OpenStreetMap

iandees's diary

Recent diary entries

Running for OpenStreetMap US Board

Posted by iandees on 7 December 2016 in English (English)

I'm excited to be running for a board position in the OpenStreetMap US organization.

This has been a busy year for the board! Most of my work on the 2015/2016 board has been behind the scenes, making it easier for future boards to do their work. After a lot of hard work we were recognized by the IRS as a 501(c)(3) tax exempt organization, allowing us to more easily accept donations from organizations and funds that require such a status. We also spent a lot of time updating and modernizing our financials and now have regular support from a bookkeeper, allowing the board to focus more of their time on moving the organization forward. Additionally, I rewrote the membership processing system so that we could send reminder emails when automatic membership renewals failed.

We ran a very successful State of the Map US 2016 event in Seattle. With the help of a really strong volunteer community in Seattle, we brought over 400 OpenStreetMappers together to share their experiences.

When I ran last year, I had lots of great ideas for how we could improve the OpenStreetMap community in the US. I still think these are great ideas, but some of them require a bit more work on our end before they'll become a reality. A side-effect of building a board out of very active members of our community is that those board members don't have time outside of performing board activities to plan events or send newsletters. We should encourage those outside the board to participate and lead the effort to make these things happen rather than trying to do them ourselves. This strategy works fairly well for the State of the Map US event, and I'd like to see it applied to other things like scholarships, event sponsorships, and newsletters.

Another important piece that I think would help the organization is to hire someone to help coordinate these volunteer efforts. For example, a large part of my time on the board was spent handling payments, scholarships, donations, sponsorships, and memberships. Others on the board spent a lot of time drawing up, proofreading, and publishing communications. Hiring someone to help with these important aspect of running the organization would give the board more time to work with volunteers and would also lead to a better sense of continuity between the different boards.

I look forward to helping build the OpenStreetMap community in the United States. Please become a member and vote to make sure your voice is heard.

Location: West Nelson Street, Lexington, Virginia, 24450, United States of America

Running for OpenStreetMap US Board

Posted by iandees on 12 October 2015 in English (English)

I'm running for the board of OpenStreetMap US.

As the current board's treasurer, I've worked hard to keep our books up to date during the largest State of the Map in the world. I've also spent countless hours working with bookkeepers and accountants to make sure the financials of previous boards are sound as well. I've made sure the membership roster has been kept up to date and have tried to dump as much historical/institutional knowledge of how OSM works and has worked as I can into places where other board members (current and future) can use it to help make their lives easier.

Going forward, I agree with several of my running mates: we need to increase the number of active mappers in the United States. In the past, the main focus of OSM US has been to run the best State of the Map in the world and I think we've proven ourselves in that respect. We've brought together more of the community than any other OSM event every year for the last several years. But even though we bring existing mappers together in increasing quantities, we aren't doing a great job of introducing new people to OpenStreetMap and having them stick around to continue mapping.

I think we should use our track record of running great events to put together several (at least 2) smaller State of the Map events around the US. We should lean heavily on existing strong communities and community organizers to help pull them off while contributing the board's time, money, and effort to make them a success. The goal of these events should primarily be to showcase local mapping efforts while also inspiring with exciting national and international OSM projects, events, and tools.

Having larger events should not detract from continuing to fund more and more mapping parties and local groups. In recent months we've called them "micro-grants" and have funded one or two events, but I'd like to see us hear from dozens of events around the country. The photos and stories from excited community members would make great material for a newsletter for our membership:

We should also organize and send monthly (at least) newsletters to our considerable mailing list of current and past members. Past boards (including myself) have mentioned doing this several times without actual implementation. My expectation is that as we reach out to and encourage local events we'll have more to share.

In short, OpenStreetMap US should shift its focus from successful, large events once per year to building more communities of mappers around the country. The goal should be to grow and support existing communities and do everything we can to help foster new communities.

Location: White Street, Lexington, Virginia, 24450, United States of America

Running for OpenStreetMap US Board

Posted by iandees on 1 October 2014 in English (English)

Hello! I am Ian Dees, a current board member running to be a part of the next board. You can read more about my credentials in last year's diary post.

This year, I'd like to continue our focus on increasing diversity in OSM through scholarships, education, events, and communication. After lots of technical challenges around maintaining our membership and interest contact lists, I think we're finally ready to start sending friendly, community-oriented newsletters. I'm also excited to help the board and local community plan and execute one of the best OpenStreetMap conferences in the world.

