decisions than absolutely necessary and as late as possible - will not prevent a critical review by the members or the community independently forming an own opinion on them, it would just make it much more awkward for all sides.

Comment from mmd on 9 August 2020 at 19:03

Here’s mine: “These software tools are very popular, and the people behind it have a track record of delivering things the OSM community benefits from”. Is that OK?

I have some doubts if a popularity contest will cut it here. How would you ever get some life back into those critical parts of our infrastructure, that haven’t seen any updates since 5 years, and have no maintainer anymore? As you might have guessed, I’m talking about mod_tile here. Others have pointed out serious issues in the queue implementation, that haven’t seen any in depth analysis or issue resolution at all since 3.5 years (https://github.com/openstreetmap/mod_tile/pull/152).

I don’t really care where/how osmf spends their money, or how strategic that investment is. I would judge those projects on a “critical infrastructure component that desperately needs some work” basis.

Comment from ᚛ᚏᚒᚐᚔᚏᚔᚋ᚜ 🏳️‍🌈 on 9 August 2020 at 20:48

@mmd Does someone want money to work on mod_tile? Who? What’s the proposal? How much? Serious question.

Comment from ᚛ᚏᚒᚐᚔᚏᚔᚋ᚜ 🏳️‍🌈 on 9 August 2020 at 20:53

@imagico you have access to the same amount information on these projects as I do. The applications were in ages ago in the form of Microgrant applications. You & I know osm2pgsql & nominatim & potlatch. So then someone suggests they get funding. And I thought “Yes, I know them, sounds good”. Just because I didn’t want to write up all the rules & procedures first doesn’t mean I am not guided by principles & ideas, it doesn’t mean I am making decisions on an ad-hoc basis with any thought. It’s unwritten rules & procedures, like lots of OSM.

Again, what do you actually want? What is your standard? Should I spend the next 2 weeks writing a 10,000 word rule for how exactly I plan to vote as a board member (instead of something else)? Would that be good enough for you?

Comment from imagico on 9 August 2020 at 22:39

Rory, i think we have quite a fundamental failure in communication here. I will try to explain my thoughts again in the terms that you seem to expect them in. I am not sure if that is going to help but there is really not much more i can do.

So trying to phrase my thoughts in form of an answer to your question:

what do you actually want?

What i want is the board to make good decisions for the long term benefit of the OSM community and the values of the project. I tried to explain where i see room for improvement here. If you want to boil it down to specific recommendations for actions here are the most important:

  • reveal information that is going to be public anyway at the earliest possible moment and not the latest possible moment.
  • do not stagger the disclosure of information (in this case the three project financing plans first and the much larger iD plans a few days later).
  • better yet: have the whole process in public - that would make things much easier for everyone.
  • on money spending decisions, in partiular such with a high impact on the social dynamics in the OSM community, design criteria and an auditable process for making these decisions. If you don’t think you have the time for that just ask for help - people will gladly support you with that.
  • either have the deliberation on such decisions in public or document and publish the key parts of the decision making process, in particular risk analysis that has been made on social implications and economic risks.
  • document and publish what alternatives you potentially have considered.

Again, these are not things i want individually, these are recommendations of what i consider beneficial for making better decisions, in particular by enabling others to provide more meaningful evaluation and review of your decisions - as explained in more detail in my previous comment.

If you disagree with either these suggestions or my analysis of the likely consequences of your decision i would be eager to hear your arguments and reasoning. If something in what i wrote - either in the diary or in the comments - just seems weird or nonsensical to you i will also gladly explain it in more detail.

Comment from mmd on 10 August 2020 at 09:05

Does someone want money to work on mod_tile? Who? What’s the proposal? How much?

I can’t really offer any insights into the commercial part here.

Given that the mod_tile project is quite dead for so many years, you can’t assume that people will reach out to the board, in this case simply because there’s noone left. That doesn’t mean that the topic is irrelevant, though.

Essentially, there needs to be some way to bring important infrastructure parts back into active maintenance. That could mean that you need to actively reach out to the community and possibly find some contractor.

Comment from SomeoneElse on 10 August 2020 at 13:16

Essentially, there needs to be some way to bring important infrastructure parts back into active maintenance.

In response to that I’d ask “What’s broken? What actually needs fixing?” In some cases it might be something external that’s changing (as is happening with Potlatch2) but in the case of mod_tile is that actually true?

My recollection of https://github.com/openstreetmap/mod_tile/pull/152 was that it was part of a discussion about whether what role OSMF and osm.org should play (should osm.org “compete with google” itself, or should it enable others to do so - my apologies in advance if I’m oversimplifying that; look at the board election manifestos from around that date for more background).

It’s true that the lack of an active maintainer means that things like https://github.com/openstreetmap/mod_tile/pull/168/files get ignored, which is a shame. Separately I’ve seen https://github.com/SomeoneElseOSM/mod_tile/pull/4 (which at some point will need looking at).

Comment from mmd on 10 August 2020 at 16:17

My recollection of https://github.com/openstreetmap/mod_tile/pull/152 was that it was part of a discussion about whether what role OSMF and osm.org should play (should osm.org “compete with google” itself, or should it enable others to do so

I don’t recall that part of the discussion. Do you happen to have some more details?

To put it very simple, I see this issue as revisiting the topic of which tile gets rendered at which time, and applying any optimizions as needed. That in turn might help us serve more users by making more effective use of available hardware, or even use less hardware. I’m quite confident that the core parts of mod_tile that are now about 7 years old (and the original author described as “good enough at the time”) would definitely benefit from doing some rework, e.g. use more modern standard instead of some bespoke code, as @apmon outlined in pull request 152.

Login to leave a comment