Cartes.app vs Murena Maps

/e/OS Foundation & Community Tech News
,
1

Hello,

I hope I’m posting this in the right place…

I recently discovered the site cartes.app, which offers a free, open-source map based on OpenStreetMap that’s already quite advanced.

I know that Murena Maps doesn’t really serve the same purpose, since it’s a smartphone app rather than a website. But I find that there are many similarities in the approach behind the product (an interface accessible to as many people as possible, the same features as Google Maps, etc.).

Do you know if there are plans to collaborate or pool resources with the cartes.app team in the development of both apps?

Or is there a plan to fork this map into the /e/ ecosystem since it’s open-source?

Anyway, just a few thoughts…

Have a great day!

Regain your privacy! Adopt /e/OS the deGoogled mobile OS and online services

phone
phone427×374 173 KB

2
3

Hello @AnotherElk.

I’m not sure to understand your post. Do you mean that the answer to my question is in the link you shared (in that case there no mention of other maps app in the Q&A for me…) or do you mean that my post was in the wrong location and should be in the thread about murena map?

Thx

4

Yes, this. Sorry, I could have elaborated a bit to not leave you guessing.

In all likelihood there will be no new fork, as Murena are already developing their own App and will put it into /e/OS as the default Maps App as soon as they can.
And I don’t see them engaging in co-development of a different App or website, but I’m not Murena, so we perhaps just wait a bit whether anybody else can pitch in with some info.

5

Hi, I’m the founder of cartes.app.

Here’s a translation of the message I’ve posted on the OSM-FR forum.

We’ve talked with Murena about Murena x cartes.app. But Murena wanted a native app, not a web app. And an online app, not an offline one. That rules out cartes.app on one side and Comaps on the other. For my part, going native is out of the question. I totally respect and understand Murena’s decision : it’s a mobile OS, a native app is in line with this of course.

For our part on cartes.app :

First, because I don’t want to deal with app stores. Developing native apps outside the two dominant stores offers no distribution advantage over the web, which has the benefit of being universal. The web’s potential for place discovery—through search engines and LLMs— has been way underused by the OSM community. We’re doing our best, but we could do 10x better.

Second, because I believe the web is more than powerful enough to do everything a modern native map app can. Granted, we haven’t had time to fix some performance issues yet. Firefox’s lag behind Chromium/Safari doesn’t help: on a modern phone though, the app is perfectly smooth. We also haven’t had time to develop a first-person GPS mode. Hence the plan to redirect to Comaps (or maybe Murena Maps someday, your choice) for GPS mode. :slight_smile:

Recently though, we’ve started implementing the GPS mode. It’s less complicated and sensor-reliant that we thought. Expect a beta version in the coming months :slight_smile:

Murena will probably need a Web version too, as Comaps and Osmand need it to cover the scope of the Web, e.g. opening a link on desktop without installing an app. Also note that on mobile, opening the cartes.app PWA is already faster than Comaps !

Third, because native development is a whole different world, with massive constraints. For me, that’s a major barrier, and a problematic distraction for a small team like ours. And we’re not alone—even huge teams struggle: no app has both a solid web map and a solid native version.

Of course, there’s no opposition between these apps. They all strengthen OSM’s legitimacy, and I think the next step will be to pool resources for everything OSM refuses to cover: place photos, place reviews (which are more or less the same thing), traffic data, GTFS transit data, etc.

Let’s collaborate on these hard-to-maintain dependencies ! For instance, the Valhalla demo server (powering cartes.app’s car mode) has been down for two weeks. Murena will need this kind of router, it will need Transitous and Motis as well !

Dispersion on the UI and end-user brand development is a problem to create alternatives to big tech, obviously. But for now, no open-source initiative has shown they are the obvious all-encompassing initiative to G/A Maps.

10 Likes
6

great you internationalized cartes.app recently!

Casual observers aren’t aware that open map stacks often share building blocks in the backend.

7

Thanks for posting in this forum :slight_smile:

I am really happy to read once again that it seems Murena /e/OS has the right vision:

Developing a user friendly map app, that first offers online maps in a native app so that a standard user has a map working without any configuration need.

The mainstream audience that Murena targets is actually what brings value to the overall project as we end up with usable tools and still keep the freedom to install more specialised apps if one want (Osmand for example).

Coming back to Murena Maps, they already said that once the basic online map function is working, they will work on adding live traffic data (it is definitely a domain from which the whole OSM ecosystem will benefit, hope we will see sharing of the servers, cost and code base) as well as offline maps (very much like what Magic Earth is offering today).

As a personal remark, those map apps are quite useful in our daily lifes and because of that and the lack of alternative (except for magic earth today) we sometimes are forced to use Google products, therefore I would be willing to pay a subscription to pay for the servers’ maintenance if it gets me a map service that does not collect my position.

8

Yes, traffic data is a good examples of winner-takes-all, the winner today is Google-Waze.

The other example is place reviews. On cartes.app we’ve built it using Atproto (Bluesky/Eurosky). Murena maps could share our place lexicon to publish in the same DB, using the same JSON schema (lexicon).

That way, it wouldn’t matter the app used, place reviews would be shared.

2 Likes
9

Great Idea! @Manoj could you share it with the murena maps Dev Team? All feeding one OSM DB would lead to More reviews and less dev work in comparison to New review/rating model :slight_smile:

10

Aren’t the dev reading their own forum ?

