Overturemaps.org - big-businesses OSMF alternative

Foundation
107

Some more interesting paragraphs and compliments from the transcript to the slide deck

Sounds like they’re impressed with the “visual” features that also enable them to enter Japan. But lack of standardization is a sticking point, e.g,. no date convention in tags?

And also the open source community has gone from strength to strength. The OpenStreetMap community is producing and maintaining a visually extremely attractive map, with a wealth of detail.
Now, our new Maps Platform is designed to bring all those sources together in a consistent way to power the most demanding applications.

But also until now, Japan was out of reach. It was simply a too expensive and too difficult market to enter. But here we are starting with a new map, with a very impressive base map already, from where we can improve using our other sources, such as trace data, sensors-derived observations, and more.

And it also lacks standardization, which is challenging. Each country runs its own community and there’s a bunch of differences, even for simple things like month-day. It can be flipped depending on the country, making it really difficult to work with on a global map. And of course, big companies trying to do a lot of work can’t do automation, which is problematic when you’re trying to leverage all the sensor data you’re getting

Automotive has been evaluating OSM but they’ve haven’t been able to really leverage it, again due to the challenges, but also because they’re more cautious with the quality, particularly in coverage and routing. Some have looked at it, but they haven’t gone very far with it. This is where TomTom’s Maps Platform comes in.

And this is where the disruption is happening. This is big. Now, those using OpenStreetMap can get the full coverage of TomTom’s road network. And that provides a significant opportunity for everyone to co-collaborate. It’s everything great about TomTom with the added richness of OSM. The TomTom Maps Platform makes OpenStreeMap enterprise-ready, commercial-grade. What I mean by this is that we’re going to add in all the extra features and capabilities that have blocked people from using OSM, like standardization of the content. TomTom’s Map Platform will read in OSM data and normalize it to a single standard globally. We’re working with a few of the top tech companies in the world on this standardization.

Now, we’ve talked about OpenStreetMap as a very important source. This is a human-curated source that is good at certain types of features. For instance, visual features. Their roadmap and the community is worldwide, meaning that this quality, they’re able to provide on a worldwide basis. Now, we as TomTom intend to be a very good corporate member of this community, also giving back

3 Likes
108

A possible take on what may be going on here:

U.S. law does not recognize collections of facts as copyrightable (the Feist doctrine). This is different from the EU, where copyright law affords a certain degree of protection for databases. So at least as the US is concerned, a large part of OSM data does not fall under copyright – the names of streets, speed limits, road surfaces, opening hours of shops, these are collections of facts.

Geometries are different, as it takes some level of creative choice when drawing lines and shapes. The fact that there’s a road above Pebble Beach which has two hairpin turns isn’t copyrightable, but mappers exercise some discretion over how many points to use to represent that road, and where exactly to put them – enough to meet the relatively low standard of copyrightability.

It looks like their plan may be to extract non-copyrightable facts from OSM, mix them with their own data, and offer the result as a proprietary product. The only parts they’ll acknowledge as being open data are those to which statutory copyright law applies.

At least in the US, much of the protection afforded by the ODbL rests solely on contract law. The big questions: To what degree is this license even enforceable? Who are the parties? What happens if someone downloads an OSM-PBF file and extracts data from it? There is no “click-wrap” where the user checks a box that says “By using this data, I consent to the terms of the ODbL.” Most mirrors don’t even have a disclaimer with the terms that apply to downloads. Effectively, a user can gain access to OSM data without ever hearing about the ODbL (Copyright still applies in this situation, but a good argument could be made that the user is not subject to the contractual terms of the ODbL, which were meant to cover the elements outside the scope of copyright).

Usual disclaimer: Not written by a lawyer, don’t treat this as legal advice.

3 Likes
109

And they would end up with a product that they wouldn’t be able to use in the EU (assuming the sui generis DB protection would actually stick).

IMHO reiterating the last dozen years of licence discussion is way off topic, particularly when it is trying to 2nd guess reasoning behind a short lived pre-Overture strategy of Tomtom that we will never have any conclusive evidence on.

3 Likes
110

Particulary version numbers.

4 Likes
111

Here you go:
https://6ug7hetxl9.execute-api.us-east-2.amazonaws.com/

Let me know if you need one that’s styled.

7 Likes
117

Please, can you stop personal duel here? That is now clearly offtopic in this thread.

