You are being too cynical. There's a lot of data outside of open streetmap that complements open streetmap. A lot of that data is open but hard to combine with openstreet maps. And some of that data is owned by companies that are willing to license the data in an open way.
Overture basically is a joint effort by several companies to combine all that data.
Speaking as someone who has worked with OSM data. It's great for maps but severely lacking elsewhere with a lot of incomplete data, poorly/inconsistently tagged data, lot's of regional variation in tagging, etc. All this presents challenges for users of this data wanting to build stuff on top of this data. And there are lots of companies that are replicating efforts to fix this between each other. Been there done that. This is hard, non trivial work.
Overture is an attempt to move on from lots of companies reinventing this wheel to get to a state where there is a decent data foundation to build their applications on.
And they are releasing that data under an open license. So, there's a lot to like here. The process of how this data is produced is not as open unfortunately and it is unfortunate that they are doing this outside of the openstreetmap community.
But then perhaps that community wasn't that welcoming to get such a thing done? Nor have they seem capable or willing to do such work themselves. Overture are clearly working around them and it's worth spending some time reflecting on whether that could have worked differently and what would have had to change on both sides for that to happen.
They seem quite happy to use OSM data. We'll see how this stacks up w/r/t license: from here (and obviously IANAL), it looks very much like "sure, give us all the ODbL data you have, but we're doing a magic trick here, and the end result is not ODbL, hee hee". https://docs.overturemaps.org/release-notes/data-attribution...
That page seems to save that almost all of the data except places IS OdBL and requires attributing OpenStreetMap?
OSM POI data is definitely severely lacking a lot of regions -- so it makes sense why they'd source that from somewhere else. It's also pretty cool that meta and microsoft's POI data is now open -- from look at it, it has the opposite problem of OSM: there are a lot of businesses in it that don't exist anymore or whose info isn't totally accurate -- but it's very usable for certain use cases (and, funnily enough, could certainly form the starting point for a clean up effort -- that COULD then be stuck in OSM and licensed under OdBL).
it seems to be making new map data under a permissive licence instead of openstreetmap's copyleft one, which makes it possible to coopt the work of volunteers
Looks like the only permissively licensed part is the "Places" dataset provided by Meta and Microsoft https://docs.overturemaps.org/release-notes/data-attribution... so the volunteers being coopted are e.g. people who made a Facebook page for their business?
Any new data not in OSM is under a separate dataset and not under a copyleft license, so anyone contributing to that in the future means that their data might be used in a proprietary map.
In general, this seems to be more aimed as a corporate-friendly replacement of OSM where instead of volunteers, its companies with satellite data and phone data and AIs and only the end result is shared, not the process. So you're right that its not really being unfair to the OSM community whose data they seem to only reluctantly want (because it's copyleft) and have kept as such.
I am worried though that if if this gets popular, that OSM will be starved for funding and contributors
> anyone contributing to that in the future means that their data might be used in a proprietary map
But overture doesn’t generally allow the type of contributions OSM allows. There is no volunteer to map out a set of houses or streets. Allowed types of contributions are large mapping datasets, typically of the type explicitly disallowed from being imported to OSM.
It's fine as long as bike paths are physically separated from cars and are maintained to the same standard as roads. My college had a network of pedestrian/bike paths that they plowed in the winter, and I and many others had no problem biking year-round.
Here's Seattle's network of protected bike lanes: <https://i.imgur.com/vEDhqoz.png>. Note that many of them are flawed in one or more ways:
* There are frequent car crossings at blind alleys and driveways, often every 20-30 feet.
* Many are "protected" by flexible posts that are easily driven over.
* When protected bike lanes run next to roads, drivers often turn across them at intersections without looking.
I generated the map with Overpass Turbo: <https://overpass-turbo.eu/s/1mT6>. To make it render nicely, check "Don't display small features as POIs" under settings. If you want to try it for a different city, drag/zoom it into view and hit "Run".
The "trekking bike" and "safety" profiles seem to work pretty well. The safety profile produces routes I would take any time of day and the trekking bike profile produces routes I would take outside of peak road rage hours.
I used to ride as far to the right as possible on roads without bike lanes, but then I got hit by a driver trying to squeeze past me without changing lanes.
It appears to use C3D as its CAD kernel, which is proprietary. You can't build it without a copy of C3D. I'm guessing the source will be available for inspection, but the binary will cost money.
A full build from scratch would require the c3d headers, which I can’t provide.
What I am planning is to something like this: divide the app into two pieces, an npm wrapper around the kernel and the electron front end that uses the module.
You will need a license key to “activate” the node module. You would then use the electron app or the module directly if you’re a programmer. You will be able to build your own fork of the electron app, but not the npm module.
