Show HN: Popcorn Messaging – Anonymously chat with people within 1 mile | Hacker News Reader

Show HN: Popcorn Messaging – Anonymously chat with people within 1 mile

appsto.re

126 pointsdrum130 comments

A lot of the people whom I met in Vancouver via Twitter I actually met via an app called Twinkle, which was essentially a wrapper around Twitter that added this sort of geographically-located content, and which was released shortly after the iPhone 3G and iOS 2.0 added location-based functionality (but long before Twitter did).

Basically, Twinkle could work on its own or with a Twitter account. When you posted a message, it would go to Twinkle's servers along with geolocation information, and then Twinkle would post it to Twitter in turn. Other Twinkle users could see your post via Twinkle's servers, as well as your actual Twitter feed. It was a fantastic way to discover new local users, and you could scale the radius up or down (from I think 1km to 10km).

I know one person in particular I ended up seeing in my stream a lot because we both lived in one area and worked in another area, and lots of others just worked in the same area as I did. You could easily Twitter-follow them, which let you take those relationships out of Twinkle.

For someone who was completely new to the city and the community, it was a great way to start getting connected to people in the area.

Wow… I had no idea people still remembered Twinkle. I was one of the engineers at Tapulous, back in the day, and worked briefly on Twinkle.

That service was actually a huge disaster for us. The users were mostly tweens, and we had a huge Chartroulette-esque problem with child porn, where users would expose themselves through the service. Our community management team is probably scarred for life. Our engineers also hated it, because its backend was horrendous and we had a lot of availability issues; the service would go down constantly.

The very last thing I did at Tapulous / Disney (on my last day) was kill off Twinkle, shutting down the servers. Everybody was quite happy to see it gone.

I definitely remember Twinkle! I used it daily.

It would be cool if the "radius" automatically scaled to ensure there are not too many or too few people in the chat. This could be accomplished with a Voronoi diagram.

hmm, the 1 mile radius is important to me. psychologically, knowing that the people you're talking to are within walking distance is exciting

Less exciting if there's no one using the app within that radius though, no?

there are a million and one apps that offer the ability to chat with people. if it's a concern people can use one of those

Wow - sounds kinda hostile and doesn't really address the original point (sorry if I'm misreading your response). What's so special about the exact mile radius that couldn't be, say, a bit more flexible if conditions aren't perfect? Couldn't the app expand that to a mile and a half if no-one's present within a mile? Etc. I think that's the original point that was being made.

sorry if it sounded hostile, i appreciated the comment! my point was that the integrity of the app relies on a fixed radius. im not opposed to increasing the fixed radius initially while there arent a lot of users. however expanding the radius according to nearby users makes the location aspect less relevant. if the nearest user is 20 miles away than they should probably use another app.

In rural areas in North America, you might expect people to be much further away, but since everyone has to drive for everything anyways, it's not that big of a deal. I think it retains much of the excitement of being within "walking distance" if you can drive 10 minutes to meet someone.

Just wanted to chime in and say I get you, man. "Chat with people within 1 mile" is just a better thing than "Chat with people within a radius adjusted based on how many users are online nearby."

Yep, that's fair enough. And you certainly don't have to make an app that's useful for everyone everywhere. Still, getting it to catch on will be hard enough even if people can find others to chat with, it makes sense to do what you can to help with that. How about keeping the tagline the same, and keeping it at 1 mile as long as there's someone there to chat with. But if not, you could prompt the user: There's no one else online within a mile. Want to look a bit further away? Or something along those lines.

I think that's a good point. If the fixed radius is an important feature you keep it. It sounds like it's the reason the app exists, and to change it would simply turn the app into something it's not.

Maybe it's a bit like Twitter's 140 character limit, ignoring the SMS size that inspired it. Sure, people might want to tweet longer messages, but tough - this is how it is and probably one reason it worked.

People have different walking distances. Nothing wrong with a user setting ...

