So basically a mobile hotspot with a webrtc webpage on which your headset microphone gets relayed? Participants can scan the QR-Code to join the hotspot WLAN and get directly redirected to the webrtc session

Oh boy, if ever there was a time to invoke the Dropbox comment!

Seriously though, I think it's an under-explored area. Several years ago[^] I thought it could be a fun way to do an ephemeral social networking type thing on the tube; then something a bit similar ('YikYak'?) became briefly popular (in the UK at least) but just using GPS or whatever for location, still with a central server.

I think there's plenty of interesting things you could do ephemerally, P2P over Bluetooth or mobile WLAN.

([^] edit: that's not some weird claim to fame, I mean it in the sense of I've long thought this should be good, but then, I've not bothered to do anything about it, and neither it seems has anybody else, so...)

Knowing Dropbox is like 10M lines of code now always teaches me the importance of not pretending things can in fact be built in a weekend.

The conceit that I could glue a bunch of libraries together in a 48 period to get an MVP-level of program that is nowhere near as polished as the collective output of teams of people, and would fall over at anything resembling load, is not the problem.

It's the implicit derision, that eg Twitter is inherently worthless because I could glue a few libraries together to a database over a weekend, and have a simplistic system where users who are able to broadcast short messages to other users, that is so odious about the "build it in a weekend" mindset. It denigrates the very real, but largely invisible effort it takes to scale a system up to that level. For any of us who have been involved with scaling a system up, it's invalidating to have your work invalidated like that. A Twitter or Dropbox clone built in 48 hours isn't going to handle the scale of actual Dropbox, nor will it be as resilient as actual Twitter in the face of eg a whole AWS region going down.

Build something awesome this weekend! (or even just something useful; it doesn't even have to be software) Just don't shit on someone else's hard work, even (especially!) if you don't get it. That Dropbox, reduced to it's basics is "just" a filesync client, and isn't actually the most technology fascinating project when described at that level, is an reflection of the oversimplification of the problem at hand, and not the inherent worthiness of solving that problem. The social aspects of what a program can be used for goes wholly underestimated and misunderstood a lot of the time. No one could deny the effect on the world that Twitter has had (though we don't have to agree on if the net effect was positive or negative), but many people still don't see the value of a product that isn't a revolutionary new technical product, like terabit wifi, but that's a reflection on their lack of imagination on their part, and not a poor reflection on Twitter as a business idea.

How could you possible get that connection going? I'm pointing out how someone could potentially implement the product the parent wanted....... The easy to use parallel like Dropbox was to network shares doesn't exist.

That sounds even better than what I was thinking of because you wouldn’t need people to download an app. Also, ESPN and Sky could have a version of that which it could license to broadcast the audio of a football game at bars or pubs

This is actually a really good idea. The hotspot could just locally serve the client and server side. Some pre connected devices could be distributed also for the elderly or the disabled.

Then you could package it with a "creation studio" to create your own version for tourguides, museums, etc. Resell some hardware also as per requirements. You've got yourself a nice little business that seems useful for the business and client.

We already have this. Just have the professor start a Zoom call on their phone and have everyone else join it using their phones while wearing headphones and their microphones muted.

I think the interesting bit is tying a stream to a (moving) location, and not having to scan QR codes, type in codes, or URLs to join.

Brilliant idea. We have thought of that before, when in a museum, where you know there is so much more history and information than just a label. Surprised no one had done it yet.

Could we geocode the lecture-snippets, so the statements persist at location? That way, one could advance at ones own pace. Or map it to a route of ones own design, while keeping the pace..

I recall projects that did something like this. Besides museums and historic neighborhoods, one memorable application was at a graveyard. Reenactors read letters and diary entries written by the deceased. The closer you were to the writer's tombstone, the louder their audio. From far away all you heard was a combination of barely audible voices. I believe the relevant terms are "audio walks" or "phonoscapes".

I have created something similar for a museum. I placed bluetooth beacons behind the exhibits, and when visitors approached them the app automatically switched to the "page" for this exhibit. The app was able to display texts, videos, images and audio files.

Bloomberg Connects[1] is an attempt to do something like this. An institution connects their CMS and then visitors can access additional text, audio, video, images using triggers in the museum space.

However, to my knowledge it does not support live synchronous audio.

[1] https://play.google.com/store/apps/details?id=org.bloomberg....

Alcatraz almost has this!

Due to the pandemic they made an app where you can follow along and listen to different parts of the tour

It’s not very “smart” I.e. not location aware to start the next prompt but honestly much better than those clunky audio tour boxes

The museum would lose revenue because you wouldn’t have to pay to rent their audio devices.

I am a software dev for a company that builds museum exhibits. Many museums these days have bespoke web or downloadable apps that serve audio tours, sometimes activated based on location through QR codes, computer vision, beacons (iBeacons, Eddystone beacons), or RFID scannable bracelets given to every visitor.

Could you please pitch this app as a more scalably-profitable revenue stream for your employer?

It's not really "an app". Every museum wants a different MVP, and I just listed some of the common tech involved in most of them. We don't have a turnkey product for this, we build out a custom thing every time, because clients come to us specifically for one-of-a-kind technical implementations.

Aruba Networks does this with their Meridian division.

You could pay for access to the audio stream just as well.

A lot of tour buses catering to speakers of foreign languages do exactly this.

Why not a 'tour guide portable FM transmitter' (less than $100) and a bunch of dollar store FM radios?

