I will just say that I’m not at all surprised this campaign is receiving some pushback. Some of the language describing it like the snippet below (emphasis mine) is almost guaranteed to elicit a negative response from some portion of the mapper population:
volunteers will be tasked with updating attributes of existing trails in OpenStreetMap using official and approved sources
There is a contingent of mappers who do not like being told what to do, and certainly don’t like the implication that official or approved sources are superior to previously mapped data in OSM. I don’t think the folks setting up these trail campaigns and initiatives mean to communicate this, but it concerns me that words have not been chosen more carefully so the message would be received positively by OSM mappers. Perhaps this whole conversation could have been avoided if the above sentence had be this instead:
Mappers will update attributes on existing trails in OpenStreetmap using maps and datasets published by land managers as a reference.
Yes @ezekielf, good call. I’m certain that “tasked with” came purely from the fact that we’re using a tasking manager to help organize the efforts for those that want to partake. And yes, as was shared in our brown bag webinar today, we want folks to use all the reference data they have, including their personal knowledge, to create the best and most complete trail data possible.
I disagree on this. I don’t like the implication that groups working in good faith to improve OSM need to craft their language to that level when the burden should be on people who have concerns to ask more questions about what those words mean rather than switching immediately to attack mode.
Word choice certainly matters, but I don’t see anything in the highlighted text that couldn’t be reasonably resolved in an open-minded discussion that starts with asking for more information, and I think that’s a fine bar for working groups and affiliates to clear in drafting language. Anything beyond that is on the people reading it, in my opinion.
I did not intend to imply any such thing. Of course nobody is ever required to carefully craft their language, but when you don’t then people can interpret your message differently than you intended. That’s all. I say this as someone who has spent a considerable amount of time participating in the trails working group (though I haven’t been involved recently). My interest is in seeing its efforts succeed rather than be met with pushback and mischaracterization simply because the messaging stepped on some obvious OSM landmines.
I can see both sides here. @ezekielf you definitely have a point that word choice matters, especially given the number of OSMers who don’t feel comfortable with any manifestation of hierarchy or authority in this context, of which there are many. It’s useful to appreciate the ways in which members of the OSM audience who share that view might read this, and I think it’s worth pointing out.
But I also agree with @Spatialia’s perspective that “attacking” is hardly the right reaction, and that questioning is.
With that said, I don’t really think anybody is attacking anything as a result of word-choice here (though, frankly, I’m not happy to have seen some of the strong nearly ad-hominems that have emerged in this thread).
Thanks - I don’t want to make it seem like you’re saying something you’re not, so I’m sorry for that. I very much respect and appreciate your contributions to this effort and the broader community.
Agreed - all parties bear responsibility for understanding each other. My point is that once the writers have put in reasonable work to be clear themselves, the reader becomes responsible for interpretation or raising questions about interpretation. Misunderstandings will happen despite best efforts of all parties, but I think how we handle them is more important than trying to eradicate them entirely. I don’t want to ask people sharing ideas to agonize over every word for fear of how a small portion of the community might react. I also know that’s not what you’re trying to suggest they do, but it’s my concern over what it means to focus more responsibility for wording on the authors - I think we need more energy as a community focused on improving how people respond to things they don’t like so that authors don’t need to worry about minor mis-wordings.
I’m also bringing some broader concerns I have with communication patterns in the community to this discussion though, in addition to separate discussions on this particular topic, so I can be quiet if I’m getting us off on a tangent - I don’t take issue with your communications :).
My objection is not with the language, it is with the risk of conflating very flawed government data with OSM (note that the data from some agencies is probably a lot better than others). My point was to provide information, express reasonably derived concern (that defective data would leak into OSM), and ask for a pause. While I can understand that people who have obviously put a lot of effort into the campaign may not have liked what I had to say, what I wrote was not an “attack.”
There are errors in the tagging/attribution in the government data in addition to errors in geometry. The government data is also missing trails, which will lead official trails being tagged as informal=yes.
I understand your concerns and that you have expertise in this area. You seem to be assuming in your posts that the trails working group will be blindly copying data from government sources without having any similar expertise to your own. I don’t think that’s a valid assumption about the group of people working on this. Wouldn’t it be better to ask the working group for more information on how they’ll determine which data source is more authoritative when OSM data conflicts with official data and what controls they’ll have in place to protect the quality of OSM data? That gives them a chance to demonstrate the same expertise.