In effect you will be paying for a license key, not the binary
I will say I’m constantly bewildered by which license to choose. I’m using copyleft to defend myself a bit. But weak copyleft because I’ve hate that the gpl prevents blender plug-ins from being closed source or linking against something closed source. Plasticity will allow plug-ins to do whatever they want, that’s the goal.
There is some possibility I will switch to MIT since philosophically I’m more in that camp
> I’m using copyleft to defend myself a bit. But weak copyleft because I’ve hate that the gpl prevents blender plug-ins from being closed source or linking against something closed source
So you're ok with using copyleft to defend yourself, but not ok with others to use copyleft to defend themselves ? This is bewildering
Their stance seems reasonable to me. They seem happy for works to be GPL, but frustrated that it means plugins or extensions or things which are not the core work themselves are then hampered.
That’s exactly my take. What exactly constitutes a “derived work” in the gpl (plug-ins do!) is counter intuitive to me.
Even the lgpl definition of “dynamic linking” is something that scares me in the context of npm modules, which is why I’m nervous I made the wrong choice with lgpl
But why would a GPL app need to force GPL on a plugin? That seems to me like a perfectly reasonable boundary between the two applications. If I wanted to leverage OBS for example and introduce my own plugin that made use of a closed-source library or tool I had the license for, I should be able to create it. My third-party plugin doesn't impede the distribution and usage of OBS in any way, only provides a separate feature that might require a different license.
As someone who spends most of his days on a GPL app which loads plug-ins, I do not want my work to serve the interest of proprietary software vendors more than strictly necessary, simple as that. I spent the effort of writing the app, defining its plug-in APIs, etc; if others want to benefit from this effort they should also spend the effort of writing GPL plug-ins. The end result is that no end user will ever be stuck because $PROPRIETARY api shut downs or change (and in my field I've seen that a ton of times).
> I should be able to create it.
Why do you feel entitled to that ? No one prevents you to go and write your own OBS-like or look for a proprietary alternative (and don't complain when they change or shut down their API with no possible recourse at all). But here you look you you want to benefit from the work of OBS contributors, without satisfying the simple obligation in return: the program that you redistribute must be open, just like it was when you found it.
Because I'm not beholden to your ideals. GPL is tantamount to a parasite, infecting everything it touches, regardless of the wishes of those who'd make use of the GPL'd software.
I was a contributor to FOSS until I had to make a choice to step back from excessive keyboard use due to RSI. I love and embrace open source software and the ideals behind it, but I believe everyone should be given their own freedoms.
The plugin boundary should be exactly that, a boundary between the primary app and any created plugins. Why do you feel entitled to force a license decision on a third party's work? Yeah, my work might not function without your code, but your code is your code and my plugin is mine. If I choose to write a utility that enhances your code in ways I want, is that not the benefit and flexibility that FOSS clamors about? Hell, what if someone else adopts a similar API and I choose to take my plugin that direction? That's my choice to make, but forcing GPL on my code may now be hampering my freedoms.
I hear the argument of "if only software X was open source, then I could fix this bug here or add this feature there". I'm right there with you on code that is in the mainline and lives inside the source, but if I'm writing a plugin, that's my work and I own that, not you.
If I choose to make use of a proprietary library because it makes sense to me, like for example CAD solvers where the FOSS alternatives aren't there yet, that's my decision for my plugin. All your GPL license does is infect my work.
That's why I don't use GPL, and why I agree with others that the end result of many GPL licenses is to simply reduce the overall amount of GPL code in the world, as people avoid it out of concern of risk.
> I believe everyone should be given their own freedoms.
Absolute freedom does not exist (or even make sense).
> The plugin boundary should be exactly that, a boundary between the primary app and any created plugins. Why do you feel entitled to force a license decision on a third party's work?
That's the expectation I set when I develop the host software. If I was fine with third party proprietary plug-ins I could set my plug-in headers as MIT or something. Just like if I only wanted my software be used by people who have no revenue or live in $COUNTRY I would put that as a license instead of the GPL.
> Yeah, my work might not function without your code, but your code is your code and my plugin is mine.
this is super super entitled. You're fine with leveraging GPL work which is release for free for your own benefit, but do not want to reciprocate. As I posted elsewhere: https://en.wikipedia.org/wiki/Golden_Rule ; this is the most basic expectation one can have.
> but your code is your code and my plugin is mine.
but the GPL code is and will forever be freely available ; anything that goes against increasing the pool of freely available thing is a no-go. The end goal of all that would be to change laws so that ownership of ideas, patents, etc..., is itself ended ; ultimately, licenses and copyright shouldn't even need to exist.
