Foundation/AGM20/Election to Board/Answers and manifestos/Roland Olbricht

From OpenStreetMap Wiki
Jump to navigation Jump to search

ABOUT YOU:

Your OSM activities

For example:

  • What brought you to OSM and why are you still part of it now?
  • What is your OSM user name?
  • What mapping contributions have you made in the last year?
  • Are you/have you been a member of any OSMF working groups?
  • Do you participate with other OSM mappers, for example in a local chapter or in social meet ups?
  • Have you run anything yourself, such as an OSM-newbie event?
  • Have you written about OpenStreetMap in the past - for example, in a blog, or on mailing lists, or in a newspaper? Please provide links if you can.
  • Do you contribute as a software developer?
  • Have you attended board meetings as a guest?

What brought you to OSM and why are you still part of it now?

I came to OSM many years ago because all in Germany available maps were awful: Maps that were at least useful for driving directions had no license beyond being printed on paper. Maps (or more precisely: geographical data) for the wildly diverse purposes produced from OSM today were so far away that people hardly dreamed of it.

I stayed because of the fantastic community:
The On-The-Ground rule resolved most potential disputes about the truth. And the decentralized structure encouraged many third party developers to come around with really useful no-nonsense tools. This in turn made mappers the life easier: for many details there is now useful feedback whether the added data makes a difference.

What is your OSM user name?

drolbr for my personal account.
drolbr_mdv for job related activities.

What mapping contributions have you made in the last year?

In my home region, initiated by the Mappertreffen_Dortmund, there is a large movement to get wheelchair routing ready for production. We are all learning a lot of important details: there is no broad agreement on acceptable incline, because wheelchair users substantially vary in their abilities, but there is a much broader agreement on which surfaces are wheelchair acceptable and which not. There are many more details, I'm stopping here before I get lost.

In addition, I'm caring primarly on keeping exisiting data up-to-date. Again, when tackling the task hands-on and appropriately deep, I learned a lot. Did you know that the best predictor whether a region needs resurveying is the ratio of reachable URLs from url tags in to totally mapped unique url tags in that region?

Are you/have you been a member of any OSMF working groups?'

I had been interested in the EWG, but it always made the impression of being inactive. From that point of view, the recent call for members is no surprise.

When the GDPR implementation nudged us, I have written position papers on timestamps and authentication to channel the future GDPR implementation in the direction most helpful for mappers.

There has also been a short encounter with the DWG:
I had tried to help with the Brazilian Turf Wars but had been unable to calm down any of the involved parties. This only heightened my respect for the long term members of the DWG. As a result, I as a board member would not have supported an overruling of the DWG as has happened with the Crimea conflict. OTRS had turned out as highly draining on motivation, thus I left the DWG.

Do you participate with other OSM mappers, for example in a local chapter or in social meet ups?

I have attended several local meetups in my home region, e.g. in Bonn, Cologne, Düsseldorf, Wuppertal and Dortmund. The Corona restrictions basically halted those activities.

I'm a regular speaker at the FOSSGIS conferences. FOSSGIS is the German local chapter of OSM. I also have given talks and workshops at various SotM conferences, and one SotM-US as well as one SotM-FR.

Have you run anything yourself, such as an OSM-newbie event?

I have run various workshops within and outside of OSM related conferences.

Have you written about OpenStreetMap in the past - for example, in a blog, or on mailing lists, or in a newspaper? Please provide links if you can.

Yes, frequently. You will find contributions from me on my project's blog, the wiki, various mailing lists, forum.openstreetmap.org, help.openstreetmap.org, the welcome mat, some diary entries, some position papers, and probably more.

I always factor in the possibility that such an artifact is the first contact of someone with OSM. For that reason I aspire to always encourage and never discourage stakeholders, in particular mappers. And I work hard to stay down with the facts and avoid theorizing or stating strategies.

Some Links:

Do you contribute as a software developer?

I am the developer and maintainer of the Overpass API.

There are also contributions of me to JOSM, the main page, and several other projects.

Have you attended board meetings as a guest?

