Slashdot Mirror
What OpenStreetMap Can Be (systemed.net)
An anonymous reader shares a blog post on OpenSourceMap: Most OSM commentary focuses on unimportant minutiae (layers, for goodness' sake, as if it's still 2004) without seeking to examine what makes OSM unique -- and whether that's still relevant in a rapidly changing market. Could OSM become a dead-end curio while Google, Apple, and an increasingly self-sufficient Mapbox hare off in another, common direction? OSM's continuing differentiation from Google/Apple boils down to two points.
First, a non-commercial imperative. Google and Apple (and Mapbox, TomTom, HERE) are beholden to their shareholders and investors. They do what makes them money, which means car navigation. (Once human-controlled, now, increasingly, self-guided. When people ask "How far ahead of Apple is Google Maps?", what they usually mean is "Who will get to self-driving cars first?") OSM, however, isn't ruled by shareholder value, but by the preoccupations of its contributor base. (We'll come onto that demographic later.) Whether that's a good thing depends on what you want from a map. But it's clearly a point of differentation.
Second, ground truthed local knowledge. Surveying by locals is the gold standard of OSM, building a rich, intricate compilation of contributors' preoccupations. The painstaking human curation of areas and topics remains unique to OSM. Neither of these are under threat from Google/Apple. Outsourced quick-fire digitisation of Street View-type imagery in cheap labour countries doesn't give you this. Nor does image recognition. OSM's points of differentation remain clear. In OSM's early days, commentators used the phrase "democratising mapmaking," and it remains true. You choose what to map; and you choose how to use the map. You participate. Other maps are a one-way street: sure, you can contribute (actively through map corrections, or passively through using a mobile app that phones home), but the provider chooses what you get back.
First, a non-commercial imperative. Google and Apple (and Mapbox, TomTom, HERE) are beholden to their shareholders and investors. They do what makes them money, which means car navigation. (Once human-controlled, now, increasingly, self-guided. When people ask "How far ahead of Apple is Google Maps?", what they usually mean is "Who will get to self-driving cars first?") OSM, however, isn't ruled by shareholder value, but by the preoccupations of its contributor base. (We'll come onto that demographic later.) Whether that's a good thing depends on what you want from a map. But it's clearly a point of differentation.
Second, ground truthed local knowledge. Surveying by locals is the gold standard of OSM, building a rich, intricate compilation of contributors' preoccupations. The painstaking human curation of areas and topics remains unique to OSM. Neither of these are under threat from Google/Apple. Outsourced quick-fire digitisation of Street View-type imagery in cheap labour countries doesn't give you this. Nor does image recognition. OSM's points of differentation remain clear. In OSM's early days, commentators used the phrase "democratising mapmaking," and it remains true. You choose what to map; and you choose how to use the map. You participate. Other maps are a one-way street: sure, you can contribute (actively through map corrections, or passively through using a mobile app that phones home), but the provider chooses what you get back.
OpenStreetMap is pretty damn good. For example, the OsmAnd android app allows you to browse in excellent detail any offline downloaded map. In this sense it's far superior to Google maps. It can even calculate offline directions.
The problem is OpenStreetMap is not easy for editing or browsing on a computer. Try finding a GUI. There are two dozen and they all suck. OpenStreetMap needs a professional editor/viewer for all platforms and it could be vastly more popular.
"What lies behind us, and what lies before us are tiny matters compared to what lies within us." Ralph Waldo Emerson
Tyrone might need a map to find Trump's phat booty in the shower!!
What we don't want is Google using this data, who's primary purpose is to spy and collect whatever it can on people.
Yessir!!!
- creimer
OpenStreetMap is pretty damn good. For example, the OsmAnd android app allows you to browse in excellent detail any offline downloaded map. In this sense it's far superior to Google maps. It can even calculate offline directions.
The problem is OpenStreetMap is not easy for editing or browsing on a computer. Try finding a GUI. There are two dozen and they all suck. OpenStreetMap needs a professional editor/viewer for all platforms and it could be vastly more popular.
One problem with Google maps is the (lack of) contrast.
