Sergey Golubev: “The best thing you can do for OSM now is to take a dump of the planet, delete it and start all over again”
Sergey Golubev is a naturalist, an experienced osmer and the author of the “City of Mine” blog. His thoughts about OpenStreetMap are always unexpected, and therefore doubly interesting. He knows how to find an unusual perspective and look at the usual from a different angle. Why OSM needs catastrophes, why it doesn’t have a community, and what the future holds for this project - he told about all this in an interview.
- How did you find out about the existence of OpenStreetMap?
- In 2007 or 2009 for work, I needed the most primitive substrate for the map, which would have the main cities, river lines, etc. At that time there was no QGIS, everything was done in ArcView GIS 3.2a. It now does not arise questions, whence in which case to take the data, so as not to trace them, and then it was a whole task. So I met OSM. In addition to working necessity, I, of course, had a purely human curiosity. When a new project appears, it is always interesting to study it, to see how it works.
')
- What was the OSM community then? How did you flow into it?
- Honestly, I still have not joined it, because the OSM community is an imaginary phenomenon, it really does not exist. All that is called the “OSM Community” is only 20-30 people who are active in Telegram and on the forum. This is an insignificant number of the number of people who really keep. After all, most simply open JOSM or iD and do not participate in any discussions. But with the 30th that are open to communication, I met in 2014, after registering with OSM. By the way, by that time there was already a large active community at GIS-Lab interested in cartography and all that is connected with it. At GIS-Lab, I have been active from 2008-2009.
- Why is the OSM community in Russia so few?
- First, Russia is a big country, because it is difficult to unite people among themselves, to make it possible for them to meet regularly. Moreover, such distances leave an imprint and on character - communication is more difficult for our people. Secondly, you need to understand that OSM is a technical project, and it is quite complicated, and therefore there is no point in meeting and discussing how someone has built houses today. This is clear to everyone. What will people who are even slightly interested in GIS technology discuss? Features of servers, Overpass code, kartostili and other troubles. But this does not bother most.
All this leads to the fact that the OSM community in Russia is very small. And I do not think that it will increase if the project will exist as it exists now.
- How is the project now? What's wrong with him?
- OSM is now in stagnation. This is a period of imaginary flourishing, but in reality - a deep crisis, which in the future will only worsen. And I think that with the departure of Steve Costa, the project has become more technological, but less alive. And I'm talking about the project as a whole, because what is happening in the Russian segment is only a reflection of the global dynamics.
- Why is OSM in crisis?
“Steve Coast, as the founder, is crazy.” For a long time he was the informal leader of OSM, rather even the leader. When a project is run by such a madman, it rarely becomes commercially successful. Moreover, it is unlikely to ever be high-tech. But such a project will always be alive, will constantly change depending on the whim and imagination of its leader.
With the departure of Steve Costa OSM numb. Now almost impossible any changes. Accordingly, many of the mistakes and problems of youth, to which no one previously paid attention, cannot be solved without serious consequences. Radical measures are needed, but how to implement them if everything has become so interdependent?
The simplest example, at one time, Steve Kost suggested not rendering those buildings that are not numbered. Thus, he wanted to solve this problem - the lack of numbering. But from the business point of view, such a cardinal decision is a disaster, because then most of the buildings would have disappeared from the map. But that was the idea of Steve Costa. He hoped that, on the contrary, this abrupt move would encourage people to “return” the buildings to the map, that is, to number them. In the future, the project would only benefit from such a decision.
At the moment, such "shares" are completely impossible. If you now propose to do something like this, then almost everyone will say that it will kill the project. But the fact of the matter is that the project from time to time needs small controlled disasters.
The best thing that can be done for OSM now is to take a dump of the planet, delete it and start all over again. Because re-drawing the entire planet is not difficult, but to correct what was originally done anyhow - it is very difficult. Of course, of course, we need to make more perfect mapping rules, which would at least somehow indicate the ultimate goal - what we want to get in the end.
I understand that this will not happen. But I believe that the OSM project itself will exist in the form in which it exists now, until some conditional OSM-2 appears - fork, where the problem with micro-mapping, with three-dimensional, with numbering and other things. If this happens, then this fork, which is likely to be commercial, will simply drag all the measures.