Only a few times.
We have nowadays excellent minutes of the meetings, thanks to Dorothea. During the meeting, guests anyway cannot speak up. Thus, reading the minutes is usually the better effort-benefit solution.

Why you want to be a board member?

  • What do you think you can achieve as a board member that you can't as a regular OSM(F) member?
  • What is the most pressing issue the OSMF board should address?

What do you think you can achieve as a board member that you can't as a regular OSM(F) member?

The board has worked hard to make as much information public as possible, and the current board has an excellent track record in incorporating feedback. Thus, there is little that one can achieve from inside the board only.

However, we need people both to keep this state and push it forwards as well as truly represent the community. One important group are core developers, and Paul Norman had execellently served to represent them so far. As Paul unfortunately does not candidate again, I would like to take over this duty. I'm pretty sue that I can represent some groups of mappers as well as data consumers, but neither of these groups is a homogenous bloc.

What is the most pressing issue the OSMF board should address?

See Alan's diary for an analysis that is up to the point or mid-term to long-term goals.

By contrast, a particular urgent matter is protecting the licensing of our data from the fallout of Brexit.

Your time

If you are currently a member of a working group, do you plan to continue your role in that working group while on the board? Do you have enough time to commit to multiple roles?

Currently, I'm not member of any working group.

Do you have any previous relevant experience?

Please describe any experience you have that might help you be a board member. Here are some examples to help you:

  • Being a board member for OSMF often involves complex negotiation and discussions within the board, with working groups, and with the wider OSM community. Teamwork and the ability to make decisions, listen (truly listen) and hear a diverse set of opinions takes humility, time management, calm process planning, and community-building skills. Do you have an experience where you managed scenarios and conversations that you may not have agreed with and/or that challenged you.
  • Do you have experience of managing a project or a team of people? Do you have any experience of coaching others to lead (i.e. managing managers)? How long have you been doing these things?
  • Have you ever managed multiple stakeholders with different agendas? What was the situation? What did you do? What challenges did you face and how did you overcome them?

Bluntly, I do not buy this list of examples. The core mission of the board is to represent the community, i.e. for every mapper here must be one or multiple board members that they trust to speak to and ask to bring forward their interests within the OSM project.

Transparency: Conflicts of interest

A. Is your main source of income related to mapping or GIS work in some way, (whether OSM-related or not)?

B. Are you an employee of, member of, or otherwise affiliated with (paid or non-paid) a company, government organization or non-profit that does work in the OSM ecosystem or might compete with it? Do you have any contracts (employment or otherwise) which would limit what you can say in public that are relevant to OSM? eg a non-disparagement clause with a company/org in the OSM ecosystem? Or an employment contract which commits you to "always work in the company's best interest"?

A. Is your main source of income related to mapping or GIS work in some way, (whether OSM-related or not)?

Almost not at all.

My day job is to write software to issue public transit tickets in an interoperable yet secure way. My employer is also a heavy user of OSM data in other settings, and I'm consulting colleagues on more complex OSM matters.

B. Are you an employee of, member of, or otherwise affiliated with (paid or non-paid) a company, government organization or non-profit that does work in the OSM ecosystem or might compete with it? Do you have any contracts (employment or otherwise) which would limit what you can say in public that are relevant to OSM? eg a non-disparagement clause with a company/org in the OSM ecosystem? Or an employment contract which commits you to "always work in the company's best interest"?

I have no relations to any organization that competes with OSM.
My current employer is a user of OSM data. In that contract, there are no explicit restrictions: The company is organized in a way such that it has anyway as few business secrets as possible. A side note is that protecting valid privacy concerns of collegues (and, by the way, any mapper or other stakeholder in OSM) by contrast of course applies.

YOUR VIEW:

What to do with the face to face meeting in Corona times?

The Board has a tradition to do an expenses-paid two day face-to-face meeting soon after the election, with a focus on agenda-setting for the rest of the year. Since there will be many new members, it also offers a chance to get to know each other better. Do you think this is a good idea? Should this be rather replaced by a video conference, given the uncertainties for travelling in the next year?

