v10 POC

[mfazekas]

V10 POC

It's a bit rough as removes most of functionality but shows up a map for the example app.
Random notes:

• Since RN requires C macros, we still need to use some Objective-C code, for the RN macros
• There is no annotation with custom view

[ferdicus]

dude... :D

👏🏿 👏🏿 👏🏿

[systemlevel]

Impressive @mfazekas! That is very cool. Excellent POC!

[mattijsf]

mapbox/mapbox-gl-native#12413 (comment)

A VERY big reason I'd like this PR to happen. I haven't looked in-depth to v10 yet but I'd like to know what challenges there are for supporting v10 and if/how I can provide support with this PR.

[mfazekas]

I think this POC shows that using v10 is possible. It is a significant but straightforward effort.

v10 does not support some stuff like custom views in annotations ATM, so no full compatibility is possible now.

Unfortunately, I don't have much time to deal with this at the moment. If someone is interested to sponsoring, that would allow to spend more time on this.

[doozMen]

I work for cowboy.com and we extensively use Mapbox. Almost all code is based on support for annotations that are views. It is nearly impossible to move to system where we snapshot the views that we update the PointAnnotation its image. Or is it really the idea to do that. Why not provide a system like apple does where for the annotation a MKAnnotationView is requested. Also updates to the view on the map are then done requesting an update of the view of an annotation to the map. Simply not allowing views seams harche. So could this PR be made a priority so we can plan our upcomming release?

[ferdicus]

I work for cowboy.com and we extensively use Mapbox. Almost all code is based on support for annotations that are views. It is nearly impossible to move to system where we snapshot the views that we update the PointAnnotation its image. Or is it really the idea to do that. Why not provide a system like apple does where for the annotation a MKAnnotationView is requested. Also updates to the view on the map are then done requesting an update of the view of an annotation to the map. Simply not allowing views seams harche. So could this PR be made a priority so we can plan our upcomming release?

I don't see any upcoming release of this PR in the near future - the only person that was working on this was mfazekas and as mentioned it's a time sink.

People are commenting that they would like to see this move forward, however I don't see that happening without proper support of someone that has time.

So, if this is important to cowboy.com maybe think about supporting mfazekas directly to move this forward.

[engelzero]

@mfazekas how much time roughly would you estimate to implement this PR?

[doozMen]

Ok thanks for this feedback. We will postphone the update and fixed some issues in the earlier version so his is less important now. Probably during the summer we would like to update do.

[sedot42]

@mfazekas We'd also be interested in a rough effort estimate!

[mfazekas]

@sedot42 @spotelf13 I'll try to came up with a rough estimate beginning of next week.

[systemlevel]

Is there any group here that is interested in helping this move forward?

We’d like to help but are not sure who else is interested in helping push this forward.

[systemlevel]

I have a company that is interested in helping move this forward but would like to work with someone or another company in our community here. They have funding, etc. Please reach out if you are interested in either receiving payment for helping with this effort or have resources to put into it.

[systemlevel]

@doozMen is your company interested in the possibility of coming together to move this forward?

[sedot42]

@systemlevel We are potentially interested, but still not sure whether not having v10 is a blocker in our project.

[mfazekas]

My initial very rough estimate is 100 hours for the iOS part and about half of that for the android part.

Some notes:

• Supporting v10 means 2 fork of the library, one for maplibre and one for mapbox v10. That likely means less focus aka less contributors will be split between v10 and maplibre.
• PointAnnotation on v10 still supports just an image, not interactive views. Not sure if that will change in near future in v10, or if we can implement an android MarkerView like workaround for the issue.

[sedot42]

Thanks @mfazekas! We will know in the beginning of September whether we need v10 and have budget to support your work. Using standard dev rates I assume we might be able to cover 10-20% of that effort at maximum. I'll put in a good word seeing that we already benefit from your (and all other contributors <3) existing work.

@systemlevel @spotelf13, what's your status here?

[systemlevel]

Thanks @mfazekas! We will know in the beginning of September whether we need v10 and have budget to support your work. Using standard dev rates I assume we might be able to cover 10-20% of that effort at maximum. I'll put in a good word seeing that we already benefit from your (and all other contributors <3) existing work.

@systemlevel @spotelf13, what's your status here?

We are good to help support these efforts. We are just trying to get a read from the community about who else will be able to support these efforts. The source I'm working with is saying they'd like to take different approaches based on whether they are the sole funders or others participate in funding it together.

[sedot42]

Unfortunately we won't have budget to support this. :(

[ferdicus]

hey there 👋🏿
To everyone who can not contribute monetarily, I'd like to encourage you to join the repo as a maintainer.
We have a notice with an attached discussion on the front page within our README.