I think you are right about wanting to know that people you are talking to you are nearby being important. I also think if you allowed a larger radius option in the form of 10 miles or whatever, you would NEVER be able to remove it without complaints. That said, I'd consider promoting the app in a specific large city.

Alternatively, a neighborhood/borough mode might be an interesting adaption that would let you shrink as you scale. In New York, maybe you only talk to people who are also in Brooklyn. Or just Nearby parts of Brooklyn (mash up with that app that allows you to see how far you can get in a time period). In North Dakota, you might have a much much bigger region.

This would also avoid the issue I have playing a game with a geolocation based chat on android, wherein I am in the Financial district in SF, and see "nearby" people in Oakland, when I would have to cross the bridge/take bart there.

this..

I'm curious about how this actually works, geometrically. Example:

A <----> B <----> C

Those are three people. Imagine each segment is 1 mile long. So A can see B; B can see C; but A can't see C. If B and C are having a conversation, does A see B talking to someone who effectively does not exist?

does A see B talking to someone who effectively does not exist?

No, why would you want them to see who else they're talking to? What messaging application allows you to see who other people are privately conversing with??? Not a popular one, I assure you that.

This is likely referring to public conversation.

I'm not sure who, but clearly one of us has not understood the purpose of the app under discussion here. My impression is this facilitates public, pseudo-anonymous conversation with anyone in a 1 mile radius of yourself.

No, I assumed it just arranges private chats with people nearby, like Chatroulette with geo features.

a solution could be the geohash

i think you might be right... hadn't thought of that

I experienced this problem first hand this morning...

I'm already seeing this happen on a consistent basis, it's making the chat basically unreadable. Seems like a major oversight.

turns out you're right. anybody have suggestions on how to fix this?

Why not group people into free-form clusters based on user density? So, if you have 5 people that are effectively in the same building, and 3 people that are outside of it in various directions, group those 3 into the 5. This has the effect of forming virtual chat rooms that come and go.

Not sure what you're using on the backend, but if it's PostGIS... here you go: http://gis.stackexchange.com/questions/11567/spatial-cluster...

How about a 1 mile (square) zone, rather than 1 mile away from you? So you just cut everything up into mile-square slices, and you can talk with everyone in your slice... Concept is basically the same, but should overcome that issue.

Make A see only see C's reponses to B.

Oh, so I am not alone in expecting Chat and IM be a different concept. I would actually like seeing a public Chat among everyone, as it was in good 2000's.

I almost created one of these, except for news. Hyper-local news is largely missing - I was going to mash up Twitter, Instagram, and in-app updates to help people get a sense of what's happening immediately around them.

This is actually a great idea to solve this problem though.

The news use-case is for city life. When you hear a gunshot or a car accident occurs on your street or you see smoke off in the distance or you hear music playing nearby or you see a drove of cop cars driving by - how can you find out more about that?

News doesn't have a hyper local channel, but it could.

This app points at a bigger market - the "directly nearby" market - that really has only been fully tapped by mapping/location-sourcing applications.

I would pay good money for this - I'm always trying to figure that sort of thing out and it always ends up as a bunch of random googling or trying to remember to check the news in a few hours.

Check out http://app.bg/. It shows local Instagrams, Foursquare tips and tweets. Also http://www.nearbytweets.com

thanks! one of our biggest goals was to maintain privacy by only requiring a username to sign up.

And we made InstaBAM for finding instagram users and photos near you (or anywhere on a map) https://itunes.apple.com/gb/app/instabam!-explore-instagram/...

So, talked with some folks about an idea similar to this a few months ago.

As the conversation evolved, they came up with "tagging" a given chat room to denote what was in it, and letting people choose their own names...some kind of "handle" on the identified "channel".

The proposed messaging fabric would use servers hosted in the cloud, each functioning as sort of a repeater to their client mobile devices.

A sort of internet-relayed chat, if you will.

There is nothing new under the sun.