I strongly suggest a video conference:
Many mapping has been stifled by lockdowns, thus travelling to mostly make good vibes will inevitably look disproportionate through the eyes of affected mappers.

What's the use of the OSMF ?

From the point of view of a small contributor, how does the OSMF helps me and could improve my "work conditions" ?

The OSMF hosts the data, both technically and legally. It also hosts a bunch of tools to build the community. The OSMF encourages on purpose third parties to develop further tools: each best idea of hundreds of smart peoply are for sure better than the ideas that of team of only seven persons can come up with.

What will you do to build a worldwide community of mappers?

My answer is my manifesto, because the task is the core mission of the OSMF.

What will you do to encourage more women leaders in OSM working groups and governance?

Remark: There are legit interests of LGBT people to get leadership as well. I refuse to restrict this to women.

As a first step, we should be bluntly honest: We as OSM have no idea what is actually keeping people from candidating for positions. Or electing them: all elected candidates last year were male despite at least on female candidate. Or contributing in general.

Honestly again, I consider the lack of diversity primarly as a missed opportunity of growth and resilience. There are probably many people out there who both can and want to engage with OSM, even if rather as a mean than an end. Our community probably could triple if we get those people involved and excited.

Various things have been tried, but we have not seen any significant progress, even after years. Again, we don't know whether we are doing not enough of the initiatives or simply not the right things. It is time to compare to other similar organizations and settings that attract minorities better, and to scrutinize whether we can learn from them. The gender ratio of math students in Germany has been eased to be almost balanced, many years after it was discovered that school teachers actively and massively talked down girl's math competences. In that case, the true origin of the problem has been entirely outside the institution where the problem has been observed. Surveying means going out on the street, and going out is not balanced on gender. Another thing is whether it is socially acceptable and subsequently economically viable to spend spare time on long term goals, and OSM is a marathon, not a sprint.

Should OSMF accept funding/donations by companies or organisations which do not want to be disclosed to the public?

Background links for context, added at the time of answers' publication:

The asset of the OSMF is the mapper's trust. Thus there must be under no circumstances any doubts that the loyality of the OSMF is to the mappers and not any donors.

On the other hand, a third party can disguise the true source of money anyway, because our legal systems at large works that way. Can we know whether the money from the Pineapple fund is actually a donation from crimials through Bitcoin? We do not know. However, the community is confident that no influence has taken place along the flow of the money, thus this donation is fine.

If a notorious attribution infringer would have funded a law clinic how to get the attribution requirement out of the license and wants to influence OSM for a license change, then this is clearly unacceptable. Tha same applies if we grow the OSMF recurring expenses to a point where we depend on the donations of notorious attribution infringers. Many other cases will lie in between.

Editing conflicts

The development of the iD editor has been classically contentious. Folks put a lot of the their time into development but made controversial tagging decisions. Do you think that they are right? What role should the OSMF and OSMF board have because it now pays an iD developer?

See a potential way out. The software resolution panel is a prospective attempt. With regard to contrained organizational resources, we should now proceed to get this thing up and running before trying something else.

How many paid staff should the OSMF have and why?

Background links for context, added at the time of answers' publication:

A responsible employment requires that the OSMF commits to large recurring expenses. Raising the income of an organization usually degrades its reputation; the last of many examples being the Wikimeida Foundation.

A second vector of reputation damage is when the OSMF picks one of multiple competing projects. Value created through volunteers on software development for OSM is estimated to be in the range of many millions of dollars, thus any move by the OSMF that discourages software development elsewhere can easily make an order more damage than the OSMF could ever spend.

Hence, the real question in every single case should be: Is the task really so important for OSM that it is worth the degraded reputation for the OSMF? One charming way out of this is to assk the Community!, in particular before the person to be employed is known: There is no better authority to discuss and confirm or dismiss that the position is proportionate than the OSMF's members.

Your views on the use and control of AI (Artificial Intelligence) edit systems?