- How is the OSM useful?
- From OSM edits there is one indisputable advantage - this is a form of relaxation for the person who makes these edits. With all the rest you can argue. I believe that even now houses can be drawn by some kind of automaton, for example, a neural network. Therefore, those man-hours that are spent on the introduction of elementary lines, I can not explain anything else but a special form of recreation and entertainment of the Osmer itself. There are some good precedents - the OSM humanitarian team. This is an incredibly beautiful and useful project. Good projects to create maps were in Russia. Several times cards were created on the requests of various organizations: ambulance and others. But in general, such phenomena are a by-product rather than the main one. In general, public good is a populist term that is commonly used when nothing else can be given as an example.
- Who then is useful OSM?
- On OSM for a long time earn. Of the major projects - Mapbox, Maps.Me, NextGIS. Even 2GIS, which states that it plans to abandon OSM, currently uses its data. Not to say that there are many such companies, but considering the narrowness of the segment, it is quite acceptable for yourself. But it is worth paying tribute that some companies that earn OSM, then spend part of their income on the development of its infrastructure. Well, and those who like mapit - they enjoy and contribute data to the project. It turns out here is a symbiosis.
- What, then, does OSM differ from Narodnaya Yandex.Maps significantly?
- I think they have a lot in common. Moreover, NK is many times cooler than OSM, especially with regard to working with users.
If you ask this question to the very “OSM Community”, which consists of 20-30 technically savvy people, they will start talking about open data, open infrastructure, etc. This is all great, of course, but to those people who directly make changes. they do not care what this data is: open or closed. When you open the Yandex editor, where everything is simple and convenient, and you open the OSM editor, the same JOSM is insane for an untrained person, you see all the differences immediately and without further ado.
To paraphrase one well-known anecdote, the People’s Yandex.Map is just about the following story: you come to the airport, sit in a huge beautiful plane, incredibly beautiful flight attendants serve drinks, soft chairs - everything is fine, but for as long as you are silent, because if you zaikneshsya that the plane flies the wrong way, then the ambals will immediately come and throw you out of the plane. OSM is a corncob on which you can fly anywhere, but you need to bring your own aircraft parts to the take-off area and be able to install them. That makes all the difference.
- Do you plan to go to NYK?
- Not. For the sake of interest, I tried, circled around several buildings, but this is solely for the purpose of seeing how NK differs from OSM, how everything is arranged there. But I would not have passed, because the edits that I make, I make for myself and I plan to use them in the future: download, process, etc. I don’t have any sense to enter this data into Yandex, because I won’t be able to use it there .
OSM is good-looking, but the project clearly stagnated, it all came down to the accumulation of data in the database. OSM started as an independent project, like Wikipedia, but only in maps, and now OSM is a project where people want them to be cooler than Google or Yandex. If a person tries to do better than someone, he will never succeed. There are millions of ways for alternative development, but for some reason they always look and want to do, to repeat the success that has already been achieved by someone.
- What are the alternative ways of development? Except how to remove everything and re-start.
- All to remove and re-start - a radical path. You need to go in the direction where neither Yandex nor Google are able to provide services. I’ve been saying for a long time that the cards are used by completely different people, including those who have vision problems. So why not create several different cartestos, for example, for color blind or visually impaired? Why not create a separate map of water areas? Where there would be rivers, lakes and everything else. Almost on all known electronic maps it is impossible to find a single river until you scale up. God bless him, with an increase in zoom. There is no direction of flow! It would seem the easiest! And this is in Russia, where a huge number of people travel by water.
- So where exactly should OSM go?
- No need to be better than Google or Yandex, you have to be different. Then everything will be fine. Now OSM from an independent project that had all the chances to develop perfectly has turned into a database that is used by commercial companies. And all this awaits the moment when this abscess breaks through or resolves on other projects, or completely flushes into another format.
- If the problems of OSM growth are clear, why does nobody solve them?
- In OSM there is no single totalitarian dictator. This is his strength and weakness at the same time. On the one hand, this does not allow the project to slip into trash through the fault of a mad manager, on the other hand, it is impossible to realize a little bit of innovation, except for purely technical ones: adding new servers or increasing the capacity of existing ones. Sometimes they still change the style. These options are still possible, but something radical - immediately stalls at the root.