EDIT:

The location functionality of this could be bodged together really quickly with the geo pg extensions, I'd bet.

After seeing the criticisms of Dropbox[1] on Hacker News, I'm much more analytical of the stereotypical "critical top post(s)" when introducing a new project. Getting candid, constructive critical feedback is certainly one of the values of posting on HN, but criticising that a product or service is just a combination of known technologies is an incredibly weak criticism: it could be applied to any innovation.

Here's my litmus test for whether something has already been done: if I asked ten random people to solve the problem this thing is trying to solve, could they name a product that solves it?

Ex:

* "How would you have a folder on your computer that stays synchronized across multiple computers and is accessible online?" -> Dropbox or Google Drive (not an acceptable answer: "I'd set up rsync to run as a daemon syncing to a VPS on each computer")

* "How would you send public messages to people you didn't know in close proximity?" -> Popcorn Messaging?

1. https://news.ycombinator.com/item?id=8863

Don't get me wrong, it's a nifty little app.

I'm just pointing out that I heard an idea a few months ago that was nearly identical. Our advice to the folks was "Yeah, this is cool--go code it!".

These toy campus apps are really great for building experience and learning; that said, they're still just that: toy campus apps.

The obligatory "but but but facebook" rebuttal is that it was a matter of timing more than anything else.

Don't you think calling this a "little app" and "toy campus app" is a little bit condescending? And that dismissing it as a simple rehashing of other ideas is not actually constructive? How many successful new products/services in the last decade aren't just rehashes of existing ideas?

I'd say kudos to the author for actually rolling up your sleeves and making this idea happen; it's way harder and more admirable IMO to do that than it is to be an armchair critic or armchair ideationist.

Would you like me to give you some Ruby code that would emulate this functionality? I am fairly sure the size of the program would be less than 300 LOC, tops.

That is indeed a "little app". The 30-something line regex parser Pike wrote (http://www.cs.princeton.edu/courses/archive/spr09/cos333/bea...) ? Also a "little app".

There is no shame in solving a small problem, solving it well, and solving it concisely.

As far as being a "rehashing of other ideas"--how many hackathons have you been to on college campuses, and/or how many students doing mobile dev stuff have you chatted with, and/or how many people with a great idea for a local messaging app have you met?

This is not, will not, and never has been a novel idea. That doesn't make it worthless, that doesn't make it pointless, but it also doesn't make it new.

As for the "toy campus app": that's what this is...a toy application (small, concise, single-function) targeting primarily college kids (see the comparisons in this very thread to JuicyCampus).

I honestly don't understand why you're getting so defensive just because I'm not getting down on my hands and knees to praise the obvious genius and ingenuity and novelty that went into this (this is sarcasm, by the way).

I can respect it as a work on its own merits, but I will not afford it more than that.

A recap of your pleasant original comment:

    "Don't get me wrong, it's a nifty little app. 
     You probably learned a lot! That said, it's still 
     just a toy campus app."

A morsel of value from the world squandered.

I'll bite your offer for you to mock it up in Ruby real quick. The problem with the "it's easy and I could do it" dismissal is that, even though you could, you didn't.

Maybe you can team up with this guy[1]. You can probably even reuse his repo[2].

[1]: https://news.ycombinator.com/item?id=678501 [2]: http://code.google.com/p/hackerexchange/

Hey, [1] was only 1584 days ago. [2]'s looking not too bad for only 225 weekends in.

https://gist.github.com/crertel/7244841

There you go--left out the boring HTML scaffolding (original chat example modified from here: https://github.com/igrigorik/em-websocket).

EDIT: It's super MVP, but the concept is there.

A very, very simple addition would have the server echo back to the clients the lat/long pairs in question to dynamically update, say, a Google Maps window.

For real usage, I'd use a PostGIS query--something like this: http://unserializableone.blogspot.com/2007/02/using-postgis-... .

EDIT2:

Looks like we're both Texans--if you're ever in Houston, I'd be glad to grab a beer with you.