  • Do you have any familiarity with OSM AI systems? (e.g. Facebook AI-Assisted Road Tracing, RapiD)
  • Do you support the development of it?
  • If the use of AI systems causes damages on existing OSM geographical data (e.g. in Philippines) do you see any needed activity from OSMF side? If yes, which solution the OSMF should provide to control this and organised editing?
  • How is your general opinion about automated edits?

Do you have any familiarity with OSM AI systems? (e.g. Facebook AI-Assisted Road Tracing, RapiD)

I do not buy the marketing term "Intelligence" at all. I'm familar with the development of AI since the discussion about symbolic and non-symbolic AI back in the 90ties. A tool is only serious if it has a clearly delimited scope, e.g. road tracing (suggesting geometry from satellite images), but not road classification (guessing the value of the highway tag), warns that it is heuristic and calibrated in region X, and does not bother the user with implementation details. If it claims to be "intelligent" but does not tell you with images of which regions it has been confirmed to work, then be wary.

Do you support the development of it?

The most recent systems had a poor ratio of development diligence to marketing budget.

If the use of AI systems causes damages on existing OSM geographical data (e.g. in Philippines) do you see any needed activity from OSMF side? If yes, which solution the OSMF should provide to control this and organised editing?

Please look at the larger picture: We live in a work where many organisations have some form of geodata, from an address list of their branches to multiband satellite images. I have seen hundreds of such datasets, and none of them were free of errors. The organizations can even for themselves only lift the true potential of their geodata if they cross-reference with a universal peer-reviewed geodata database, and this is OSM. This applies even more if they want to give back and publish the data in a useful way. Thus, there is a constant pressure to pour bulk data into OSM.

The long-term solution is to offer two databases:

  • the peer-reviewed database that today is the sole database
  • a heap for bulk data such that data from third parties becomes interoperable without tainting the peer-reviewed data

The challenge is to develop software to connect both databases smoothly, i.e. allow for cross-referencing from the heap to the peer-reviewed database, deal with contradictions within the heap, and make it easy to promote data from the heap to the peer-reviewed database when confirmed by mappers.

A short term solution is to clarify roles:

  • our first line of defence against damage is the peer-review of the mappers.
  • our second line of defence is the DWG.
  • the OSMF is a merely legally not factually competent third line of defence

The only case so far where the OSMF as third line of defence has been involved is the Crimea conflict. The course of events were not exactly helpful to build trust, thus I prefer to avoid any other case where the board of the OSMF becomes involved in mapping decisions.

How is your general opinion about automated edits?

Writing software is much easier than understanding large databases, dealing with corner cases, handling internal contradictions, and, most importantly, work out with the databases' stakeholders how to proceed safely.

A similar situation exists with car traffic: pushing throttles and spinning the driving wheel is much easier than understanding physics and the needs of fellow street users. Part of the solution for car traffic have been obligatory licenses for driving. Our nowadays much less bureaucratic approach of just doing some communication might be good enough for the moment, but unless we are quick enough with the heap database, we may sooner or later need to tighten the permissions.

What is your opinion of the proposal for a software dispute resolution panel?

Background links for context, added at the time of answers' publication:

The whole thing makes only sense if both sides, mappers and developers, trust the commitee. But it is at least a thing worth trying.

Independend of this candidature, I volunteer for the panel, building on my long experience both in software development and mapping.

Should we do anything about EU database rights?

The OSMF is incorporated in the UK. The UK completely leaves the European Union on 31 December 2020 and so EU database rights held by UK entities are impacted. Do you think that changes that keep our license clearly enforceable in the EU (e.g. moving the OSMF) are important? Would you give them high priority for your involvement in the OSMF board?

Definitely yes:
The attribution requirement is substantial part of the motivation for many mappers. The share alike provision helps to be taken seriously, otherwise the attribution requirement would lose too easliy against politics considerations of large organizations.

It looks like we are falling out of legal protection in the EU if no action is taken. A definite answer is difficult because Brexit negotiations are ongoing. But I honestly do not have a straightforward plan to fix the problem, in particular which part of the OSMF needs to move where.