I didn’t mean to imply that the WG members were not experts, only that I had some information that they might not be aware of. Also, just to be as honest as possible with everyone, all of my work has been done in Colorado so far, and mostly with USFS data and Colorado State Park data. Utah may be different. To be fair, the NPS data from NPS directly that I have looked at so far looks pretty good (Rocky Mountain National Park near here was updated a couple of years ago).
I am not assuming people will be blindly copying data, but in some cases there is no way to tell there is an issue, no matter how much online research one does, without a ground survey.
I will try to formulate my specific concerns as questions. That is a good suggestion, thanks.
Unfortunately, there is no sure-fire method of gauging governmental data. E.g. just computing deviations from OSM data will not work, as openstreetmap data also cannot be blindly trusted.
Download OSM highway=path/footway/bridleway/cycleway/track
Buffer by 15 meters (~49 feet). This is somewhat arbitrary, but the US National Map Accuracy Standard for 1:24K maps is 40 feet if I recall. So this gives the government data the benefit of the doubt.
Subtract buffered OSM from government data
Flag any trail in the government data where more than 40 meters (somewhat arbitrary) or 50% of the original length remain after the subtraction.
Manually review all of the above flagged trails using imagery, Strava Heatmap, USGS 3D Elevation Data, other government sources (e.g. USFS has written descriptions for many trails, which doesn’t necessarily agree with its own GIS data, also, any new trail is required to go through an approval process that includes a spatial component, and those documents are available online), my own GPS traces, GPS traces others have uploaded to OSM, geotagged photos I have taken (for name, number, and allowed uses). I also sometimes make phone calls to local officials, visitor bureaus, etc to ask about the existence of certain trails and whether they are official, what their names are, and what uses are allowed.
Any government trail that remains after the review is considered an error, although in some cases there is conflicting evidence, or no evidence, and these trails go into a third category which will require an in person survey.
In other news related to this general topic, I have been added to the invite for the TWG Zoom calls.
Not to speak on behalf of the Trails Working Group, but this wiki documentation was posted last month. I don’t see a link to this document in the announcement blog post, although it does link to static pages that include more or less the same content, as far as I can tell.
I can’t find the consultation that was done before edits started. I might be missing somewhere since I’m travelling but I didn’t see anything on the community about this before this post.
I should also point out (for anyone else reading – not in direct response to Paul, who obviously knows this), that like the OSM Foundation, OSM US has an elected, unpaid, volunteer board of directors that oversee the organization’s activities and its paid staff. We are community-led organization at its heart and soul, and this ensures that the organization operates consistent with the values of our open mapping community. If anyone feels that I or my fellow board members are not doing our jobs properly, there will be an election this winter and I would encourage anyone interested or motivated to step forward and run.
Other than the poorly-worded press release which has already been picked apart and clarified further upthread, I think that OSM US has been quite vocal about our activities for anyone choosing to pay attention. If the OSM Foundation would like to make a rule that any edit conducted by two or more editors must be preceded by a two week consultation on Discourse with a post in a certain place in a certain format, they are certainly free to do so. However, I believe we’ve met the intent and spirit of operating within community norms, including the Foundation’s guideline on organized edits.
I’ve just read through the Organized Editing Guidelines to see exactly what is considered “organized editing” and what is not. Unfortunately the document provided no clarity on the issue. The first sentence of the Scope section seemed promising:
The organised editing guidelines apply to any edits that involve more than one person and can be grouped under one or more sizeable, substantial, coordinated editing initiatives.
Uh oh! I have worked with other people on coordinated editing initiatives and some of them have been substantial. Perhaps I have been unwittingly participating in organized editing! However, in the Preamble it says:
They are not meant to apply to community activities like mapping parties between friends or doing a presentation on OSM at a local club.
Hmm, well the coordinated editing initiatives I’ve participated in certainly felt like “community activities”, but they weren’t exactly mapping parties or local club events. I guess I’d better check what “community activities” means in the context of this document. Oh, it isn’t defined anywhere. I guess it can mean whatever I want it to then. The Preamble ends with:
They are not a policy, but following them is the best way to make your organised edit successful and receive constructive community feedback.