You may have heard an identical idea a few months ago, but I've never heard of it.

I have heard of Popcorn.

I am imaging you in the Amazon meeting a previously uncontacted tribe and asking, "How could you eat today if you didnt feel like hunting and gathering?" When the elders answer "we dont know" you would go back to the lab and try to invent Cheerios.

Just because people dont know about something doesnt mean that it doesnt exist. Just because something exists doesnt mean that one shouldnt work solving the same problem.

Not sure what your point is here, but I'd reinforce that in your Amazon example that would be perfectly good justification to create a food product directed towards Amazonians (or perhaps just a Cheerios distribution network in the Amazon?).

i see a huge difference between location based chat and subject based chat.

The market problem is a very real reality. Popcorn is pretty simple, you need to invite you friends - but do you have to do that for every area to unlock that area?

We've been working on the same problem for a few months now but haven't quick cracked the market issue. Our app, racut, has a similar feature called Shout where you can say something and it appears over a 6 mile radius.

In addition to Shout you can also start a private conversation, or have a group conversation outside of the Shout stream.

It's good to see others entering the space.

From my own research of the space while working on near.im, it seems like the companies that have done the best are the ones who make agreements with events (i.e. a networking app for people at a conference.) This is great because it has a clear source of monetization and has the added bonus of having the conference advertise that the app exists so that people might actually use it.

Seems logical to partner with a physical event or space, because it helps to provide some of that social proof that people often crave to use something.

I think of it like a restaurant. You're probably going to eat at the restaurant that is busy, not the one that is empty.

where's the link to your app?

http://racut.com

Currently only available for iOS. The version in the store is a hybrid, we're working on an objective-C build that should be done in the near future.

It's a fun project.

just downloaded. i can't seem to hit "register" for a new user after i fill in my information.

It looks like button is slightly jacked up. Tap the register text. Thanks for the 411.

I also did a little adventure into local-anon-messaging, but my solution was to use wireless access points and udp for text, relying on a server only for image share. In retrospect it is primitive and ugly as sin, but I was excited about it so I put it on the app store:

https://play.google.com/store/apps/details?id=com.constantiz...

The only problem? wireless routers that block udp ;)

JuicyCampus for the mobile age. And this time administrators banning it from campus wi-fi won't do any good, with everyone on 3G and LTE.

http://en.wikipedia.org/wiki/JuicyCampus

That´s very exciting.

Introducing restrictions like Twitter with 140 characters is very powerful.

How to get users apart from PR? Spend $500 of Facebook ads per city, that gives you around 300 users. Done. Prove for one city raise Angel Money and do it for other cities.

You have something really powerful in your hands, however it might take a while until you really find what it is exactly. Take your time to find it. In terms of what features to add, as you read in many blog posts that the most important thing about finding new features is to learn how to say no. Your app needs to do one thing really well and only that one thing.

My gut tells me that there needs to be an upvote mechanism or an ability to communicate with individual users. However, then it would be whisper already. But something in that direction.

Good luck!

http://near.im/ is a mobile web implementation (in Meteor) of the same concept that I work on occasionally, although the frontend needs to be redone. No longer requires a Facebook login.

Really like this idea. There's been a deficit of anonymous messaging services in the college market since JuicyCampus went down.

I remember making an app similar to this a few years ago. Would automatically create a room if one did not exist, scale-able search range, anonymous users with an option of providing a name, all relied on a jaber server hosted in a micro aws instance. Was a fun project, but the critical mass factor makes things like this very hard to create.

If anyone's curious, github source:

https://github.com/quackware/Crowdsource

In the early 2000s there was a device/toy called the Cybiko (http://en.wikipedia.org/wiki/Cybiko) which offered similar functionality. It was amazing in theory, however, adoption was never very high so it never took off.

There were also something known as Citizen-Band radios, but that was a long time before all this computer nonsense.

yea that thing is ugly as sin

My location isn't unlocked yet, but this has the ingredients for brilliance. Marries the desire for privacy with the desire to feel connected to a local community... without being Whisper.

