As many of you may be aware, Microsoft will be discontinuing their SoundScape app in June. This news is undoubtedly a devastating blow, as there are currently no apps available that can match its accuracy or provide the same level of positional audio and navigation.
Thankfully, Microsoft has released the source code for this app, which means I am considering taking on the task of reintroducing it to the App Store once it is discontinued. However, I will need the help of an experienced app developer to make this a reality.
I am willing to pay for this service, and if anyone can connect me with Apple developers who are interested in this project, I would greatly appreciate it. It is my sincere hope that we can preserve this valuable app for future generations.
I don't want to discourage you, but bringing that application back online is not an easy thing to do, nor is it cheap.
In addition to the costs of the developer account, you will need to assume the costs of servers and probably more than one third party service, it is not enough to clone the repository, modify four things and upload the application ...
if the app had a cost that could be maintained by a single person, I highly doubt Microsoft would pause its development and remove it from the App Store.
For one thing, what Microsoft is offering is not the complete sources. There are parts of the app that would have to be recreated.
On the other hand, the application uses cloud services such as Azure. Those services depend on the requests they receive. The costs may be too much for one person.
You would also have to pay the developer (or developers) who would not exactly charge you cheap, you have to put a lot of hours into that project.
I quote the original source and leave a link to the repository.
This open source project is not a turnkey equivalent of the Microsoft Soundscape product offering. The sources have been modified to remove branding and IP. References to the production resources were also altered. Further elements too specific to Microsoft's internal environmen were omitted.
To bring this up this open source project, you'll need:
iOS experience -- Apple developer account, experience with Swift, Xcode, AppStore processes, etc.
Cloud experience -- The services ran in the the Azure Cloud, though could be adapted to run elsewhere. In particular, note that no automation has been included to provision the required resources and services.
The core elements of the service eg. the OSM ingester via imposm3 and serving the ingested data as GeoJSON are provided and packaged as containers in svcs/data.
You can find the gitHub repository here: https://github.com/microsoft/soundscape
I applaud your efforts. It…
I applaud your efforts. It might well end up being a bigger project than you first envisage, most projects are, but it's certainly a worth while one and I think there is enough of a user base to make it feasible. The fact it was free and, any development costs you undertake will have to be remunerated, may be an issue. I'm not sure how many potential users will be willing to pay for it, especially after it has been free.
It would be interesting if someone such as be my eyes could take it on and integrate it into an all in one assistive app for the blind, both micro and macro navigation etc... Not that I wish to undermine your intent.
Very much hope it does find new life. I simply enjoy wearing AirPods and, whilst in the back of a car, understanding the city around me. Such a shame they discontinued it.
I would love to see the Soundscape app continued in some form, but you need to be aware going in to this that doing so is likely to be a significant financial commitment.
An experienced developer with the specific knowledge and expertise to pull this together could easily cost you in excess of $10000 simply to get you an initial release. You would then need to have the developer on some form of retainer to cover any bug fixes and updates needed to maintain compatibility with future OS releases. You should expect to budget at least $5000 annually for this. These figures may seem steep, but I suspect that I might actually be under-estimating the level of expertise and developer time that this project would need.
Unless you are planning on providing support yourself, you will need to source an agency to provide user support.
You would also want to obtain legal advice to make sure that you are satisfying any requirements or constraints contained in the licenses of any software that the app uses. You also need to ensure that you are legally protected from any liability from people's use of the software. Because of the nature of the app, it's possible that this might require some form of paid cover.
It's likely that the backend infrastructure needed by the app could also be a significant expense. I don't know enough about the specifics here to suggest what that might be, but I am confident that it won't be the cheap hosting that you typically find advertised for a few dollars a month.
My apologies if you had already considered and accepted all of the above. I certainly don't want to discourage you from pursuing this goal, as I wish you all success, but I do want to make sure that you go in to this with a clear indication of the potential financial commitment that might be required.
I wish you all the success, and please keep us updated!
This is a really good idea, and wouldn't take all that much. You would need a $20 a month virtual server to host the server assets.
So the financial burden doesn't fall on one person charge a small $5 per year subscription just to pay for the hosting.
The source isn't all that big and doesn't require all that much space for hosting. The data would be pretty minimal as it mostly sends longitude and latitude.
There could also be a charge depending on the size of the map data.
lets assume that the map data is 100 GB. On the higher end you might have to pay $0.10 per gig. So, add another $10 a month to host maps and supporting data.
Furthermore you could probably do a go fund me to raise a few thousand dollars to pay someone for their time.
Keep doing what your doing, and don't let the nay sayers discourage you.
Its not unreasonable and I think the community would chip in a small amount to keep it going.
In response to the previous comments, I have no real knowledge of app development i just would be helping monetarily. However i certainly didn’t see that steep of a budget being needed. That all being said, do we know of any developers who may be interest in helping or how to find them?
The first thing I did when Soundscapes became open source was to download the source and import it in to Xcode. I have been developing apps for over 10 years, so I wanted to see what it would take to get this back in the App Store.
I was able to get the app to compile and run on my phone and I was very excited. I made it through the tutorial, and then the app crashed. I checked the debug information, and it is at the point that you need to connect to a backend server.
I checked out the rest of the project, and there are discussions of what is needed for the server to work, but this code is not included with the project as it is proprietary code.
