Slashdot Mirror
The Future of Free and Open-Source Maps (emacsen.net)
Grady Martin writes: Former OpenStreetMap contributor and Google Summer of Code mentor Serge Wroclawski has outlined why OpenStreetMap is in serious trouble, citing unclear usage policies, poor geocoding (address-to-coordinate conversion), and a lack of a review model as reasons for the project's decline in quality. Perhaps more interesting, however, are the problems purported to stem from OpenStreetMap's power structure. Wroclawski writes: "In the case of OpenStreetMap, there is a formal entity which owns the data, called the OpenStreetMap Foundation. But at the same time, the ultimate choices for the website, the geographic database and the infrastructure are not under the direct control of the Foundation, but instead rest largely on one individual, who (while personally friendly) ranges from skeptical to openly hostile to change."
Eh, quality is not declining.
They could have a chat with the what3words folks...
https://map.what3words.com/kicks.pasta.steer
You have the right to remain sentient. If you give up the right to remain sentient, you will be elected to public office
This is sorta peripheral, butt I have been having fits with my Garmin updates. The maps are definitely not up to date. the last trip I took, the speed limits were wrong more than they were right. With the insulting requirement to buy the same map over and over for the devices I own, and their arrogance, I am considering ditching all my Garmin devices. My better and I spent three months in Chile and we tried using OpenStreet maps. They were unusable. We bought the Garmin S. America set and it was marginal at best. We used Google maps and Waze when we had cell service and the Google maps are much much better. And free... as long as you have cellular service. Question: How is it that Google maps are head and shoulders better than Garmin maps and Garmin charges out the wazoo???
Dear Microlimp: I give you 2 valid product keys for win7 and you reject both of them. Piss off you wankers!!!
They can't take the fork in the road?
Table-ized A.I.
Whatever... It's futile anyway. I'm merely battling windmills.
? Seems to go against the whole 'open' thing, or perhaps what we really need is FreeAndOpenStreetMaps
Open isn't as open doesn't.
But really. What's to stop someone from forking the whole shebang and then running the fork properly? The nuclear option.
As opposed to merely being possessed or controlled?
If you can fork it - without legal repercussions - it can be free. It's just a matter of effort and organization. How badly do we want freedom? How capable of we of protecting said freedom?
Can we just take our ball and go home?
Tooling is. It shouldn't take 2 weeks to import the DB. A 65GB DB shouldn't turn into a 2TB one once restored. Serving tiles shouldn't require a convoluted stack of 20 apps which ultimately take up gigs of RAM and 20% cpu just to sit idle.
Just for starters...
"instead rest largely on one individual, who (while personally friendly) ranges from skeptical to openly hostile to change."
Lennart Poettering finds another good thing to screw up.
Reading his rant I could identify with many of the problems. But the solution lies with the foundation itself and the board of directors. The board of directors needs to be ran by the president in a professional, transparent, ethical manner, that follows its rules especially as it might relate to conflicts of interest. I suggest finding a way to invoke parliamentary procedure to the board even if they need to hire a professional parliamentarian for meetings. Rewrite the bylaws as necessary to correct the deficiencies.
So you need a sympathetic president, and enough like minded members to constitute a majority. Then you start voting. Get the new bylaws approved. Then the board needs to gain control of the entire project head to toe. Find a way to raise *some money* to hire an attorney. You may need to step on some toes. Get rid of the frenemy gatekeepers at that time.
Then it becomes about money. Wikipedia manages to fundraise. Wikipedia sucked forever until it got a critical mass to start hiring programmers, and this project might need to follow the same path. Until you can hire on programmers, look for free help from college programming departments and others who could glean something for doing so. I'm not sure what would be best route for fund raising, but as an idea maybe the Pokemon guys might be interested in buying custom mapping or data (just as an example). Eventually you will formally need to hire a programmer. You need to chop up the project in smaller individual chunks. You need to create an easy to use standard for data transfer to make the data you do have easy to use and the different chunks of the project to interact with and standardize around. The interface needs to be dead drop easy to use with no command line knowledge necessary. Over time, you will get more users who might be willing to donate that $3 to the project when asked. Again, you might want to look at Wikipedia's successes and its mistakes (hard to use editing interface) and learn from them.