Please have a look and talk to your lead/ manager/ partner if you can spare a bit of time every week to help us keep this project alive.

Cheers 🍻

[mvanroon]

@ferdicus @mfazekas have you guys considered asking Mapbox for support? They mention this library on their website and might be interested in helping out (financially or otherwise). I can reach out to them if you like. The more people using this library, the more people end up paying for Mapbox, right?

[ferdicus]

@mvanroon, may I present to you exhibit A

Executive Summary:

• Was mapbox supported
• support dropped

[doozMen]

@doozMen is your company interested in the possibility of coming together to move this forward?

sorry for the very late reply but we are yes

[mvanroon]

@mvanroon, may I present to you exhibit A

Executive Summary:

• Was mapbox supported
• support dropped

Objection your honour! Even if they don't want to put engineering effort in it, they might be willing to sponsor @mfazekas so he can make those 150 hours. Worth a shot right?

[ferdicus]

I mean, yeah, sure - go for it.
Worst that could happen is them saying "no" 🤷🏿‍♀️

[mvanroon]

@mvanroon can we find some time for a call to determine what this would take? I'm tlee (at) mapbox (dot) com -- please send me an email so we can explore the possibility.

@ferdicus @mfazekas Would either or both of you be up for a call with Tom Lee to discuss moving forward to Mapbox v10? I could join too, but I don't think I could be of much help :-)

[mfazekas]

@mvanroon can we find some time for a call to determine what this would take? I'm tlee (at) mapbox (dot) com -- please send me an email so we can explore the possibility.

@ferdicus @mfazekas Would either or both of you be up for a call with Tom Lee to discuss moving forward to Mapbox v10? I could join too, but I don't think I could be of much help :-)

I'll contact Tom Lee.

[Dakuan]

I'm a (paying) mapbox customer. I won't be for much longer if there's no likely support for React Native, for what its worth I've messaged Mapbox to tell them this.

[EricPKerr]

@TruckMap would be interested in helping sponsor the development costs of this. It would be great if we could get a few companies to pool together to support the engineering time.

It sounds like @systemlevel and @doozMen represent companies that would be interested (in addition to hopefully @sbma44 at Mapbox). Are there any others?

@mfazekas is this still something you'd be able to take on if we put together the funds? Any updated available start time, or changes to the 150-hour estimate you mentioned above?

[systemlevel]

V10 Update

I'm happy to announce that we have worked out the funding details and have been working on the V10 Mapbox library. We are roughly 75% to 80% through the transition to V10 working on the initial pass for iOS and Android.

As a result of this, we want to make the work we have been doing available to the open-source community. We also wanted to get feedback from the community in relation to where this open-source repo may reside.

1. Where is this particular repo headed? Is this going to eventually turn into a maplibre repo or stay a V6 Mapbox repo?
2. Do we want to use this repo for V10 as well? I am not so sure about this as the namespace indicates this repo is based on gl. V10 is based on Metal for iOS now among other major changes.
3. If we decide to use this repo, I currently do not have any authorization to make updates to this repo or any level of control of the repo.
4. Yes, we would be open to coordinating additional funding. As of right now, we've been the sole funders of the V10 transition but we can discuss how that might work.

A big thank you to all of those who have supported this repository up to this point.

@ferdicus @mfazekas

[mvanroon]

Fantastic news guys! Looking forward to the update

[ferdicus]

V10 Update

I'm happy to announce that we have worked out the funding details and have been working on the V10 Mapbox library. We are roughly 75% to 80% through the transition to V10 working on the initial pass for iOS and Android.

As a result of this, we want to make the work we have been doing available to the open-source community. We also wanted to get feedback from the community in relation to where this open-source repo may reside.

1. Where is this particular repo headed?  Is this going to eventually turn into a maplibre repo or stay a V6 Mapbox repo?

2. Do we want to use this repo for V10 as well?  I am not so sure about this as the namespace indicates this repo is based on gl.  V10 is based on Metal for iOS now among other major changes.

3. If we decide to use this repo, I currently do not have any authorization to make updates to this repo or any level of control of the repo.

4. Yes, we would be open to coordinating additional funding. As of right now, we've been the sole funders of the V10 transition but we can discuss how that might work.

A big thank you to all of those who have supported this repository up to this point.

@ferdicus @mfazekas

Thanks @systemlevel , I opened a follow-up discussion to collect feedback on this topic here: https://github.com/react-native-mapbox-gl/maps/discussions/1680

Let's move the conversation there.