On Google, choose any town or city and Zoom down looking for streets. The streets are juuuuust sligtly darker than the background, and there's no variation in the background. It gives the impression of a map of highways with lots and lots of empty space in between.
This was driven home to me recently when I wanted to find out where a road went in my area... and couldn't. It's impossible to trace the road with your eyes at a reasonable zoom level, and at the level where the road has labels you're too close to get an idea of the road relative to anything else.
Choose any town or city and Zoom down looking for streets. Roads in OSM are perfectly readable at reasonable resolutions, and you can even see the difference between different types of areas.
I don't know how "this makes money" on Google, but maybe somehow it does.
Coward, Here used to be Navteq.
Garmin ships some maps based on OpenStreetMap data. Mostly outdoor maps in Europe, where OSM is truly outstanding, having much more data about hiking and biking than other providers.
When you're planning a route using Google Maps you can drag the route around to set waypoints and try to find the best route that works for you. I often use it to find a good speed vs simplicity balance or to avoid roadworks as there is no way to tell it to avoid a specific road.
If OSM made it easy to set waypoints and block roads, I'd use it all the time because its maps are much more readable and easier on the eyes.
This article attempts to summarize what is wrong with OSM and why it has not caught on. It stirred a community debate when it was written: https://blog.emacsen.net/blog/2018/02/16/osm-is-in-trouble/
I like OSM, but just today I tried printing directions off of it and couldn't figure it out for the life of me, even with RTFW. So a simple 'Print' button for maps and directions would do wonders.
The main reason I only use OSM and the OSMAND app, is that I don't want my map searches and travelpatterns to become available to third parties.
None of the competitors can beat OSM on this. Not in functionality (offline maps, search, and navigation), but also not in trust (OSM is a non profit, so there is less incentive to sell or derive data).
the main thing that I don't like on OSM is how the zoom works... it would be an easy fix but I guess most people like it like that?
Google built their own mapping dataset and I believe almost their entire dataset is now their own, basically anywhere that is street view is their data.
Apple is doing the same but it will be even better as itâ(TM)s doing 3D scans of the surrounding area https://techcrunch.com/2018/06/29/apple-is-rebuilding-maps-from-the-ground-up/
...to travel from Maine to California, you are fucked.
All you need to do is look at the overhead signs on expressways. They will point to to major cities. You don't need to know navigation on a micro level. You don't even need to know the names of the highways. Just city names.
Google Maps uses artificial intelligence techniques with its gathered data so it needs less manual labor. OSM depends exclusively on manual labor. The bad is that if contributors aren't interested it's less complete than Google Maps (many places lack street numbers or aren't mapped). The good is that with people's interest the maps can be more refined than Google's.
What I completely disagree about is the unimportant minutiae. It's these things the ones that separate great products from the rest.
By now, Google probably has data that they have made themselves from satellite images and their Street view fleet, but at the start they licensed the map data from others, including HERE (under their previous name Navteq).
And that's good. Because it's a fucking map. Some things shouldn't be drum circles.
Maps are about getting somewhere, going places and getting back again. Simple.
A walk in the woods, drop into a single track to nowhere or two track over the horizon; you go because its there. 1x1,2x2 and 4x4 all scale the same paths. Jeep made a brand out of getting you there. Famously, Jeep emerged out of WWII and into everyman's driveway. Jeep and 4x4 didn't takeoff for decades. But one thing changed everything - automatic scaleability. Jeep replaced manual shift transmissions with automatic shift. What happened next led to SUV's and changed the industry to the point that FORD simply dropped its car lineup keeping SUV's.
The automatic transmission meant that the ' lady' of the household could drive the Jeep in the driveway. That made getting her signature at the bottom of the car loan easier. That led to leather seats, cup holders, butt warmers and the SUV's today. That's scaleability. A simple automatic shifting transmission allowed everyone to enjoy going places - off road even.
SO build the UX element that defines Open Map's automatic transmission where anyone who wants, can map their trail, single track or off road adventure where no map has gone before.
A primary benefit to such a platform is that it can field and apply corrections and updates to the street graph. They collect this data, but NEVER process it. Source: years of observation of my correction not being incorporated.