Manifesto

About Me

I'm Roland Olbricht (drolbr). In OpenStreetMap, I'm best known for the Overpass API. I have developed it and maintain the public servers that are available now since more than eight years.

My largest mapping project so far has been to map and complete the about 1500 bus stops in my home city Wuppertal. For that purpose I have written (many years ago) the public_transport plugin for JOSM. Now my mapping interests are in particular on pedestrian routing, in particular for non-standard people like wheelchair users. Part of the challenge there is to figure out what is actually relevant to map, thus communication on top of surveying.

I'm Dr. rer. nat. since 2008 for my works in Pure Mathematics. Since October 2014 I'm a software developer at Mentz in Münster (Westfalen), Germany. My business account is drolbr_mdv.

My day job is developing software for public transport ticketing:
It is quite common in Germany that dozens of operators cooperate to offer a uniform combined tariff to the passengers. Thus, dealing with multiple stakeholders with often conflicting interests is part of my day job. And even more: helping out customers whose tariff data turns out to be so full of faults that it needs to be sanitized or completely rewritten, although the customer is sure its data is perfect. This is always a communication challenge.

Beside living in Wuppertal, Münster, Frankfurt and Bonn in Germany, I have also spent four month in Grenoble in France. So I'll answer mails in German, French, and English.

About OSM

More important than my personal background is my understanding of OSM: There are many sources of free data, and geo data in particular.

The heart of OSM is the mapper community. The data structures in OSM are designed to facilitate peer review, i.e. the mapped data is not so much the physical world but the consent how to interpret it. The On-the-Ground-Principle ensures that it is as close to the physical world as useful.

It also means that we offer a virtual venue to communicate about general purpose geodata, and to help out people who want to get done something new or specific. It also means that it is a place where a developer can make a real difference by helpful software.

In addition, OSM offers fine grained general purpose geo data in a worldwide uniform format, and means to stay always up to date. Thus, we are attracting data consumers as long as we ensure that the mapper community maintains the map.

Mappers get their positive feedback from applications where they can see that their contribution makes a difference. This is a task often overlooked by data consumers:

  • Not all data consumers have the attribution as clear as we mappers require by the license.
  • Very few data consumers tell explicitly which tag structures they process and which not. We see this even within the project by disputes over the map style rules for the default map.
  • We are missing out many mappers, in particular from minorities, because there is no respective service that shows the data that really matters to the potental mappers.

Why Do I Candidate for the Board?

There are two competing paradigms about what the board should be:

  • The expert paradigm: The assumption is that the voters should elect the super-experts of OSM, and that those do not have to take too serious working groups because the board members were anyway greater experts. It were ok to have a hierarchy of informedness.
  • The trust paradigm: Voters should ensure that there are members on the board whom they trust that they represent their interests. Expert knowledge is more likely to indicate a conflict of interest than being a requirement.

I subscribe to the trust paradigm:

  • Keep the trust: The board's communication is only open enough if the community has no doubts left.
  • The experts are at working groups: The board will not override the (recent and concious) decisions of active working groups. Beside working groups, reaching out for experts for ad hoc topics on the OSMF communication channels makes sense as well.
  • Sustainable funding: Assure that OSM will remain here and stay independent for many years ahead.

I'm confident that I can speak up for developers in the OSM ecosystem (currently assured by Paul Norman), and I will do my best to speak up for mappers, but I'm aware that interests are diverse enough that conciliation like in the iD case may also be required. I'm happy to help data consumers to understand mapper's needs and OSM intrinsics.

I do on purpose refrain from stating strategic goals.
The active board of this year has started quite a number of initiatives, and external shocks like the Brexit make the to-do-list even longer. It is an act of good teamwork to let existing ideas flourish and getting things done within the trust paradigm instead of throwing in further ideas and completing none.

Retrieved from "https://wiki.openstreetmap.org/w/index.php?title=Foundation/AGM20/Election_to_Board/Answers_and_manifestos/Roland_Olbricht&oldid=2066132"