Partly, the democratic approaches adopted by the OSM are to blame for this - everything needs to be discussed and coordinated with everyone. While there are endless discussions, they continue to accumulate problems and slow down the development of the project. Why endless:? There will always be someone who will say that he does not agree and against.
Democracy is a good thing, but sometimes it must be interspersed with periods of hard totalitarian regimes, when a person comes in and says that he doesn’t care who he or she thinks will be like he said.
- Are you talking about the traditions of the domestic OSM or foreign community?
- Foreign colleagues have much more than what revolves around OSM. I like things that are not related to technology and hardware. These are questions of different kartovstrech, humanitarian team.
- Why such a difference?
- We in OSM are people of technical mentality, and there - located to the humanities. I think you know how our techies traditionally belong to the humanities? Therefore, we probably have more interest in pure IT. What is pop, so is the parish.
- I can assume that people of humanitarian mentality are more concerned with philosophical questions: public welfare, development of society, open data, equal opportunities, etc.
- I do not really believe in all these good aspirations, I also do not believe that a person, because of the conviction that he has the opportunity to make the world better, will do something. He can say to himself that he does it because he wants to change the world, but in fact he does it because he likes it, because he satisfies some kind of his own complexes, problems, or gets his own selfish pleasure.
Often, people simply deceive themselves and talk about what is not. I do not believe that OSM will save the world. Well that he is, but do not overestimate him. In addition, OSM actually made all other open cartographic senseless. The negative role of OSM is usually hushed up.
- Tell us about this role, which is silent. What are the myths about OSM?
- First about the myths. The biggest myth is about the big fullness of the base. Now OSM - these are the balls of cities that are connected to each other by threads-roads. All other settlements between them are marked by dots or two or three streets and that's it. There is nothing else. OSM is not a complete database. She still has to grow and grow. The data in OSM is extremely broken and, let's say, “dirty”. If you are going to do something with them, they should be pre-processed and cleaned.
How did OSM hurt? Not even OSM itself, but the infrastructure around it: the ease of downloading data, QGIS and Overpass - thanks to this, programmers unexpectedly came to cartography. Cartographers left and visualizers arrived. Now it is the cards, in the former understanding of the word, no. There are visualizations of various data sets. The notion of generalization is completely forgotten. Now she has become what Mapbox is doing, when not all points from polygons are loaded on different scales and zooms, which is why maps become ugly. From 16 to 18, zoom companies still strive to make the map beautiful, and everything else is monstrous. I note that such cards appear not because the market asks for such, but because few people are familiar with good examples.
“What good examples are you talking about?”
- You can take any card until 1990 and look at the quality of its performance. These are primarily signatures and fonts.
- How would you give advice to a newcomer in OSM?
- Do not map with Google and Yandex, because it is not a thrill. OSM is brilliant in its principles, I really like them, both in the project and in life: firstly, do not steal data, secondly, do some garbage and harm the project, thirdly, enjoy it. This is the absolute minimum, which is sufficient with excess. If you feel that you can follow these rules, then act. If you do not get pleasure, then you can always say boldly: "Sorry guys, I can not participate." People often ask me why I began to map some territory and did not finish it according to the OSM rules. I answer that I stopped enjoying the drawing of this particular area, as I am in a mood, I will continue, and therefore, sorry, this is the rule.
- How can you attract people with a technical mindset to RU-OSM?
- In chat and on the forum? I do not know. It seems to me that people come there when they have a specific OSM question. As for the appearance of new Osmer in the project, for this it is necessary that as many people as possible know about the existence of OSM. Let's be honest, people do not know about him. And all the talk about the fact that OSM is so popular and everyone knows about it for a long time, ends with the Moscow ring road.
The fact that Yandex or Google have maps, the user finds out almost immediately, as soon as he takes advantage of these search engines. How should he know about the existence of OSM and its benefits?
- What do you say at the end of the conversation?