11

No, they are deving, hopefully. Reading this forum is potentially very time-consuming. (I know.)
For organised interaction with them there’s Report an issue.

12

Hello again,

Thx for this vert rich discussion.

I totally understand that the goal of cartes.app and Murena are not the same, but as mentioned in the discussions I was mainly wondering about the work which is done in both side and could have serve the to project. And as you mentioned, if one brick of one project could serve another it should’nt be underestimated.

Murena is focusing on the development of an app which is not a web app. Nevertheless I wonder If the next logical step would not be having a web app in the ecosystem of murena. Gmaps is a big brick of the G ecosystem so… and at this point wouldn’t it be possible to integrate a redirection to cartes.app with maybe some modification of the UI (like what have been done for Murena search which redirect through Qwant)?

13

Thanks @laem for reaching out and sharing your comments and inputs. Have passed it on to the development team.
As mentioned in @laem’s post we are looking for a native app to work seamlessly with /e/OS.
Will share the teams’ comments and response.

1 Like
14

Yes it’s possible, it’s what we’ve proposed to Comaps. Because recoding a Web UI is about the same overall complexity as building the native app itself (well I’ve never tried to code a native map app, I’ve heard that app store problems are very annoying).

It would be quite easy to change colors and icon of cartes.app to serve as a third-party instance. Server-side, our stack is quite easy to deploy since we’ve scripted everything with Ansible.

Changing other things by diverging from the base branch would be a lot harder though, because we’re moving fast and the Web app is obviously a monolith.

We already redirect our users to Comaps for navigation. Tell us if you’re interested (the app is not officially available yet, is it ?).

15

This, I think, is the latest from the murena-maps team Murena Maps: questions and answers about our open source mapping app.

Test versions are downloadable but we are expecting news of the debut at /e/OS 4.0 live launch - Murena - deGoogled phones and services

16

Also, here Milestones · e · GitLab I can currently see a milestone for an alpha version expiring on June 11th.

17

As a GIS specialist, cartes.app is appealing, nice map theme, congratulations. Unfortunately the website is clearly oriented for french users (near realtime traffic is only for france, I recognize that is difficult to handle such services but it requires a lot of efforts to provide worldwide data like Google or mozilla have)

Some of cardinal map (renamed murena map on the last versions) Ui are counter intuitive and parameters are ok but not for all users … especially pelias and valhalla paramètres.

As Eu will move to qwant from google which system of reviews users will have ? I think this question will come in the next days … may be if all align to the same reviews system it will be a sustainable approach.

I mainly use comaps as it provides offline maps and I don’t always have network where I live, in that situation cartes.app will not work … Building a native app in my opinion is not durable as the next step is probably not to stay with android but to use the fisrt vision of Gael Duval a linux phone like Jolla phone.

18

Yes, that was our first aim : grow in France and prove an interest. Now we’re scaling to Europe.

We don’t have real-time traffic for cars though, not even for France, maybe you’re reffering to live transit times ? We’ve started integrating them for France on our server, and relying on Transitous for the rest of the world, so this depends on their GTFS-RT coverage.

cartes.app will not work

Yes but we’re planning on releasing an offline version soon. It’s theoritically possible as a PWA, let’s see how hard it is :slight_smile:

Building a native app in my opinion is not durable as the next step is probably not to stay with android but to use the fisrt vision of Gael Duval a linux phone like Jolla phone.

Yes, we’re wary of the Android ecosystem that is being locked by Google more and more every semester. The Web is here to stay, even more if we show its power to users free from app stores :slight_smile:

Still, Android won’t disappear soon, so having great and free native map applications is key. The more people use OSM-based apps, the better its data quality will get. The main think missing right now is show-owners. If we get these to fill their opening times, links and hopefully pictures, we’re conquering whole new end-user markets.

3 Likes
19

Hello,

Another important aspect motivating the co-development of a web interface, is to be able to share locations with other people.

For me that’s a key problem of the major part of app today. In magic earth for exemple, if I want to share a place with a friend I can only share GPS coordinate…

So we need (in my opinion) to have a way to share infos with people that will not have murena maps on their phone. Creating link for G maps doesn’t seems to be a good solution… but to be able to share link that redirect to a web app seems to be a good alternative (Additionally the website can propose a download link to install murena maps on non /e/os phone…).

But that’s just my point of view :grin:

20

Yes, that’s why Comaps introduced this web app Salle associative Armorique | CoMaps

And Osmand OsmAnd Map

Both are sub-par, because it’s hard to recode a Web version. Osmand’s take isn’t even vectorial, and comaps has almost no features.

Of course it’s not that difficult to implement a complete Web interface, that’s why we’ve managed to build cartes.app :smiley: , but it’s still a 100 % focus for a whole dev team.

What makes me think that it’s an underestimated problem is to look at the search engines’ take on this. Kagi first went with Apple Maps but then switched to Mapbox. I wouldn’t want to see their $ bills, it’s probably huge. Kagi has no itinerary feature (!).

Qwant dropped their maps. It was great, all in-house, all open-source, kind of the ancestor of cartes.app, but management killed it because judged too expensive to develop.

Ecosia has no maps, despite its huge size for a european tech service.

2 Likes