OpenStreetMap
Versão em Português link
Disclaimer: All the text you read below is based on an online research aimed at the Brazilian market. survey link
It isn’t news - or shouldn’t be - that OpenStreetMap has a great value in sharing local information. It is a social impact tool for minorities and humanitarian aid policies. Despite the big efforts thousands of mappers who dedicate themselves every day to improve the maps, added to the diligent efforts by the BigTechs to promote their own services, the OSM is far from being the most ppopular map in the world. Or, at least a reference as big as its sister, Wikipedia.
A Hidden Gold Mine
Out of a group of geotech enthusiasts, the OSM is totally unknown to the general public. As we could see in the survey, about 68% of the participants had never heard about OpenStreetMap. Although most likely having had previous experiences with one of the various contact points (Tesla, SnapMap, BigMaps, Apple Maps, Facebook, Strava, Mapbox), most people still didn’t know about OpenStreetMap.
If you’re concerned about whether people see or not your work on OpenStreetMap, the answer is simple as that: Of course they do. They just don’t know where it comes from!
The main reason for this omission is that companies often try to hide our contributions in the smallest and most obscure lines of terms of service while building their own proprietary database, giving grumbs to the community. No need to name names, right?
But this isn’t an exclusive practice of large corporations. How often do apps like Maps.me, OSMAnd, OrganicMaps stop to remind their users that the data they’re using came from OpenStreetMap?
Truth is, no one wants to tell you about their Goose that laid the Golden Eggs!
Not everyone wants to map
In our survey, 47.4% of respondents don’t have the habit of contributing with the map they use. This doesn’t come as a surprise, after all, not everyone has the patience or even the interest in mapping. According to another study of pirâmide de conhecime (https://www.gislounge.com/defining-the-role-of-gis-and-needed-skills/)nto em GIS, the more knowledge is required for the task, the less the number of “enabled” people”.
OpenStreetMap took a big leap with ID to popularize mapping. Tools like Zapto (https://zapto.openstreetmap.org.br/) to add a locati (https://zapto.openstreetmap.org.br/)on, and initiatives to bring in more collaborators like Youthmappers, Mapathons, etc., [falta informação]
Within the scope of the common user, there is still a big barrier so they can effectively contribute as a costumer. If you don’t bring the product for as many people as possible to use, there’s no way to know what to update or add on the map!
Google’s omnipresence
It’s not a secret that BigG dominates our lives. From the searches you do, the way you access the internet (you’re probably using Chrome right now), to how you get around in your city, it’s there.
There’s just no functional alternative to Gmaps that works in Brasil, and even those who try to replace it end up having Gmaps as a backup. BigG conquered this monopoly using its position in the smartphone market and most popular search engine. It has since established itself as the only map provider for many regions, and with that many companies who need data have no other option but to surrender to BigG’s domain and use Gmaps in their products. This often takes the OSM to the bench, as a map post to backend or a cheaper alternative.
The lack of alternatives able to compete directly is glaring, even considering OpenStreetMap’s avid community of mappers. As of today, I am not able to recommend any of the avaliable clients to a colleague or friend as a better alternative to Gmaps.
People want to find places!
GIS Lounge Defining the Role of GIS and Needed Skills A recently published open access article presents three case studies to help analyze the level of GIS expertise needed for interdisciplinary studies. If someone asked me “what is my biggest disappointment with OpenStreetMap customers?” I would say that I have difficulty finding an address!
This is a serious problem in Brazil and other countries that insist on maintaining their proprietary address bases, which cannot be imported directly into the OSM.
Among those interviewed, 55.7% said they found it easier to search for addresses online. Can you imagine one of these people using one of the OSM-based apps to find an address she doesn’t know and that no one has bothered to map before?
This leads the user to frustration and ends up turning the OSM into a tool that cannot fulfill its own basic function.
Projects like Open Addresses (https://openaddresses.io/) try to fill this gap, but their data source in Brazil is not very accurate.
OpenStreetMap itself never wanted to be a flagship, and I doubt that will change in the short term. We need someone to guide us forward, but don’t forget those who came first.