OpenStreetMap has made huge gains in the United States. With OpenStreetMap US's help, I think we can make it even better, and I want to continue leading the way. With your vote, I think we can work together to make that happen.

Thank you!

Location: South Main Street, Lexington, Lexington City, Virginia, 24450, United States of America

Running for OpenStreetMap US Board

Posted by iandees on 4 October 2013 in English (English)

Hello! My name is Ian Dees and I'm running for one of the five OpenStreetMap US, Inc. board member positions.

As a member of the 2010 board I solicited hardware donations that gave OpenStreetMap US it's first web presence and provided utility to the community. Later on I purchased and configured the hardware used to host numerous helpful tile layers and data for the community.

Outside of the OSM US board, I've contributed hundreds of hours of my time to improve the map: I've run several mapping parties (in Chicago and Minneapolis), talked about OpenStreetMap at several conferences around the country, guest lectured in University classes on OSM, and contributed code to the project (jxapi, JOSM notes, fieldpapers, and mbtiles plugins to name a few). I've also committed tons of data to OSM: the city of Chicago has very nearly complete address coverage due to my year-long work to get the city to release its buildings data with a free license.

As a future board member of OpenStreetMap US, I will continue to work diligently to improve the map while advocating for its use wherever I can. Specifically, I will:

  • Push for using organization resources to collect, collate and properly license address data from as many municipalities as possible in the US
  • Work to increase visibility and diversity of the OpenStreetMap project in the US by reaching out to organizations like Boy Scouts, Girl Scouts, Sierra Club, and local student organizations
  • Make it easier for those with questions about OpenStreetMap to get answers from a knowledgable, friendly person in a structured manner (advertising a phone number, e-mail address, or using something like UserVoice)
  • Improve communication with our community: an e-mail newsletter pointing out great events, particularly well-mapped areas of the map, and community-contributed writing about OSM.

OpenStreetMap in the US has improved by leaps and bounds since I last sat on the board and I hope to have your vote so I can help it continue it's stellar growth.

Google Doc Summit 2011

Posted by iandees on 22 October 2011 in English (English)

Over the last week (17 - 21 Oct), a small cadre of OpenStreetMappers joined the Google-sponsored Doc Summit on the Google Campus in Mountain View. The Doc Summit is part of the Google Summer of Code and occurred immediately before the Mentor Summit. Kate Chapman, Ian Dees, and Shaun McDonald represented OpenStreetMap while members of KDE, OpenMRS, and Sahana Eden along with some unaffiliated document specialists participated as well (for a total of roughly 30 people). The objective of the Doc Summit was to participate in a "book sprint" facilitated by members of the FLOSS Manuals community, writing an open source and Free book for the community to use.

We began with an unconference on Monday discussing how book sprints work led by a very exciting and inspiring guy named Alan Gunner. His excitement and energy made for a great experience overall, but he is especially good at making these events fun and interesting. We learned about how a doc sprint works, what sort of things to write about, about the FLOSS Manuals platform and tool, and learned about each others' projects.

The following three days we spent heads-down in our book. The first day we defined our table of contents. This forced us to define the scope and audience for the book. This turned out to be the trickiest part for us since we didn't have a very good idea of what we wanted to discuss. Since Kate's LearnOSM.org project was well written and a great starting point, we thought the book should start out as a beginner's guide for OSM. Shaun and I wanted to include some more technical information that has not been well documented in other areas. After a 1 hour brainstorm and some help from external facilitators we ended with a great outline for our book.

It was especially useful to discuss a couple personas to help define who this book should be written for. I don't remember exactly the stories behind the personas, but one of the "floaters" Anne Gentle was particularly helpful in nailing down this part of our book's scope.

Once we had the scope of our book defined, we began writing content for the chapters. We spent the rest of the first day writing and moving content from learnosm.org over to the book. After the first day we left Google's campus at 22:00 and had roughly 9000 words in our book. It was a surprisingly successful day and gave the entire group a boost of confidence that we could actually write a whole book in a week. After some more writing, we had a day of review and cleanup on Thursday and a quick unconference/discussion/reflection on Friday morning. We published our books to the on-demand publisher lulu.com that evening (over some beverages of celebration) and called it a night.

Google's Open Source office was kind enough to purchase 20 copies of each book for the organizations to hand out as they see fit and I have those copies today at the beginning of the Mentor Summit. You can purchase a copy of the book from lulu.com and participate in further revisions on FLOSS Manuals.