Some, many? most? cell phones have a builtin FM receiver so in many cases no additional HW should be needed

[0] https://duckduckgo.com/?q=cell+phone+FM+tuner

FM audio on yoru phone usually relied on the headphone cable being used as an antenna to receive the FM signal. With the demise of wired headphones, especially for high-end phones, this is unlikely to be viable.

that is insidious. Hopefully there is an alternative dangling a charging cable or someone coming up with an antenna case that can boost the signal. So far I'm good with just not buying one like that.

Because few people already carry FM store radios in their pocket, so relying on them requires solving a logistical problem.

True, but the FM solution could really be low latency, unlike the digital rube goldberg machine equivalent.

It will be more reliable than any solution based on either (i) a battery-powered web server, or (ii) mobile internet.

And latency will be constant and near-zero.

I suspect that's why a couple of the use cases you mentioned (city tours, dance parties) are usually addressed with FM radio and not apps.

Perhaps there's a way to do this with a modern bluetooth stack. But none of the offline messaging apps I've seen (Bridgefy, Meshenger) support group audio.

What's wrong with the reliability of mobile internet? I'd expect that in the target environment (e.g. some building in the middle of a city) mobile internet would be like 100% reliable, that would be a problem only in remote wilderness areas.

The things I had in mind:

1. I sometimes participate in Zoom/Meet calls when walking around SF. Not often enough to have a representative sample, but I know I've had a non-zero number of 1-2 second drops.

2. Students are often on a budget, and may have data caps on their mobile plans. Running out of data towards the end of the month might be OK if you're mostly in your room or the library. But it might impede you from participating in a lecture if you're walking out of wifi range.

Could be done with WiFi direct on phones that support it.

The fact that WiFi didn't originally ship with (and in many cases still does not support) simple P2P with nearby devices is just unbelievably stupid. Apple still insists on their own stupid standard instead of the actual standard too, at least on iOS devices. (I think Macs can be coaxed into the standard mode, but Macs are real computers.)

I am lookng forward to SDR-capable hardware falling in price to the point that it becomes the default for most devices. At that point, not only do wireless protocols become field-upgradeable (including more exotic options like UWB), but decentralized mesh-network protocols become installable as well.

I think this is planned for a future version of bluetooth.

I imagine that a silent disco setup would provide this rather well. Might need one with a better form factor for walking outside, but the technology for broadcasting locally to headsets has existed for a while now. Museums likely have these deployed already.

The trouble is that you need to distribute headsets to people, hope they don’t get broken, and then collect them back. If there is both an android and iPhone app, then you can let people use their own earbuds. You’d merely need to tell people “open Soapbox and pick #gymnosperms. We’ll set out on the trail on 8 or so minutes”

And then you have to deal with people who don't have earbuds with them.

Distributing headsets is a lot more reliable

Until you think about people that have devices that don't use headphone ports, or don't have bluetooth on their phones (both, IIRC, are very small counts, but potentially a problem). Not to mention if you're sharing headsets that, unless they're all coming in plastic bags suggesting that they're new from the factory, everyone else's germs are all over them.

> headsets to people, hope they don’t get broken, and then collect them back

Many museums do this already for their self-guided audio tours, it's pretty old tech but it works super well.

Your comment struck me as odd but then I thought maybe you live in the US or a country like that with a language largely used by everyone. In France and Belgium and Europe it's really common for museums to offer audio guides. Which are either headset with a radio receiver that plays whatever the radio emitter is cycling or a headset with a small media player that plays tracks when you enter a room or switch a button.

You need zero-latency for this to be comfortable if participants will be in partial-hearing range a portion of the time.

An analog solution (e.g. "walkie talkie") would work well, though.

You could do a group call where everyone else is on mute.

Horrible idea. I want it silent while walking around solving a complicated problem. It's only partially the walking, it's also the silence which brings out the best solutions to think over. But a good idea is to walk along a very loud river, the noise also clears your brain.

Random talkers certainly not.

Could be accomplished with BLE broadcasting.

I've tried using bluetooth in a museum exhibition before and noise from wifi and other equipment rendered it pretty hopeless... but may have been the modules I was using...

Flawless at home/office. Useless on site

This app indicates that existing Android and iPhones can already act as receivers. Can you (or anyone more familiar with the tech) verify that is true?

https://stackoverflow.com/questions/27294649/i-want-ble-broa...

Receivers of BLE broadcast packets? Yes, that's how devices are discovered. It's just that you can also stuff those packets with a payload and transfer data (one way) without a connection.

Something like SoundSeeder (works via WiFi)

Kinda near-range walkie, nice. Bluetooth transmission should do, shouldn't it?

Is this an audio guide available at lots of museums?

I think Crowdcast is what you are looking for.

That would be short range FM-Radio.

Couldn’t you use Zoom for that?

Zoom isn't really setup for something like this. Ideally you want something that is one to many (broadcast), is low latency (since the participants would be able to hear the professor talking both in person and over the app), and is reliable.

Conference call

Latency on that would be even worse than a zoom call.

...that's AM/FM etc radio.

Yes. I think I’ve even seen this implemented.

Certainly some drive-in theaters in North America worked like this. There would be signs telling you to tune your car radio to such-and-such frequency to hear the film audio. Here’s a picture of one: https://c2.staticflickr.com/4/3186/2718300357_ec1b9f3f45_z.j...

These still exist :)

https://kendadrivein.com/faqs

Yeah, this is totally doable with walkie-talkies. The biggest drawbacks I can identify would be audio quality and interference.

Within a distance you can walk, it should be pretty easy to find a band with minimum interference even in a crowded metro. Also, quality won't be as much of an issue as range. There are pretty low limits on how much power you can use on the FM bands without a license.

Source: I drove all over a crowded metro in the early 2000s with one of those fm transmitters for playing audiobooks to my shitty car stereo. 3 station presets was enough for the whole area.

You just invented poorly implemented radio.