After a few “shakedown cruises” with the Max2 I began collecting imagery in earnest in August, and learned a few lessons.

1. Massive 10Gb files overwhelm older MicroSD-to-USB card readers. I struggled to copy the .360 files from the SD card to my desktop computer’s SSD drive, and resorted to Disk Drill (sometimes worked) and Roadkill’s Unstoppable Copier (sometimes worked) to copy them when copy, xcopy, and the Windows copy utility failed. I asked GoPro if it could identify the problem. When GoPro asked if I had tried a different card reader, I blew $7 on a new one rated USB-2 and it worked without problem. Issue solved, and something learned–just because a piece of hardware worked flawlessly for a decade doesn’t mean it will work today with the larger file sizes inherent to 360-degree imagery.

2. The battery lasts about 2 hours or less and a 64Gb SD card fills up in about 2 hours. Letting the camera battery die while filming is a headache, because the file isn’t closed properly. This is where Disk Drill came in handy to repair it so it could be copied and uploaded to Mapillary. A similar issue appears if you run out of storage. I blew $99 on a 256Gb SD card to avoid running out of space and put my wife in charge of monitoring battery life on the Quik smart phone app. Her reward for this is lunch at restaurants we have discovered while collecting imagery.

3. The Max2 can be toggled between collection at 5.6K and 8K resolution. I have experimented with both. Mapillary recommends 5.6K video at 30 frames per second but I have found that this makes reading house numbers difficult to impossible, and signage of businesses becomes very iffy. When experimenting with 8K resolution, I found both house numbers and signage become clearer and thus readable for addition to the OSM database. The tradeoff is file size, which quickly becomes an argument for a larger SD card (the Max2 can accommodate up to 1 Tb if you are willing to spend the money).

… See full entry

A while back a message popped up in my inbox inviting me to compete for a GoPro Max2 360-degree sport camera, courtesy of Meta, which of course now owns Mapillary and thus has a vested interest in supporting crowdsourced collection of street-level imagery. I submitted the application assuming it would come to naught and was pleasantly surprised when a GoPro Max2 appeared on my doorstep on July 15th. What follows is a review of that camera for collecting street-level imagery. Amazon sells this camera for $299. To put that in perspective, I paid about $200 for a RealMe model RMX3241 when we got to Italy for SOTM a couple of years ago and discovered that my wife’s iPhone was not unlocked as advertised, so a local Italian SIM card wouldn’t work.

Description:

The GoPro Max2 features back-to-back fixed-focus-to-infinity fisheye lenses. The built-in processor stitches those images together and stores the combined image as a 360-degree view of everything, not just horizontally but vertically as well. A smartphone app, Quik, can be downloaded to control the camera via a wifi connection (and that is vastly easier than trying to control the camera via the tiny LCD screen). The camera can collect both video and still images, 360-degree JPEGs, which can be displayed by VLC Media Player as well as GoPro’s proprietary Player. GoPro Player, a free download, is a desktop imagery editor that allows you to trim video files and also to edit the point of view of video and still photos.

After charging the batteries I did some initial filming by walking around my back yard and viewing the videos on the camera’s LED screen. That led to basic familiarity; after a short time the controls became relatively intuitive. I downloaded the Quik app to my Android phone, connected the camera and phone, and used it to start and stop the camera, select modes, and generally experimented a bit, then deleted those test videos. The Quik app is not hard to figure out.

… See full entry

I have figured out how to get mkgmap to produce fully customized, routable Garmin-compatible IMG files for my nüvi GPS navigator. Here’s the story, with the helpful hints so you can produce custom IMG files to your own specifications, too.

Background

I first got serious about mapping in OSM in 2014, during a diplomatic posting to New Delhi. I had a Garmin nüvi GPS navigator, but Garmin provided no maps for India. A search of the OSM wiki revealed Lambertus’ website that provided ready-to-use, downloadable IMG files, which proved a godsend. Fast forward to 2015 and my diplomatic posting to Turkmenistan. That story is told in a recording of my 2019 banquet speech to the North American Cartographic Society. People in Turkmenistan and elsewhere relied on Lambertus’ IMG files to navigate in countries for which Garmin provides no maps. Alas, a couple of years ago Lambertus’ website went defunct. I tried bbbike’s IMG downloads, but they leave out numerous POIs I know are either useful or necessary for both diplomats and local residents in Central Asia.