Thanks again to Google's Open Source Office for funding, FLOSS Manuals and Aspiration for organizing, and to our fellow open source organizations for making it such a successful event!

Location: Amphitheatre Parkway, Mountain View, Santa Clara County, California, 94043, United States of America

State of the Map 2011 Summary

Posted by iandees on 29 September 2011 in English (English)

Some thoughts a few weeks after I got back from State of the Map:

Since this was my first State of the Map I don't have much to compare it to, but I felt like it was the year for OSM to dip into the "Trough of Disillusionment" and start moving up the "Slope of Enlightenment" (see [0] for the Hype Cycle). After Haiti and similar disasters surprised everyone with the flexibility and instantaneous usefulness of OSM, there was a lot of interest in trying to mold OSM into something that could be applied to other situations and give a similar result. At the State of the Map US conference last year we heard from tons of GIS departments that wanted to integrate their data into OSM. After a long while of not coming up with technical solutions to this problem, it seems like those users of OSM data have lost interest a bit and commercial entities have started to express interest.

MapQuest, Bing, and ESRI (among others) all had a presence at State of the Map this year and were showing that OSM data is useful to their fight against Google's surge in geo data. After a huge push earlier this year (http://open.mapquest.com, routing, map tiles, and other developer services/APIs all based on OSM data) MapQuest's interest in OSM as an alternate data source seems to have faded a bit as AOL is trying to find their way in a challenging market. Bing hired Steve Coast (the founder of OSM) earlier this year and has slowly rolled out support for OSM by hosting tools to help mappers create more data for OSM. Most importantly they also gave permission for OSM mappers to use the Bing satellite imagery while mapping. ESRI showed off their OSM editor plugin and related services.

As OSM data matures and becomes more complete, it seems like the focus moves from the mappers (there weren't any new announcements geared towards the mapping or developer community) to the data consumers: not only did the commercial entities I mentioned above express interest, but developers from firms interested in using OSM data in their products were there and discussed ways of making it easier to use OSM data. For example, Mike Migurski from Stamen came up with a way to create weekly dumps of OSM data on a metropolitan-area level [1] rather than forcing everyone to download the 18GB dataset that includes the entire planet.

We also heard from some researchers who discussed various methods of improving the existing OSM data, especially in the US. The holy grail of OSM (at least for places like MapQuest and Bing): complete address and routability is within reach and seems to be getting closer every day.

We just need more mappers!

[0] http://www.gartner.com/it/page.jsp?id=1447613
[1] http://metro.teczno.com/

The Java XAPI

Posted by iandees on 26 January 2011 in English (English)

I recently released the first version of a mostly-complete Java implementation of the XAPI server. The source code is available on my GitHub with a readme describing generally how to get it set up.

Running this database with a full-planet requires some large, fast disks. The database is roughly 350GB and is slowly increasing:

CPU and memory usage are useful for the initial import (indexing especially), but any subsequent queries will be limited by the disk speed.

Feel free to use it! Serge W. created an XAPI UI framework that I set up at http://openstreetmap.us/uixapi/xapi.html to build the XAPI query URLs.

At the moment, the Java XAPI seems to have a problem returning relations and doesn't implement child element predicates or special tags. If you find other problems, please file a bug.

Generating a Planet PBF

Posted by iandees on 23 December 2010 in English (English)

After some discussion on #osm IRC, I started running the following command:

curl -L http://planet.openstreetmap.org/planet-latest.osm.bz2 | bzcat | osmosis --read-xml file=/dev/stdin --write-pbf file=planet-latest.pbf

After 15 minutes or so I have a 1GB (and growing) .pbf planet file. I'll probably generate these weekly and put them somewhere with a quick network connection. Stay tuned if you're interested in a planet-wide .pbf.

Only 400 Million Maps? We've Got ESRI Beat!

Posted by iandees on 23 December 2010 in English (English)

Yesterday Tim O'Reilly made some waves by posting a tweet from within ESRI who claimed that they served 400 million user-generated maps in October. Needless to say, this number was called into question by those that heard it.

My theory (along with other commentors and tweeters) is that they are counting each map tile generated for this number. Based on this assumption I asked User:Firefishy to dump some logs from the main tile server so we could compare our "map generation" statistics with theirs.

Let's look at the graph:

Tile Server Request Graph

For the last two months, our tile servers "generated" 50-75M "maps" each day. Over the course of those two months we served over 3.7 billion maps.

I think we win, ESRI.

(Data is here if you're interested.)