- Viva la revolucion! If you do not want revolutionary changes, then you need to do at least something that nobody else does. Map houses is already a move. It is necessary to pick up something that is not found anywhere else on other maps. Then it will be cool. Or walk along the street and zamapit trees indicating the breed - that's cool. This will definitely not be on any map, and if you walk down the street and zapapit house numbers - for this there is Yandex or 2GIS.
- How did you find out about the existence of OpenStreetMap?
- In 2007 or 2009 for work, I needed the most primitive substrate for the map, which would have the main cities, river lines, etc. At that time there was no QGIS, everything was done in ArcView GIS 3.2a. It now does not arise questions, whence in which case to take the data, so as not to trace them, and then it was a whole task. So I met OSM. In addition to working necessity, I, of course, had a purely human curiosity. When a new project appears, it is always interesting to study it, to see how it works.
')
- What was the OSM community then? How did you flow into it?
- Honestly, I still have not joined it, because the OSM community is an imaginary phenomenon, it really does not exist. All that is called the “OSM Community” is only 20-30 people who are active in Telegram and on the forum. This is an insignificant number of the number of people who really keep. After all, most simply open JOSM or iD and do not participate in any discussions. But with the 30th that are open to communication, I met in 2014, after registering with OSM. By the way, by that time there was already a large active community at GIS-Lab interested in cartography and all that is connected with it. At GIS-Lab, I have been active from 2008-2009.
- Why is the OSM community in Russia so few?
- First, Russia is a big country, because it is difficult to unite people among themselves, to make it possible for them to meet regularly. Moreover, such distances leave an imprint and on character - communication is more difficult for our people. Secondly, you need to understand that OSM is a technical project, and it is quite complicated, and therefore there is no point in meeting and discussing how someone has built houses today. This is clear to everyone. What will people who are even slightly interested in GIS technology discuss? Features of servers, Overpass code, kartostili and other troubles. But this does not bother most.
All this leads to the fact that the OSM community in Russia is very small. And I do not think that it will increase if the project will exist as it exists now.
- How is the project now? What's wrong with him?
- OSM is now in stagnation. This is a period of imaginary flourishing, but in reality - a deep crisis, which in the future will only worsen. And I think that with the departure of Steve Costa, the project has become more technological, but less alive. And I'm talking about the project as a whole, because what is happening in the Russian segment is only a reflection of the global dynamics.
- Why is OSM in crisis?
“Steve Coast, as the founder, is crazy.” For a long time he was the informal leader of OSM, rather even the leader. When a project is run by such a madman, it rarely becomes commercially successful. Moreover, it is unlikely to ever be high-tech. But such a project will always be alive, will constantly change depending on the whim and imagination of its leader.
With the departure of Steve Costa OSM numb. Now almost impossible any changes. Accordingly, many of the mistakes and problems of youth, to which no one previously paid attention, cannot be solved without serious consequences. Radical measures are needed, but how to implement them if everything has become so interdependent?
The simplest example, at one time, Steve Kost suggested not rendering those buildings that are not numbered. Thus, he wanted to solve this problem - the lack of numbering. But from the business point of view, such a cardinal decision is a disaster, because then most of the buildings would have disappeared from the map. But that was the idea of Steve Costa. He hoped that, on the contrary, this abrupt move would encourage people to “return” the buildings to the map, that is, to number them. In the future, the project would only benefit from such a decision.
At the moment, such "shares" are completely impossible. If you now propose to do something like this, then almost everyone will say that it will kill the project. But the fact of the matter is that the project from time to time needs small controlled disasters.
The best thing that can be done for OSM now is to take a dump of the planet, delete it and start all over again. Because re-drawing the entire planet is not difficult, but to correct what was originally done anyhow - it is very difficult. Of course, of course, we need to make more perfect mapping rules, which would at least somehow indicate the ultimate goal - what we want to get in the end.
I understand that this will not happen. But I believe that the OSM project itself will exist in the form in which it exists now, until some conditional OSM-2 appears - fork, where the problem with micro-mapping, with three-dimensional, with numbering and other things. If this happens, then this fork, which is likely to be commercial, will simply drag all the measures.
- How is the OSM useful?