I decided to study mkgmap so as to generate custom IMG files. As an aside, when I mentioned this to a longtime OSMer friend, the response was, “Good luck with that.”

The documentation

The documentation on mkgmap is a bit sparse, and it is not really comprehensive. The base documentation found at https://www.mkgmap.org.uk/doc/index.html is enough to get you started, but is written by computer experts for other computer experts. It is inaccessible in places to those of us who majored in political science or economics.

Andy Townsend’s instructions at osm.org/user/SomeoneElse/diary/38613 proved very helpful in shedding light on how things work.

Then I stumbled across ligfietser’s style sheets at https://github.com/ligfietser/mkgmap-style-sheets. This shed even more light on how to get rendering to work the way I wanted it to.

Phase I - getting it to route

… See full entry

March 10-11 I will present a series of three Big Blue Button video conferences open to the OSM community 8 hours apart to answer questions about results of the OSMF community survey, at noon UTC and 2000 UTC March 10, and 0400 March 11 UTC (these correspond to 0700, 1500, and 2300 EST). I will present some summaries of the data in graphic form, then take questions. I will use my Big Blue Button home room at https://osmvideo.cloud68.co/user/all-3t3-ekg

I also plan to open an instance of Microsoft Translator and to speak into it in English, and will share the access code for the instance at the start of the presentation, so that anybody with Microsoft Translator’s app (on desktop, notebook, tablet, or smart phone) can follow my oral narrative in the language of their choosing. I will also use the Power Point 365 facility for rendering speech to text so that anyone who can read English can read subtitles as I speak. Anyone planning to view the video conference should download the Microsoft Translator app in advance and become familiar with it first.

The anonymized, raw results of the 2021 OSMF community survey have been posted here in .ods format spreadsheets. There are two spreadsheets, one containing anonymized raw data, and the other containing summary statistics calculated from the raw data before anonymization. I believe I have caught all errors made in the initial trials of calculating summary statistics (I had to brush up some long-dormant spreadsheet skills), but will appreciate being notified if anybody reading this notices anomalies or errors in the statistics.

I am continuing to analyze the raw data, and if there are specific requests for cross-sectional analysis not already presented in the summary statistics spreadsheet, please let me know, and I’ll try to fit them into my schedule.

We are working on normalization of the data against mapping activity by country. Since 84% of respondents identified themselves as mappers, this should help us to weight (normalize) the responses by geography. Initial examination of normalization trials revealed, however, that the differences between raw results and normalized results do not appear large. We are therefore proceeding with release of the raw data.

Please feel free to amplify this message via social media and other communications channels and in languages other than English. Many thanks in advance for your help.

With two days left to take the OSM Foundation’s community survey, we have 2835 full responses out of 4059 total responses. The survey is available in 18 languages. If you are an OSM community member and have not yet taken the survey, please do! If you know a community member who has not taken the survey, please encourage that OSMer to take it! https://osmf.limequery.org/281662 is the link to the survey, take your pick of the languages!

Please spread the word through social media, your communications channels, to the whole OSM community in your country of residence! Please feel free to translate this message into other languages and to spread it around!

And if you have taken the survey already, the Board of Directors and I thank you!

The survey now offers 18 languages (Arabic, Chinese (Traditional; Taiwan), English, French, German, Indonesian, Italian, Japanese, Korean, Kurdish (Sorani), Persian, Polish, Portuguese (Brazilian), Russian, Spanish, Turkish, Ukrainian, Vietnamese).

As of today (3 February) the survey has received 2381 full responses and 962 incomplete responses, for a total of 3,343 responses. A total of 3,875 tokens have been requested.

Responses are predictably tapering off from the peak of 461 on 23 January (the day after a banner went up on osm.org), but remain relatively high with over 100 responses daily. The survey will remain open until February 14th, so if you have not taken it yet, please give the OSM Foundation five to ten minutes of your time to share your opinions!

The OSM Foundation Board of Directors needs your feedback on actions it took in 2020 plus your sentiments regarding controversial questions, and asks for some demographic data (optional) so we can understand better both the survey data and the structure of the community. Coincidentally, the survey is divided into “Feedback”, “Community Sentiment”, and “Demographic Data (optional)”.

After seven days, the OSMF 2021 Survey of the OSMverse has received 1,918 responses in total, of which 1,424 are “full” responses (that is, they include the optional demographic data).