WhileI can completely get the app to work, I do not have the understanding to set up or maintain servers that use GeoJSON or geolocation database structures. I really think we could get this thing off the ground if we could find someone who can help us set this up.
I personally would very willing to volunteer time to put this app back in the app store, and I would not mind hosting the app in the store since I already have apps in the store.
If anyone knows anyone who can help me with this then I would love to have a discussion as I really do not want to see this app completely disappear.
There is a petition to try to stop Microsoft pulling this completely. Maybe too late but i have signed anyway in case. https://www.change.org/p/don-t-let-microsoft-take-away-navigation-for-the-blind?cs_tk=Ah2FH4ASp8ouQmadSGQAAXicyyvNyQEABF8BvDdzmblcfikpk28kYQnH3ps%3D&utm_campaign=861c340b21234870ba2277f547d987d0&utm_content=initial_v0_2_0&utm_medium=email&utm_source=recruit_sign_digest&utm_term=cs
If you make this app workb again is there any possibility that it could come to android?
I’m just warning you, since the original soundscape works of Microsoft online servers, you’ll need to try and make a source code for that server, or make your own online server
So sounds like we need to find someone who can host the data on their server correct? if so where can we look to find them?
Would it be possible to integrate Soundscape with Google Maps data?
In response to the last comment, I’m not sure we’ve got a find someone who has the knowledge to work with and get these servers running and the integration process.
I don’t know anything about coding, get me nearest server and I’
I am very smart when it comes to my technology, however, so I could try and keep the servers up and running, at least temporarily until we can find a permanent solution.
Hello, I unfortunately think very differently. I put a cross on it, that app will never exist again.
This is because not only would there be costs, but it is the least of the problems. The real problem is that the app needs to constantly communicate with servers that provide information about maps, shops, etc.
I don't know how many thousands of api calls it would have to handle at the same time if multiple people accessed the app. Also, map data is not free. To be clear, it is forbidden to use google maps.
Sure you can use open street maps and probably Apple maps too, but please note that this app was run by a giant like Microsoft, not an independent developer.
I'll be happy to be wrong, but in my opinion it's a waste of time.
Rather, Microsoft should be persuaded to backtrack on its decisions…
Does anyone even know what the cloud part does yet>
Assuming we did manage an Applevis version of Soundscape....
Obviously we need some map data from somewhere. But I'm not sure we would necessarily want to hots that on our own servers. I think this is more likely something we would connect to a 3rd party for.
So, other than that, what exactly do we need the cloud for?
Our own user data could stay local to start with. Maybe we can use iCloud storage or something and maybe then just as a backup.
But before buying a server farm somewhere, might be worth figuring out what's needed. Physical servers aren't the only option.
Has anybody tried reaching out to Microsoft to see if there’s a way we can keep it alive at through them
@Michael -- if you have the developer account to host apps in the store, please chime in on the Github issues, as this is a need that is not yet claimed: https://github.com/microsoft/soundscape/issues/12#issuecomment-1509915900
The backend is still an open question. The current hosting is terabytes of OpenStreetMap data on some pretty industrial-grade infrastructure (Kubernetes on Azure) -- probably overkill for the size of the user base. It would require some engineering to come up with something simpler, but there are enough smart and interested minds that I expect it can be done.
Keep Soundscape online in some form
I would also love to help with this as this app is amazing but I am currently at school so sadly don't have much time to code any of the stuff that needs re-coding. In a car the best thing to get your bearings used to be just scanning the nearby WiFi networks and working it out that way (on familiar routes I could instantly tell where I was), but Soundscape took this to a whole new level as I didn't have to always refresh the WiFi list and try and remember that "WiFi-Network-ABC" was halfway along a certain route (although I quite enjoyed doing that as I like networks anyway). The ability to share a beacon for another blind user to find you was also invaluable.
I think that keeping it going in some shape or form is much much better than nothing (although it would be better to try and get an organisation like Guide Dogs UK to take it over). So for version 0.1 you could take out all the server stuff, give up on hearing shop names for the moment (don't get me wrong, I love this too but its just too much to do in one go), and purely make it about setting markers yourself which you could use as beacons and use a clever URL scheme (e.g. soundscape://marker?name=x&lat=0.0&lon=0.0) to share them between iPhones without a server in between. This would still require some work, and this is assuming all the beacon code was released by Microsoft, but it would be better than just letting it all disappear as the beacon feature is not used in any other app and is great. You could then expand to a simple search for beacons (which wouldn't be as hard as downloading the entire map area around you), and then progress to full functionality after everyone's had time to get it all working.
What do you think?
Do also sign the petition if we can as the more people the better and I think we all agree it would be better if we didn't have to do all this, but if we have to and its the only option I'm 100% up for it.
If you have a computer, if you know cutting, anything else
Do you know how to use service, and if you have a computer and if you know some kind of bit of programming and coding, including the UI element, and the server identifies, and if you can get through Apple review process, you should be good to go.
But you need a developer account.
Hi all -- I hope I've moved the ball forward on this effort, by writing code for a replacement backend server for Soundscape. It can use public and free-to-use Overpass servers to get OpenStreetMap data, and convert it to the format that Soundscape expects. https://github.com/steinbro/overscape
To make this usable in the app, the backend server URL would need to be a configurable setting. Right now, it's hard-coded to a Microsoft domain that goes dark next month. If an iOS developer can compile that minimal change and get a new release in the app store, we just might be able to keep the lights on.