It seems that this type of “what is wrong with OSM community taking your behaviour as an example” thread is doomed to fail and likely against at least some rules, but here it is also offtopic (see the thread title)

EDIT: this applies also to message below.

17 Likes
118

is blatantly false. I most certainly have something to say about it (and rightly so, and I’m not being dramatic by doing so) when someone blithely negates my actual experiences in this project: the gaslighting is obvious and unacceptable.

The counter-evidence is my own experience of dozens, even hundreds of positive, collaborative, instructive, results-oriented, look-at-what-we’ve-done, look-at-what-we’ll-do human relationships. These real, actual experiences are built out of good intentions, good work, trust and results. I, along with I’m sure, millions of others in OSM can say this, too.

These real, actual experiences are positive, willful, constructive, growing, adapt and strive for genuinely high quality all at the same time. Yes, I set a high bar for myself and then offer my contributions (data, code, wiki, collaboration, syntax, proposals…) as examples, good ones, I hope. I don’t look for accolades or recognition, I mostly map, document some things in wiki, talk to others where both chin-scratching and head-nodding take place (usually in that order) and I do believe, millimeter by kilometer at a time, our map gets (our map data get) better. Wanna build community? Try that.

People who think they can dox where someone is a former employee and at the same time challenge its truth speaks for itself. This is all bait being dragged behind an ugly fishing vessel, everybody, although I’m sure many, even most already see that.

Community is much more than pub-based mapping parties or whatever a (stunted?) limited-experience or can’t-seem-to-spread-wings-and-fly experience of community might be. Community is relationships (like we find in this forum) that are built over months and years, decades in some cases. Built out of constructive collaboration, yes, criticism at times, but constructive criticism, built out of making something, rather than tearing something down, throwing rocks at it, disparaging it as if everybody does that: everybody does not do that. Stop saying so, I insist. We insist. Build community in positive ways, I’m all for that, but the rock-throwing (seeming like it is flying under radar, or masquerading as “but I’m BUILDING community by saying what is wrong with it!”) must stop. Some people simply cannot learn what constructive criticism is, and if so, in this project, they have no business criticizing.

Sure, we can, do and should “better build” our community / communities in OSM. But (I’ll say it again) I will not stand idly by while someone makes blanket-statement, wholesale falsehoods about it. Having trouble building community? If so, do not assume your problems exist in the wider community. I’m all for identifying real problems we have or obstacles that prevent someone from building community. But, I myself (and literally dozens of people I know in this project, along with thousands and millions of others) don’t seem to have that trouble. Listening to you, maybe that means OSM has big problems. But for “many of us,” not really. Actually, I repeat myself here (and again), something I find myself doing a great deal, but only with a single person in this entire project (out of millions). This is tiring.

Others do not say “essentially the exact same things I have said about it.” Stop making up things (like that) which simply aren’t true. I (and others) don’t buy that, as we see through it. This is tiring, even exhausting.

I purposefully do not address other points as so obviously inflammatory they do not rise to my concern.

Like the signs say at certain places: “Do not feed this animal.”

I’m out; this seems a lost cause.

1 Like
119

I’ve spent a few days on reading everything related to Overture Maps, and summarized it in an article: Overture Maps расправил плечи (it’s in Russian, please use your favourite translation service).

11 Likes
120

This thread has gone in many directions over the past few days. As a reminder, when you reply to a previous message, you can click on the :arrow_right: button above the comment box, to the left of the topic subject line, to reveal options for splitting out a new topic or moving to direct messages. This keeps the rest of the topic’s participants from getting dragged into tangents they’re less interested in. Thanks for understanding.

12 Likes
121

This is a well-written and seemingly comprehensive article, right on topic! Anyone reading this thread likely benefits by reading it, whether you agree with it or not. (Though, it does try to “stick to facts,” largely speaking, and that is my quick opinion).

An excerpt:

The problem with this whole system is that it is based on OpenStreetMap. A million people with whom it is impossible to negotiate, and a stupid data format. The OSM Foundation is unaccountable volunteers, the same can be said about the Council. No one knows what they want or what their plans are. No one expects them to do something useful to anyone other than the editors (although it’s sad). The whole system is extremely antagonistic. That’s why the Overture Foundation emerged.

Along with some amusing Russian-to-English translations (“Believe the Foundation’s statements: they really want to make a new potato basis available to anyone”), it purports to be a blueprint for the often-worried-about hand-wringing regarding all that is wrong with the OSMF and how “it will be assimilated.” (A reference to the “Borg,” a Star Trek, fictional, star-sized, hive-mind / AI which “eats good people”).