A sum of 1,918 total responses means the survey features a 2.24% confidence interval at the 95% confidence level, and a 2.95% confidence interval at the 99% confidence level. The survey features a confidence interval under 3%! This is good validation of the sample, though I suspect it still suffers from a European and North American bias. Male respondents also dominate the sample quite heavily. Therefore: please spread the word, especially to colleagues in Africa, Asia, and Latin America, and to women and non-binary mappers! And data users! And software developers! And systems folks! Everyone in the OSM community!

We are shooting for a 2% confidence interval at the 95% confidence level, for which we need a minimum sample of 2,401 “full” respondents, and preferably they will be distributed geographically. Since about a quarter of respondents so far have declined to provide demographic data, this implies we need to hit 3,201 “total” responses or thereabouts.

… See full entry

At the moment, after 5-1/2 days, we have 1496 responses, of which 1116 are “full”, meaning they include demographic data that will allow us to do some cross-sectional analysis (almost 29% of respondents decline to provide demographic data). But keep spreading the word. We have 16 languages available: (English, Chinese (Traditional; Taiwan), French, German, Indonesian, Italian, Japanese, Korean, Persian, Polish, Portuguese (Brazilian), Russian, Spanish, Turkish, Ukrainian, Vietnamese). If you know of a group that needs another language and can link me with someone to translate the questionnaire, please do so! I have become adept at copying and pasting languages I don’t understand into Lime Survey.

We needed 1067 respondents to achieve a 3% error term at the 95% confidence level, but now I’d like for us to push for a 2% error term. As you all know, the larger the sample, the more accurate the estimates, and the wider we cast our net, the less the selection bias. For that we need 4160 respondents. We have three weeks left in the survey so it is possible if we continue to spread the word. Please help! The URL is https://osmf.limequery.org/281662 Please spread the word in whatever other languages you can, please encourage OSMers to participate, please use all available communication channels to reach as many corners of the OSMverse as you can!

The 2021 OSMF Survey of the OSM community has been activated. It will be open for participation until February 14th. Please surf to the following URLs to register for the survey. You will receive an email in return with a single-use token that can be used to take the survey.

Please be aware that the Persian (Farsi) translation is 100% machine translation, via Microsoft Translator. I apologize in advance for any mistakes, since this is a language I do not know, and I could not find volunteer translators to edit and correct the machine translation. There will be errors, but I hope not serious mistranslations that make the survey impossible to understand. If you can help with editing the Persian translation, please message me.

Please amplify this message throughout the OSM community–spread the word! Please use your preferred communications channels to encourage everyone in the OSM community worldwide to participate in this survey! I thank you in advance for that.

English (Base language): https://osmf.limequery.org/281662?lang=en

Chinese (Traditional; Taiwan): https://osmf.limequery.org/281662?lang=zh-Hant-TW

French: https://osmf.limequery.org/281662?lang=fr

German: https://osmf.limequery.org/281662?lang=de

Japanese: https://osmf.limequery.org/281662?lang=ja

Korean: https://osmf.limequery.org/281662?lang=ko

Persian: https://osmf.limequery.org/281662?lang=fa

Polish: https://osmf.limequery.org/281662?lang=pl

Portuguese (Brazilian): https://osmf.limequery.org/281662?lang=pt-BR

Russian: https://osmf.limequery.org/281662?lang=ru

Spanish: https://osmf.limequery.org/281662?lang=es

Turkish: https://osmf.limequery.org/281662?lang=tr

=== UPDATE ===

Indonesian (Bahasa) https://osmf.limequery.org/281662?lang=id

Italian: https://osmf.limequery.org/281662?lang=it

Ukrainian https://osmf.limequery.org/281662?lang=uk

Vietnamese https://osmf.limequery.org/281662?lang=vi

Very best regards, and happy mapping!

To the OSM and OSMF communities:

I convey the following information on behalf of the Board of Directors of the OpenStreetMap Foundation.

An outcome of the current controversy on the osmf-talk mailing list over misogynistic language is a decision by the Board as follows:

The Board will find partners to help instate a moderator team for the OSMF-talk and talk mailing lists. 
These moderators need to have the trust of the community subject to the moderation (consent of the governed) by some kind of approval mechanism. 
This moderator team will start to work on enforcing the current Etiquette guidelines as soon as possible. We will also start work on updating/replacing our Etiquette rules, which must focus on balancing all participants' interests.