- From OSM edits there is one indisputable advantage - this is a form of relaxation for the person who makes these edits. With all the rest you can argue. I believe that even now houses can be drawn by some kind of automaton, for example, a neural network. Therefore, those man-hours that are spent on the introduction of elementary lines, I can not explain anything else but a special form of recreation and entertainment of the Osmer itself. There are some good precedents - the OSM humanitarian team. This is an incredibly beautiful and useful project. Good projects to create maps were in Russia. Several times cards were created on the requests of various organizations: ambulance and others. But in general, such phenomena are a by-product rather than the main one. In general, public good is a populist term that is commonly used when nothing else can be given as an example.
- Who then is useful OSM?
- On OSM for a long time earn. Of the major projects - Mapbox, Maps.Me, NextGIS. Even 2GIS, which states that it plans to abandon OSM, currently uses its data. Not to say that there are many such companies, but considering the narrowness of the segment, it is quite acceptable for yourself. But it is worth paying tribute that some companies that earn OSM, then spend part of their income on the development of its infrastructure. Well, and those who like mapit - they enjoy and contribute data to the project. It turns out here is a symbiosis.
- What, then, does OSM differ from Narodnaya Yandex.Maps significantly?
- I think they have a lot in common. Moreover, NK is many times cooler than OSM, especially with regard to working with users.
If you ask this question to the very “OSM Community”, which consists of 20-30 technically savvy people, they will start talking about open data, open infrastructure, etc. This is all great, of course, but to those people who directly make changes. they do not care what this data is: open or closed. When you open the Yandex editor, where everything is simple and convenient, and you open the OSM editor, the same JOSM is insane for an untrained person, you see all the differences immediately and without further ado.
To paraphrase one well-known anecdote, the People’s Yandex.Map is just about the following story: you come to the airport, sit in a huge beautiful plane, incredibly beautiful flight attendants serve drinks, soft chairs - everything is fine, but for as long as you are silent, because if you zaikneshsya that the plane flies the wrong way, then the ambals will immediately come and throw you out of the plane. OSM is a corncob on which you can fly anywhere, but you need to bring your own aircraft parts to the take-off area and be able to install them. That makes all the difference.
- Do you plan to go to NYK?
- Not. For the sake of interest, I tried, circled around several buildings, but this is solely for the purpose of seeing how NK differs from OSM, how everything is arranged there. But I would not have passed, because the edits that I make, I make for myself and I plan to use them in the future: download, process, etc. I don’t have any sense to enter this data into Yandex, because I won’t be able to use it there .
OSM is good-looking, but the project clearly stagnated, it all came down to the accumulation of data in the database. OSM started as an independent project, like Wikipedia, but only in maps, and now OSM is a project where people want them to be cooler than Google or Yandex. If a person tries to do better than someone, he will never succeed. There are millions of ways for alternative development, but for some reason they always look and want to do, to repeat the success that has already been achieved by someone.
- What are the alternative ways of development? Except how to remove everything and re-start.
- All to remove and re-start - a radical path. You need to go in the direction where neither Yandex nor Google are able to provide services. I’ve been saying for a long time that the cards are used by completely different people, including those who have vision problems. So why not create several different cartestos, for example, for color blind or visually impaired? Why not create a separate map of water areas? Where there would be rivers, lakes and everything else. Almost on all known electronic maps it is impossible to find a single river until you scale up. God bless him, with an increase in zoom. There is no direction of flow! It would seem the easiest! And this is in Russia, where a huge number of people travel by water.
- So where exactly should OSM go?
- No need to be better than Google or Yandex, you have to be different. Then everything will be fine. Now OSM from an independent project that had all the chances to develop perfectly has turned into a database that is used by commercial companies. And all this awaits the moment when this abscess breaks through or resolves on other projects, or completely flushes into another format.
- If the problems of OSM growth are clear, why does nobody solve them?
- In OSM there is no single totalitarian dictator. This is his strength and weakness at the same time. On the one hand, this does not allow the project to slip into trash through the fault of a mad manager, on the other hand, it is impossible to realize a little bit of innovation, except for purely technical ones: adding new servers or increasing the capacity of existing ones. Sometimes they still change the style. These options are still possible, but something radical - immediately stalls at the root.
Partly, the democratic approaches adopted by the OSM are to blame for this - everything needs to be discussed and coordinated with everyone. While there are endless discussions, they continue to accumulate problems and slow down the development of the project. Why endless:? There will always be someone who will say that he does not agree and against.