Check it out. The link is clean and many browsers offer in-browser translation.

122

Maybe you aren’t aware of them, but there’s been multiple discussions about how new users don’t respond to welcome messages. It’s rare that people even respond to changeset comments. While I agree that people like us who put a lot of work into the social aspects have some experience of community, most people don’t. To give another example, you can look through most discussions here, on the Wiki, and on the mailing list. Most of the it’s the same 15 people asking questions and discussion things that it’s always been. Usually less. I’m sure you’d agree there’s way more then 15 users out there actively mapping.

It’s not doxing when you’ve openly said where use to work in public discussions, including in a mailing list discussion a few weeks ago. Nice try though.

Sure. But most people clearly aren’t forming those relationships. I’m not the one saying it either.

100% the reason I pointed out there are no community based groups in the United States cleaning up the TIGER data is because I want there to be and this seems like the type of discussion while floating those kinds of ideas would be useful. It has nothing to do with throwing rocks at anything. Let alone I’m disparaging “everybody” by pointing out something that’s an obvious fact. Anymore then would be if I said there’s no one mapping glaciers in Antarctica if there’s no one mapping glaciers in Antarctica. Whatever rock throwing you’d take out of something like that is 100% on you for being unwilling to show me the slightest amount of good faith.

I never claimed I was building community by saying what is wrong with it. Don’t put words in my mouth. You can never build or improve on a community if the slightest suggestion that it might need improving is met with gas lighting and vitriol though. It’s fine if we disagree on the specific details. I’m not the enemy here.

I can back up most of what I’ve said with evidence. And like I said other people are saying the same things I have been saying ad nauseum. In the social sciences there’s a concept that I think is pertinent to this. It goes something like this, there are personal troubles and public/social issues. Personal troubles involve an individual’s private problems in relation to others. Whereas, public/social issues are forces which are outside of the personal control of an individual. Community building by it’s nature is not a personal trouble, it’s a public/social one. I can’t not privately build community on my own. Me and my counslering are not the OSM community, obviously. Community building is also something other people in the project have a problem doing. So whatever dirt you want to throw about how I should go see a counselor or whatever if I think there’s no community, it’s not on me.

All I can do is point out where I think it’s lacking and hope that others will agree and we can come up with a remedy. That’s all I’m doing here. Nothing more, nothing less. The reason I decided to bring it up here is because I think it’s semi-related to why Overturemaps was created. Others disagree. That’s fine. We should all be to say what we think the causes of this might be.

I don’t have the time right now, but if you want I can find some quotes from other people that are extremely similar to what I said.

123

@Adamant1 @stevea your personal back and forward replies have been flagged as both off-topic and offensive in some cases. Can I ask you to move your personal conversation over a PM please?

Thanks for helping keep this space on-topic.

11 Likes
124

Sure. Sorry about that. I’ve actually sent private messages about similar topics to SteveA before but he mostly ignores them. So I think I’ll just end it there. I apologize if it distracted from the broader discussion though :+1:

@steveA your welcome to send me a private message anytime if you want to discuss it further. If not, I at least appreciate that you explained what you thought was wrong about what I said to the small degree that you did.

125

What the heck does it mean to “co-collaborate”?

Thanks for your article! I have a question about these bits (translated with DeepL):

The license is already known: it is CDLA Permissive 2.0: analogous to MIT and CC-BY, requiring only a source. (…) A special permission will be given for use in OSM. Of course, all OSM-based data will be published under ODbL.
(…)
We even win in terms of sources: we used to suggest that companies and government agencies open data to OpenStreetMap. That led to long discussions about why and why. Now everyone will be opening up data for a solid project with millions of dollars in funding and dozens of developers on staff, supported by all the well-known companies from the top of the rankings. And that data will all be in a single format. It will be elementary to take them to refine OSM.

So OSM data will be “released” by them under ODbL, and I assume only their own data will be released under the CDLA. You mention that CDLA is something like MIT, but if that’s the case why the need of a “special permission” to be used in OSM?

1 Like
126

With permissive licenses, it’s not always obvious whether our way of attribution (via a link under osm.org/copyright → wiki/Contributors) would be okay with publishers. The OSMF LWG analysis suggests it’s fine, but you can never be safe with licenses. Hence the explicit permission is better than nothing.