We have asked the Local Chapters and Communities Working Group (LCCWG) to take the lead on this and to work with signatories of the open letter to the Board as well as members of the Diversity and Inclusion Special Committee to produce proposals for the Board to consider at its January meeting. The LCCWG has accepted this task. This issue will be on the agenda of the January meeting of the Board of Directors, exact time and date yet to be determined, though as is customary it will be posted to the Foundation’s website well in advance.

Members of the OSM community are, as always, welcome to share their opinions and any relevant information on this matter, either publicly via osmf-talk, or privately in direct communications to the LCCWG. I feel compelled to remind all members of the community that a Code of Etiquette has existed since June 2011 and shall be observed by all community members.

Very best regards to all, apm

My report for 2020 has been posted here: https://wiki.osmfoundation.org/wiki/Annual_General_Meetings/2020/Chairperson%E2%80%99s_report#OSMF_Chairperson.27s_Annual_Report_for_2020

The 14th Annual General Meeting of the OpenStreetMap Foundation will be held online in the IRC chat room #osmf-gm on the IRC network irc.oftc.net, at 1600 UTC on Saturday, 12 December 2020. If you are a member of the Foundation, you are encouraged to log in.

Please see the following wiki page close to the AGM date for information on how to connect to IRC: osm.wiki/IRC You will also be able to join via your browser: https://webchat.oftc.net/?channels=osmf-gm

Agenda

• Apologies for Absence
• Minutes of the last Meeting
• Chairperson’s report
• Treasurer’s Report
• Election to Board
• Results of voting on the special resolution and the other suggested resolutions
• Any other business

The agenda can also be viewed at https://wiki.osmfoundation.org/wiki/Annual_General_Meetings/2020, where links to the reports will be available before the meeting.

The OSMF Board is asking the membership to approve an amendment to the Articles of Association that will allow Board committees (specified in the AoA as only consisting of Board members) to include any Foundation member, associate or full, to serve on a Board committee. The reason for this is that some of the Board’s administrative work, such as handling our finances, has proven very time consuming, more than one person can handle. Another sphere is the budget preparation, and yet another is fundraising. Since the Board is also hiring fulltime staff and engaging contractors, it needs help with oversight.

Some board members have been asked if this is intended to supplant the Working Groups. At least one diary entry has been posted by a community member asserting that this is the case, and urging Foundation members to vote against the amendment.

The proposed AoA amendment is NOT intended to supplant Working Groups. The Working Groups handle the substantive and administrative issues of the community, which is separate from the Foundation and the Board. The Working Groups would therefore not be affected. As I envision it, the Board committees would deal with personnel, budget, and fundraising, none of which fall in the remit of any Working Group.

I urge Foundation members to vote in favor of the AoA amendment, and then to volunteer to serve on one of the Board committees (and on Working Groups, too, but separately!)

Yesterday I had the pleasure of participating in the Local Chapters and Communities Congress, where we celebrated and Joost Schouppe ceremoniously countersigned new local chapter agreements. At the beginning of 2020, the OSMF had 8 local chapters, all in Europe. Today we have 14 local chapters, on five continents: Europe, South America, North America, Asia, and as of the last OSMF Board meeting, Africa. Joost as Board secretary has done the lion’s share of the work in processing six applications so far this year and deserves kudos for his efforts.

Among other things we discussed what the role of local chapters should be in OSM governance. Nothing has been set in stone, and indeed the conversation in a breakout session on the topic was free-flowing. Rob Nickerson of OSM UK suggested that the local chapters might be asked to help decide on admission of new corporate members, and also to help the Board prioritize paid work (e.g., to help decide when volunteer contributions of labor are not keeping up and OSM needs to pay somebody to fix broken software). I think these are ideas worth considering.

The conversation yielded a suggestion that the OSMF Board assist local chapters and communities with fundraising and perhaps help coordinate fundraising.

I have my own thoughts on this–I don’t think the Board itself should be a source of funds for local chapters and communities, because that could lead to dependency relationships that would undermine the decentralized nature of the OSM community, one of its strengths. That said, there is perhaps a role for the Board in cooperating with the LCCWG to assist local chapters in raising their own funds and to ensure that we don’t interfere with each other in the process.