In short, we need a client made for the Brazilian reality…
Discussion
Comment from lyx on 29 August 2021 at 08:58
Hi, even the best possible client can not find data that is not available. To make searching for addresses functional there is no other way than to acquire these addresses. If the government does not want to make their address databases available, you could try other sources. Home owner associations, chambers of commerce and the like probably are interested that customers find their member businesses or delivery drivers there houses. If road geometries and building outlines are already mapped, they could just place notes on buildings and leave the information for mappers to add. If road geometries are missing, how about asking the people driving garbage collection trucks to take a gps logger along on their tour for a few days. I’m sure you can think of other ways to get information from people who have some and might have an interest in finding it on the map.
Comment from PlayzinhoAgro on 29 August 2021 at 11:44
Hi Lyx, thanks for the comment,
Yes, I agree and have been making an effort to do so, always adding POIs and buildigns footprint, so that it can be added the addresses using street complete.
For clients, would it be so complicated to merge the data from OpenAdresses in their geo encoders … even if it is a stopgap measure I think it would impact a lot on the user experience, ultimately who knows even acquire this database with the government that is for sale.
I believe it would be a minimal investment for a client and can also bring great returns as the sale of geocoding services to third parties.
Comment from SimonPoole on 29 August 2021 at 14:36
@PlayzinhoAgro openaddresses lists quite a large number of data sources in BR, supposedly with “attribution only” licenses, are these available on OSM compatible terms or not?*
Further I think the comparison with google is a bit Apples and Oranges, a more correct question would be to ask for googles data suppliers, a question that would likely get essentially blank stares. That said IMHO the result for OSM is far better than expected.
Wikipedia while in many ways a similar project differs strongly in its “business model” and how its contents are distributed, not to mention that it operates in a very different, non-competitive market.
Simon
Comment from DeBigC on 30 August 2021 at 12:13
I realise you raise a wide variety of issues. Forgive me, as I only have sufficient time for one. “People want to find places”, absolutely right they do!
A collective decision to work on things that directly fix that, as a stated priority would help. I don’t mean “tell everyone to map X”, I more mean a consensual and more visible high priority to be set, which acknowledges that everyone might have their pet projects, but the community asks for a more specific effort to be front-loaded.
In my view we should be concentrating on adding buildings, and adding their addresses. In my experience nothing other than this stimulates macro mapping of landuse and micro-mapping of street infrastructure as much as clumps of well defined buildings. This is what consultant-speak usually means by “scalability”, - activities which when completed support and make other activities possible and make them easier or more focussed.
I also think that this is what might convert the 47% who are part of the community into higher levels of contribution activity, and would allow the Foundations, chapters and community meetups to convey a coherent narrative about what we all need to work on. It might even help to stimulate recruitment.
Comment from PlayzinhoAgro on 30 August 2021 at 15:41
@SimonPoole, hi… What I say in the text is that the OpenAdresses data could be merged by the OSM app into a geocoder. About the comparison with Google I don’t think it is different, I think that for OSM to grow it has to be more than a back-end tool. If we want OSM to grow we will have to leave this passive position and start investing in the end user as well…
Comment from PlayzinhoAgro on 30 August 2021 at 15:52
Thanks for the comment DeBigC, If you can take a look at my old texts you will see that in the last year I have been trying to build a base with mapped buildings, in Brazil there are not many incentives for recruitment and considering the financial situation of much of the country we have even fewer people to collaborate voluntarily. What I advocate is that we find an effort in mapping along with palliative solutions by the Apps for OSM.
Comment from Aury88 on 30 August 2021 at 19:45
I see 3 other big problems:
One big problem with our map is that our main tile render displays the local name and not the (user) localized name…so when we explore the map often we are not able to identify easily what we are seeing and almost over the half of the globe is displayed in foreign/an-know language/characters for the user/consumer … this is confusing and not very helpful.
Secondly our main website is seen as a service provider rather than the portal serving the QA tools for the mappers and, let me say it, it is seen as a poor experience for a consumer perspective…
Third problem is that business have a clear advantage to be displayed on GoogleMaps so they are the first that put their data on that map. In osm that advantage is not so clear, and their information can be vandalized/changed without their consent by anyone… whe don’t have stable node/way/relation id so any information in our database can be changed anytime by anyone…so why put effort in a so incontrollable database?
Comment from PlayzinhoAgro on 30 August 2021 at 20:51
Thanks for the comment Aury88, I agree with you, I think it might be necessary to have a tiler server or a separate frontend for end users, about vandalism I think solutions like Facebook’s Daylight Map has done a great job in giving stability to the data but the lack of a unique identifier for POIs affects the data a lot plus it would be hard to build an evaluation system without it