> If I choose to make use of a proprietary library because it makes sense to me, like for example CAD solvers where the FOSS alternatives aren't there yet, that's my decision for my plugin. All your GPL license does is infect my work.
Again you seem to forget that your plug-in depends on an existing host software and seem to want to use the work done there without giving anything in return. If there was some kind of public standard for CAD solvers so that the GPL part of your plug-in could be also used with free solvers, that would be a non-issue - ship your plug-in, and end-users can install a free solver or a proprietary one separately if they are fine with proprietary code.
I think it's fine for people to have varying philosophical takes on what should be the normative goals of open source software licenses. Obviously the FSF takes a rather maximalist stance. I personally find this aggressive goal and some of the personalities behind it to not match my vibe. And the reality is that more permissive licenses like MIT are becoming more and more popular.
What I've seen some Blender plugins do to get around this is to have an open-source plugin communicate via IPC to a closed source library. It's totally legal. It's a performance and implementation tax that just seems silly to me.
If the IPC is here only for the sake of circumventing the GPL (e.g. there aren't any meaningful use of the "GPL plug-in" part through this IPC channel outside of communication with the non-GPL program), it's totally not. What are such plug-ins ?
Why would it be illegal to use IPC to communicate from one GPL executable to another non GPL executable?
Files are interprocess communication too. If files are fine and other forms are not, then it starts to get into an argument of speed making something illegal.
Not only that, but network communication is IPC and that would imply that it's fine to have a GPL program go to another computer holding a non GPL program, but not if they are on the same computer. There are a lot of holes in this argument.
Not only that, but any program that can't communicate with other programs (do drivers count?) through file formats, networks operations or anything else is probably limited anyway.
I think so. I originally said there were a lot of holes in their argument, but I think they didn't actually make an argument in the first place, they just made a claim with no evidence behind it.
The GPL does not care about stuff such as libraries, executable, etc. The only relevant concern is whether your IPC-splitted thing is part of the program, or is a public API expected to be used by many unrelated other programs.
Quoting the GPL:
> The “Corresponding Source” for a work in object code form means all the source code needed to generate, install, and (for an executable work) run the object code and to modify the work, including scripts to control those activities. However, it does not include the work's System Libraries, or general-purpose tools or generally available free programs which are used unmodified in performing those activities but which are not part of the work. For example, Corresponding Source includes interface definition files associated with source files for the work, and the source code for shared libraries and dynamically linked subprograms that the work is specifically designed to require, such as by intimate data communication or control flow between those subprograms and other parts of the work.
The relevant part is "intimate data communication": querying for instance some other program's publicly-advertised D-Bus API, such as a notification service expected to be used by the whole operating system wouldn't be "intimate data communication". However splitting a single executable into two parts which don't individually make sense are obviously "intimate data communication" - it may be two binaries, but it's still one single software from the point of view of the end-user.
If the GPL part of your plug-in cannot do meaningful things (basically still be useable for its original goal by the end user) without the other, proprietary, side of the IPC up and running, then they have "intimate data communication" and distributing it is breaching the GPL.
I have a hard time imagining how one would justify that a CAD plug-in split under two parts for IPC wouldn't constitute a single work for instance, with the following GPL paragraph in mind:
> A compilation of a covered work with other separate and independent works, which are not by their nature extensions of the covered work, and which are not combined with it such as to form a larger program, in or on a volume of a storage or distribution medium, is called an “aggregate” if the compilation and its resulting copyright are not used to limit the access or legal rights of the compilation's users beyond what the individual works permit.
You mentioned drivers: they are specifically denoted in the paragraph above ("it does not include the work's System Libraries").
First is a plugin that would be shared library loaded into the same virtual memory space as a running program.
The second is GPLed header files that define function signatures and data structures.
The third scenario would be passing already defined data formats like JSON or something serialized through the file system, shared memory, or the local network. In this situation, there would be separate processes, separate memory spaces and no interface definitions would need to be included into the non GPLed program. This situation can't violated the GPL because it would imply that almost any communication with a GPLed program would violate the GPL.
> This situation can't violated the GPL because it would imply that almost any communication with a GPLed program would violate the GPL.
any intimate communication would indeed. Any communication that goes through a standard (e.g. ISO-defined or industry standard, so not something specific to your company) would not.
Quoting the FSF:
> Using shared memory to communicate with complex data structures is pretty much equivalent to dynamic linking.
You can obviously extend "communicating complex data structures" to RPC through JSON & websockets for instance as it's this "exchange of complex data structures" which is the defining matter in what constitutes a single program from the point of view of the GPL.
I gave specific examples and you keep using an abstract label.
> Using shared memory to communicate with complex data structures is pretty much equivalent to dynamic linking.