I’m in Iowa on a political campaign, but took last Sunday off to tour the six surviving covered bridges of Madison County. If you are interested in a virtual tour, check out the images I uploaded to Mapillary. I used a GoPro Hero 7 Black, newly acquired after the GPS function of the GoPro Hero 5 Black refused to work.

I have finally bitten the bullet and acquired a GoPro Hero 5 Black as well as a windshield mount for it, and tomorrow will test it out for collection of Mapillary imagery. Until now I have used smartphones and the Mapillary app exclusively, but that only works if someone else is driving and I can restart the video when the app invariably either crashes or locks up at some point. Since retiring from the ambassadorship, I am now my own chauffeur and no longer enjoy the luxury of being a passenger who can serve as the photographer. Mrs. Mustard kindly granted permission to splurge on a dedicated camera for collecting ground-level imagery.

The OSMF Board decided in August to bite another bullet and resolved to finance a full-time developer of iD, the OSM default editor, plus a full-time system reliability engineer position to ensure maximum reliability of the hardware. The Potlatch maintainer also received a grant for porting Potlatch, since Flash is soon to be deprecated, and some mappers are still using Potlatch very heavily. The Board expressed a strong desire to maintain diversity in the editor ecosystem despite the dominance of JOSM and iD. JOSM maintainers were pinged but indicated that financial support for JOSM is not needed. These decisions were not taken lightly, were discussed with the community, debated within the Board itself, and in the end adopted unanimously because maintenance and support of core infrastructure are crucial to the OSM project, and demands on that infrastructure are only growing.

Today as the remnants of Hurricane Laura dumped rain on my house, I chose to stay indoors and resume reading “Land of the Turkomans”, an anthology of reports delivered in the 1800s to the Royal Geographic Society of expeditions and explorations of what is today Turkmenistan. As I delved into the reports, I realized that I had been in many of the places described, and if not, at least knew where they were. Out came the Soviet military maps, and I began trying to plot the routes of the explorers.

One noted the names of water wells as he crossed the Karakum Desert led by a local guide. It struck me. Of course, a semi-nomadic people like the 19th-century Turkmen tended not to reside in villages out in the middle of the desert, but to move their flocks of sheep and herds of camels from pasture to pasture, always bearing in mind the locations of wells for watering themselves as well as their animals. Thus the wells have names! A review of the Soviet military maps revealed that some of those wells existed at least as late as the 1980s (and probably still do). I’d never thought of that…where there are few to no villages, names are assigned to water wells.

A trip from Gokdepe to Khiva, which today might take perhaps 10- to 12 hours by automobile (including up to two hours to clear Customs at the border), back then involved a minimum of two weeks’ “march” across the desert, including one five-day stretch without water, and was only possible in practice during spring and fall, when temperate weather was somewhat assured.

(Guillaume and Allan wrote this together, and will therefore share the verbal abuse it will generate)

In its first seven months, the ‘new’ (well, no longer all that new) OSM Foundation Board of Directors has made progress on issues facing OSM and the OSM community:

… See full entry

I have completed an overhaul of Turkmenistan’s municipalities in the OpenStreetMap database. All cities are identified, plus all but four towns (those pesky four are still hiding), and 256 out of 481 “village councils” are now on the map, as well as 504 out of 1,717 “villages” known to exist. If anybody knows offhand where the four missing towns (şäherçeler) are located, please drop me a line.

Each geolocated village now also has addr:city tags so we know to which municipality it is subordinate (this is important because so many village names are used over and over again), as well as addr:district tags to aid in identification.

Misnamed municipalities or those with obsolete names have been corrected or updated, and old_name tags have been added where appropriate. Mapillary key numbers for images of signs have been added in several cases to assure positive identification.

On the wiki, the “Turkmenistan Geoname Changes” page features a list of cities and towns with hyperlinks to their respective nodes and ways on osm.org, and “Districts in Turkmenistan” features a list of villages with the same for them, wherever a village has been geolocated. This will ease finding specific municipalities by province and district.

I spent a bit more than four years mapping Turkmenistan en situ and did not want the accumulated knowledge to go unused, hence this effort to get everything I learned into the database..There is still more to do: 1,213 villages remain unidentified, and those four towns are out there somewhere. But this is a good start, and the map is in much better shape than it was in 2015.