As far as I got, companies (including Meta) plan to continue improving OSM with the externally provided data, which would be aggregated on Overture Maps portal. Hence the permission would be obvious.

3 Likes
127

TL;DR: I think they’re going for UUIDs as concept identifiers and their file format will be inspired by the IMDF (vocabulary very, very similar to OSM Wiki, but uses “.” over “:” / “_” on keys)

Based on quick casual browsing on Twitter for “overture maps”, then people proud about this, I found someone’s from one of these company days before mentioning this (which might give a hint on their mindset to push for broader adoption of whatever data exchange format to sideline OSMF over time):

From their “solutions” one is the GTFS from Google (pointer as success case for bus routes) and a new concentioned data package, Indoor Mapping Data Format, which seems to be around since at least 2019. But is from Apple, not AWS/Meta/Microsoft/TomTom:

At the https://www.ogc.org/pressroom/pressreleases/4415 explains a bit how this has been standardized, but I’m somewhat a bit disappointed if Overture Maps Foundation will either use IMDF done by another company or create a conventional in similar way. From the “war” on open vs closed specifications, I remember the case of Office Open XML had a major discussion, even an ISO, to brag about be an open standard with wider discussion, but OMF (if going this path) will effectively have an “open standard” not more than an informational IETF RFC. And from this link, a paragraph:

An OGC Community Standard is an official standard of OGC that was already available as a widely used, mature specification, but was developed outside of OGC’s standards development and approval process. The originator of the standard brings to OGC a “snapshot” of their work that is then endorsed by OGC membership so that it can become part of the OGC Standards Baseline.

So, this approach on file format is actually less than the OpenStreetMap Wiki: the actual “points of interest” are poorly defined, if defined at all over mere key use in en-US. Pretty lazy job, I mean is not even monolingual. The JSON/Geojson is not even JSON-LD/GeoJSON-LD. Not just this is less semantic, but seems that the whole “Global Entity Reference System” is a buzzword for long random UUIDs that any tool can generate but are unlikely to conflict with each other (actually, "globally unique identifier GUID are another name used for UUIDs).

Despite them saying in the FAQ they aren’t focused on community, and informally we know here they’re sometimes upset with DWG not allowing mass changes, from the “OGC Community Standard” thing, plus all the idea of trying to appear friendly to OSM, seems otherwise. Also even GTFS from Google had very slow adoption despite having no competition of file formats, so Overture from medium to long term attempt requires developers to promote use of their exchange format (if they don’t, it will make them work very, very, very manual, unable to scale up).

Anyway, they’ll be very susceptible to public criticism about technical implementation (because otherwise their data would be more outdated pretty quickly). But if they’re going for an IMDF-like approach, from their way to think like layers, to even the way they would do conflagration at later stages, becomes predictable, regardless of they avoiding leaking any code before going public.

2 Likes
128

OSMF official replay in the blog.openstreetmap.org on December 22, 2022 by Mikel.

https://blog.openstreetmap.org/2022/12/22/views-from-the-openstreetmap-foundation-on-the-launch-of-overture/

5 Likes
129

Discussion is happening next door: Views from the OpenStreetMap Foundation on the launch of Overture

4 Likes
130

A few more blog posts from various people in the OSM community and beyond:

(please add your own… this post is a wiki)

12 Likes
131

As far as I can remember OSM has always been community-driven raw map data project, with focus on manual, on the ground mapping of reality with no one specific use case in mind. Now any data usage (especially complex data like maps) is not just taking raw data and use it as is. You need structuring, cleaning, quality control, mixing with other datasets etc, and all these steps are purpose-driven. Therefore from the ancient times of cloudmade (and mapbox, grofabrik, skobbler etc etc) OSM users for real business (or fun) had to build their stack on top of OSM. I’ve been involved on a few these initiatives in past, same thing every time: inconsistent coverage, tags etc. Amost all of these downstream companies have kept it private and closed in terms of both tech and “clean” data, and until OSM attribution is included it has been ok.

Now Overture seems to build same old stack, but in completely new level: with backing and collaboration of major players and at least promise to make resulting data also open. Thats something mapboxes of the world never did. And this can be a game changer thing, it can become partial replacement to these OSM commercial sellers, but not to OSM community. It is a play to outcompete no less than Google Maps. I see OSM has to win from here more than loose.

6 Likes