Oh okay. So nobody is actually required to follow these guidelines at all, they are just polite suggestions.
If the OSMF intends for these guidelines to hold any weight then I’d suggest they be re-written as actual rules with clarification of exactly who must follow them and under what conditions. If a polite suggestion is all the OSMF intended, then we can all move on since there are no actual rules to comply with.
Such clarification “we” might issue in the spirit of what Zeke suggests above seems it should differentiate between the good intentions of how an individual contributor setting a good example (as I here and now state I do, when, for example, contributing various recently-introduced national bicycle routes on a regular basis, sometimes with the participation of others who “choose to enter a route,” sometimes not) and the systematic entry of data by a single mapper, but which hasn’t a well-documented process being followed.
100% of the time, me entering USBRs (and posting a request for others to participate in the semi-annual “Round” where we do this) is much more the former (it began initially a “WikiProject” to systematically correctly enter these, but by now I hope has become “a good example to others in the project” who might do something similar), rather than a “scattered” or “random” pattern of (similar) editing. The latter might also (100% of the time?) be OSM data entry we wish to characterize as “just fine, to be encouraged,” but this can be difficult to characterize. It might be “organized editing,” it might not, it might be “keeping within the guidelines and practices we wish to encourage as good mapping,” it might not.
The language to do this can be admittedly difficult to craft properly to convey what we mean. Let’s do this carefully, with deliberation and with an eye towards that there are “organized” editing practices which might walk up to an edge of needing supervision or triggering wider review, but not actually cross the line of requiring this.
Historically, that has not been how OSM(F) has operated. Just because there are no hard and fast rules that say “thou shalt not do X” doesn’t mean it’s OK to do X without a community consensus. Plenty of organisations who wanted to “improve” OSM data from Facebook down have found that out the hard way.
That doesn’t mean that there is a problem with this project, but it seems only polite to explain how the concerns raised at the top of this thread will be addressed or already have been. The linked blog post is largely content-free on that issue; if anything, it suggests that government data is treated as authoritative. Where two data sources will be used (e.g. what is already in OSM and government data about where they think trails go and what the access rights are) there needs to be some sort of conflation where one source disagrees with the other. That level of detail would fit well into an OEG wiki page; perhaps better there than on a self-promotional blog post.
With a DWG hat on I can absolutely see the logic in improving access tagging on this sort of data, and also tagging genuinely “informal” trails as such; it would reduce the number of complaints that we get. As a member of the OSM community in the UK I’ve talked to the National Trust (one of the largest land managers here) about their editing and tagging; including defending their edits against criticism from other mappers where appropriate. Sometimes, however, land managers do get it wrong, and we sometimes have to explain how to tag edge cases like “there is a legal right to walk here, but the land manager would really rather you didn’t, for safety or other reasons”.
Of course. This is the standard we all hold each other to whether editing is “organized” or not. I take issue, not with the spirit of the Organized Editing Guidelines (OEG), but with the lack of clarity . I know they are recommendations for companies with paid mapping teams and organizations like HOT. However, it sure does seem that some folks are quick to use the OEG (as if it were a hard and fast set of rules) against other community members who have decided to do some editing in an organized way. I find this tiring and not in the actual spirit of the OEG.
Again trying to add some clarity to the discussion by way of my own experiences of forging consensus and having “channeled consensus” turn into wiki I (and others, but in some cases, primarily me, as I’m standing tall on the shoulders of the consensus of many) have written. There is some organization here (as “channeled consensus,” pointable-to in wiki) while it might look to others like I’m being a single person “building something” (rail routes, long-distance bicycle routes…) in OSM. In some cases, yes, that IS what it is, a single person doing constructive, informed edits as a network of routes is built. Though, the rules and conventions on “how we do this” emerged, over years, with consensus and got documented in wiki as the bricks were laid on “how we do this.”
With US Trails, I hear many voices and some consensus, but not 100% consensus. This is natural, expected and can have its difficulties (in OSM) all at the same time. It is a process where consensus continues to emerge…on “how best to do this.” Then, that turns into “here’s how we do this,” but we aren’t completely “there” yet (for US Trails).
Good discussion is wonderful, and what this project is really made out of. Clarity is closer to consensus.
will not work, as openstreetmap data also cannot be blindly trusted. 