Democracy is a good thing, but sometimes it must be interspersed with periods of hard totalitarian regimes, when a person comes in and says that he doesn’t care who he or she thinks will be like he said.
- Are you talking about the traditions of the domestic OSM or foreign community?
- Foreign colleagues have much more than what revolves around OSM. I like things that are not related to technology and hardware. These are questions of different kartovstrech, humanitarian team.
- Why such a difference?
- We in OSM are people of technical mentality, and there - located to the humanities. I think you know how our techies traditionally belong to the humanities? Therefore, we probably have more interest in pure IT. What is pop, so is the parish.
- I can assume that people of humanitarian mentality are more concerned with philosophical questions: public welfare, development of society, open data, equal opportunities, etc.
- I do not really believe in all these good aspirations, I also do not believe that a person, because of the conviction that he has the opportunity to make the world better, will do something. He can say to himself that he does it because he wants to change the world, but in fact he does it because he likes it, because he satisfies some kind of his own complexes, problems, or gets his own selfish pleasure.
Often, people simply deceive themselves and talk about what is not. I do not believe that OSM will save the world. Well that he is, but do not overestimate him. In addition, OSM actually made all other open cartographic senseless. The negative role of OSM is usually hushed up.
- Tell us about this role, which is silent. What are the myths about OSM?
- First about the myths. The biggest myth is about the big fullness of the base. Now OSM - these are the balls of cities that are connected to each other by threads-roads. All other settlements between them are marked by dots or two or three streets and that's it. There is nothing else. OSM is not a complete database. She still has to grow and grow. The data in OSM is extremely broken and, let's say, “dirty”. If you are going to do something with them, they should be pre-processed and cleaned.
How did OSM hurt? Not even OSM itself, but the infrastructure around it: the ease of downloading data, QGIS and Overpass - thanks to this, programmers unexpectedly came to cartography. Cartographers left and visualizers arrived. Now it is the cards, in the former understanding of the word, no. There are visualizations of various data sets. The notion of generalization is completely forgotten. Now she has become what Mapbox is doing, when not all points from polygons are loaded on different scales and zooms, which is why maps become ugly. From 16 to 18, zoom companies still strive to make the map beautiful, and everything else is monstrous. I note that such cards appear not because the market asks for such, but because few people are familiar with good examples.
“What good examples are you talking about?”
- You can take any card until 1990 and look at the quality of its performance. These are primarily signatures and fonts.
- How would you give advice to a newcomer in OSM?
- Do not map with Google and Yandex, because it is not a thrill. OSM is brilliant in its principles, I really like them, both in the project and in life: firstly, do not steal data, secondly, do some garbage and harm the project, thirdly, enjoy it. This is the absolute minimum, which is sufficient with excess. If you feel that you can follow these rules, then act. If you do not get pleasure, then you can always say boldly: "Sorry guys, I can not participate." People often ask me why I began to map some territory and did not finish it according to the OSM rules. I answer that I stopped enjoying the drawing of this particular area, as I am in a mood, I will continue, and therefore, sorry, this is the rule.
- How can you attract people with a technical mindset to RU-OSM?
- In chat and on the forum? I do not know. It seems to me that people come there when they have a specific OSM question. As for the appearance of new Osmer in the project, for this it is necessary that as many people as possible know about the existence of OSM. Let's be honest, people do not know about him. And all the talk about the fact that OSM is so popular and everyone knows about it for a long time, ends with the Moscow ring road.
The fact that Yandex or Google have maps, the user finds out almost immediately, as soon as he takes advantage of these search engines. How should he know about the existence of OSM and its benefits?
- What do you say at the end of the conversation?
- Viva la revolucion! If you do not want revolutionary changes, then you need to do at least something that nobody else does. Map houses is already a move. It is necessary to pick up something that is not found anywhere else on other maps. Then it will be cool. Or walk along the street and zamapit trees indicating the breed - that's cool. This will definitely not be on any map, and if you walk down the street and zapapit house numbers - for this there is Yandex or 2GIS.
Source: https://habr.com/ru/post/447850/
All Articles