en.osm.town is one of the many independent Mastodon servers you can use to participate in the fediverse.
An independent, community of OpenStreetMap people on the Fediverse/Mastodon. Funding graciously provided by the OpenStreetMap Foundation.

Server stats:

261
active users

#odbl

0 posts0 participants0 posts today
Replied in thread

@karussell They don't explicitly say that they're incorporating Open Street Map data---instead saying that you can bring "your" data alongside Google's---but, as evidenced by your screenshot, and by Google's intro video here at the 41 second mark:

youtu.be/g9F-_tCakL8?t=41

It certainly seems like Google views Open Street Map data as fair game for inclusion within their "Geospatial Reasoning" product. Their promotional material shows a map rendering that bears a very strong resemblance to Open Street Map's default "Carto" raster tile rendering style, encouraging their potential customers to believe the same.

That is as clear a violation of the ODbL as can be, which was written in large part specifically to exclude Google from using OSM data in its own customer-facing products. Laundering license violations through AI doesn't make them not violations.

Replied in thread

@datagouvfr euh... uniquement les bases de données dérivées, pas toutes les réutilisations.

Exemple:
- je produit une carte à partir des données #OpenStreetMap (qui sont en ODbL),
- ma carte peut être dans la licence de mon choix (y compris non libre) et sûrement pas #ODbL car le plus souvent une carte ce n'est pas une base de données.

Non-altération et date du millésime ne sont pas des obligations de l'ODbL et le #RGPD est un tout autre sujet à ne pas mélanger.

Replied in thread

@hareldan I'm not sure #STAC (SpatioTemporal Asset Catalogs; stacspec.org) support for OSM editors would be worth the effort: Is there enough STAC-indexed imagery (or other STAC-indexed data) useful for mapping that is licensed such that using it for OSM mapping (and re-licensing the result under #ODbL with just "central" attribution on the OSM wiki rather than attribution embedded in the thusly created or modified parts of the OSM data and any derived DBs & works) is allowed?

stacspec.orgroot.meta.titleroot.meta.description
Continued thread

Well, I'm reading through it now, mere days before the book report's deadline (life has been kicking me while I'm down this past month; just lemme have this one). Keeping in mind that this was published in 2015, and likely worked on in 2013-2014, I'd say, so far (I'm finished with the first two, of eight, chapters), that the authors are... wildly optimistic in their prognostications. Their faith in capitalism is thoroughly baked into -- and yet completely unspoken in -- what I have read so far, and it borders on naïveté. Ethical questions are barely even raised, and there is no investigation made into them at either the textual or subtextual level.

As a former software engineer, their hope in Big Tech -- and particularly companies like Facebook and Google (the authors go so far as to nonchalantly suggest that a "genomic 'Facebook'" might one day exist, without casting any skepticism or aspersion toward that idea at all) -- is particularly misplaced when viewed through the lens of a decade of hindsight into the mess that surveillance capitalism has wrought. #23andMe has been presented as a success story, a particularly tragicomical specimen in light of its recent financial troubles and questions over ownership of the genetic data should it go bankrupt.

And that leads into what I suspect will be my primary gripe with the book: its (suspected) dearth of considerations for the licensing of genomic data and technology, about who gets to benefit from amassing such a corpus of genomic information, and on what terms. I have found myself at multiple instances, just in the first two chapters, already exasperatedly asking the authors: haven't y'all seen or read anything put out by @aral on the questions of data ownership? Is the field of #bioinformatics truly sleepwalking into the exact same quagmires that the #FreeSoftware world contended with, all without ever reading anything put out by @rms, by the @fsf, by the @conservancy, or by the @fsfe? Is our species, in yet another instance, going to prioritize private profit at the expense of public good? Nay, even publicly subsidize a for-profit sector that has no obligations to the public in turn?

If genetics truly is the source code of life, and our existing licenses for such source code are truly so "permissive" (or even non-existent) then our species is in very sorry shape indeed. To avoid a #cyberpunk future, I implore and entreat that we take inspiration from nature's viruses, and adopt viral, copyleft licensing for our collective genetic source code as a species. Genomics must belong to everyone, lest it become yet another tool of oppression. "Free Genomics, Free Society," as it were.

I see the nascent optimism of authors Field and Davies and raise them an equal dose of pessimism, administered one decade later.

Replied in thread

@march ja, finde es eine tolle Entwicklung, dass das mittlerweile fast selbstverständlich ist, wenn neue Datensätze/Dokumente veröffentlicht werden.
Leider ist "CC BY 4.0" nicht mit der #ODbL kompatibel, weshalb dieser Datenschatz leider nicht zur Verbesserung der Datenlage in #OpenStreetMap verwendet werden kann:
wiki.openstreetmap.org/wiki/Im

@BNetzA

wiki.openstreetmap.orgImport/ODbL Compatibility - OpenStreetMap Wiki