These are two separate things unless they are talking about shared memory meaning sharing the same memory space as another process, which is what dynamic linking does.
Communicating through complex data structures can describe files too. If you save a file from one GPLed program and open it in a non GPLed program, that is interprocess communication. If both programs are using mmap on the files instead of writes, does that change things? If files are being sent over local loop back does that change things?
Are there any examples of any of these ideas being tested? I've never heard of anyone except for you asserting that two processes communicating without including GPLed source code was somehow a GPL violation.
> I gave specific examples and you keep using an abstract label.
I am quoting the license text.
> Communicating through complex data structures can describe files too. If you save a file from one GPLed program and open it in a non GPLed program, that is interprocess communication. If both programs are using mmap on the files instead of writes, does that change things? If files are being sent over local loop back does that change things?
I don't thinkg it would at all - it would still breach the license if e.g. you are dumping raw struct contents in your files that are only meaningful for the proprietary side of things.
The part you are quoting is still an abstract label. You can't take specific common scenarios and say 'intimate communication is banned' when 'intimate communication' is not defined well.
you are dumping raw struct contents in your files that are only meaningful for the proprietary side of things.
What does that even mean? You could work with the bmp file format internally as a frame buffer in a program because it is uncompressed. Does that mean if you write a bmp file out of gimp and open it in photoshop you are violating the GPL? If so, why has that never been brought up or asserted by any person or company ever?
The things you are saying don't seem to line up with what anyone else thinks. When has anyone else tried to claim that you can't send files from one GPL program to another non GPL program?
I think you should consider whether you are trying to claiming something is true because you want it to be true, not because it makes sense.
IPC is totally worthwhile for native applications in any case, when separate complex components are combined into a single user session. Running the code in separate heaps reduces the chances of bugs in one application code corrupting the entire application heap.
Very hard to argue that IPC would ever be just for avoiding license. It always increases total robustness.
I agree with you -- it's hard to safely kill a runaway thread but it's not hard to kill a runaway process. But I think it's worth noting that the performance characteristics of IPC can be very bad
As a very tangential side note, behavior like this is fairy common, and a good reminder human being are not rational actors. People drive like asshats when they're in a hurry, but are very upset when someone else drives that way. Rather than bewildering, I would call it human nature. What is rare though, is being very honest about it.
> People are free to use whatever license they want and I’m free to explain why I choose lgpl over gpl
choosing a license is a political act whether you want it or not - you are free to choose whichever license you want, that does not mean you won't get called out (like for any other choice with political relevancy even tenuous - eating meat, driving a car, etc. to speak about things I am guilty of for instance)
We shouldn't have to live in a culture when being called out for making a 'choice with tenuous political relevancy' is socially expected, and something someone is 'guilty of'.
surely you see how your position is not workable in any decent society ? this is literally one of the only common rule across humankind: https://en.wikipedia.org/wiki/Golden_Rule
I thought I am the copyright owner and I can offer my code under as many licenses as I want. Many people dual license gpl code is my understanding. Converting my own code to mit should be allowed right?
Every Google result I’ve found says I’m the copyright owner and I can offer as many licenses as I want. Are you just trolling?
So external contributors can only provide patches without being able to test themselves I suppose.
Is the c3d dev license cheap enough for determined contributors to join? As in, cheaper than a f360 maker license?
I do not mean this in any pejorative sense. As a dev/maker I kept an eye on c3d for a long time, since that seems the only advanced-enough and commercially affordable brep kernel around to get off the ground quickly.
However, there's no discussion the closed nature pretty much bars any sort of in-depth contributor.
At least, contributing to a project like this would be extremely off-putting for me, to the point that besides having the ability to look a bit deeper than usual, I question whether keeping the source open does much.
> So external contributors can only provide patches without being able to test themselves I suppose.
My intention is that if you buy a license for plasticity, you can then build locally and test locally. You can contribute back or not depending on your interest.
Think of it like this. There will be a typescript/javascript wrapper around a limited version of the c3d kernel. this is the plasticity api. You will call plasticity.Enable(license_key) at the top of your program and you will be good to go. You buy a license key from me.
Although I do hope people will contribute the plasticity's development, my main goal with it being open source is that people will write plugins that they can then give away or sell themselves.
I have used commercial software that I pay for -- like Fusion 360 and MoI3d -- where I ran into bugs that I could have fixed for myself if only I had the code. I'm still happy to pay for them. Instead I literally waited 2 years for Fusion to fix a bug I cared about.
It is for sure a murky area where I'm working out the details. You don't have the full kernel. But you can think of it a bit like a scripting API on top of the kernel, which is not uncommon in CAD apps, like OnShape's FeatureScript or something.