Eventually you will succeed in overcoming the things mentioned in the rant. But the very beginning starts with cleaning up the board of directors first that run the foundation before the project itself can be cleaned up.
At least that's how I'd do it. I'd love to hear other's opinions.
Take the cheese to sickbay, the doctor should see it as soon as possible - B'Elanna Torres, "Learning Curve"
> "intellectual property" is a crime scheme to get money without working for it.
Based on what? Are you saying that if I do the work of launching satellites, taking photographs, catalogue and index them, and then build the earth-based infrastructure to load and display them to end users while simultaneously building an interface that a user can then put origination and destination points and get accurate directions to those points that all of that is not considered work? I don't deserve to be paid? Pray tell, where do you live? I'll be taking your possessions - you didn't work for them. You didn't earn them in my eyes. You fucking retarded motherfucker.
It has more information and less wrong information than Google Maps in my area, and it's OPEN. Don't be so quick with your judgments.
Garmin must have fired or lost some of their key technical employees, and/or decided to let marketing run things, because they have gone straight into the toilet. My first GPS, which I still own, is a Garmin GPS 12. This was one of the early 12-satellite GPSes and it has a serial port connector, which is why I bought it in the first place. It is great in every way for what it is; durable, usable (the UI was good for its day) and above all reliable.
I also own a Garmin Nuvi 1450LMT. It is a total POS. Touch recognition is complete garbage. But what's even worse than the unit itself is the management software. The software to load maps onto the device has gone through several revisions, a complete rewrite, and several more revisions. During that time I've tried to update my maps about ten times, and succeeded twice. One of those updates left the device unusable for literally years. It actually erased the maps from the device, then refused to load any new ones, saying it couldn't find it. They eventually updated the software and it found it, so I did a map update... which claimed to work, but didn't actually load anything but highways. I did zoom in, nothing. So I did another map update, and that one actually worked. The GPS works as well as it ever did now, which is not that well. GPS reception is good, everything else about the unit is bad.
But let us not forget Garmin Viago, their GPS software for Android. It came and went in about three months, it literally never worked (I tried to use it a dozen times, got it to find an address about twice, got it to actually route me there literally zero times) and I'm just out the money I spent on it now. They literally threw up their hands and walked away because making an Android GPS app was too hard for them.
In short, Garmin is totally incompetent today, and giving them money is a total noob move.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
The quality of the data is very much dependent on volunteers. In the Netherlands, the data is actually pretty good and detailed.
Certain details are in the map that HERE/TeleAtlas/Google/Waze don't have, like pedestrian tunnels, paths in woods, animal-crossings, etc. There are many people keeping things up-to-date.
But, if you're somewhere in a third world country, outside of a city, chances are higher that your road is missing or incorrect.
Geocoding requires two things:
1. Accurate and complete data. It is well acknowledged that in many areas, full street addresses are not yet there in Openstreetmap, but you can geocode at street-level. I've had many cases where Google's geolocation for addresses is way out, and doesn't even give a warning that it's only managed a partial match on the geographic centre of the district.
2. Context. Depending on your application, you may only be interested in the 'Station Road' nearest you. How's the geocoding server supposed to know if the client application doesn't give it a hint as to how the results should be ordered? Google's geocoder is likewise hopeless with giving relevant results if the client application doesn't give it a hint that you'd prefer a local result over a result from a different continent.
some issues as any map. Still it is the good map.
I use mostly the OpenStreetMap, Maps.ME on smartphone, where I can download the whole country, and I also use Google map.
But the guy who owns the IP didn't do anything. They just had money to pay salaries, and he didn't work on 99%of the Intellectual Property and now has billions. You can call us retarded if you like, but what you are is blind. IP is a way to make money off the labour of others. That scientist who worked in the lab don't get crap but another meal to do it again tomorrow.