A great example of digital technology facilitating in person communication.

Look forward to playing around with it.

hey thanks a lot! thats pretty much the same way i look at

The chicken and egg problem can be pretty frustrating with an app like this. There needs to be a critical mass of users in a certain area for it to be reliable, which is easier in some places than others. Best of luck though.

One option for a location-aware system like this: the chatroom-defining radius could be adjusted elastically to make it so there are always enough people to chat with, even if the nearest is many miles away.

true! right now im targeting college freshman. if it catches on in the dorms i think itll work.

How will I know my area is unlocked unless I'm constantly checking the app?

Also, does it take -just two- people to unlock? I'm guessing so because there is no way anyone within a one mile radius also has this app.

yep, just two! i just pushed an update that includes push notifications. you might need to log out and back in, but it should show the prompt. from there im coding in a notification for when the area opens

Just logged out and logged back in, but no alert view for receiving notifications. Do I need to create a new account?

dang, friend reported the same issue. not sure why they're not going.

Also I logged out, killed the app, and went back in, but to no avail.

This app clearly reminds me of the device John McAfee promised us: http://www.theverge.com/2013/9/30/4786008/john-mcafee-d-cent...

"the device [...] has a range of around three blocks. Everyone in those three blocks can then communicate with each other and that will obviously change as users move in and out of a local area."

Will definitively try to maybe see people from HN in Paris.

I'm kinda sad I didn't capitalize on this idea when I had it awhile back. I may still. I really like radius-based anonymous messaging.

Good to see the idea getting some attention anyway!

dave? pretty sure i met you a few times in SF at the TRUE car meetups. thanks for the comment!

Great idea, nobody in my area yet so I cannot check the execution. (Barcelona, Europe)

Please, don't let this become something bad as happened with ChatRoulette when it became viral.

The grindr app is very popular in Barcelona and will let you chat with people nearby in seconds.

Why does it require iOS 7.0?

Because it's a new app, and actively doing work to support anything older than iOS 7 is a waste of time.

Unless the app uses some iOS 7.0 specific API, it makes sense to support at least 6.xx considering 7.xx is just over a month old.

The UI changes in iOS 7 make supporting both 6 and 7 more of a pain than most previous iOS releases. Certainly possible and not a huge deal, but it doesn't just come down to iOS 7 specific APIs.

Supporting iOS 6 is a non-trivial amount of work for many apps. Furthermore, you also have to test on iOS 6, and on more devices.

All that for a piece of the market that now amounts to 25% and is shrinking every minute.

To target a lower iOS can be as simple as selecting from a menu. For most apps, it really is as simple as changing the target to iOS 5.0.

You also need to do QA on the older version(s).

(That said, I personally appreciate allowing it to work with older versions since I haven't upgraded to iOS 7 yet and waited a long time to upgrade to iOS 6.)

You actually need to test your app on the lower deployment target.

There are quite a few crash bugs that are specific to, say iOS 6.0.1, but work fine in 6.0 and 6.0.2. And there are definite style and layout issues when using UIAppearanceProxy stuff on iOS 6 and iOS 5 (iOS 5 having a larger number of bugs related to this API).

Supporting both iOS 6 and iOS 5 is actually hard work and days, if not weeks of additional testing and bug reports.

Supporting iOS 7 and iOS 6 is much harder, since you have to have two different designs — you'll need separate code paths to style the app nicely on both versions.

This reminds me a lot of a webapp I built called Whisper:

http://zyu.me:5000/ https://github.com/tlhunter/whisper

You aren't limited to just one mile though, you can change the loudness, which affects the duration, and it's anonymous.

I really hope this takes off - it would be great to use at events!

Yobongo was trying to get this going a while back. For some reason they focused on specific cities (Austin, NYC, and SF) instead of using a generic 1-mile approach like this. I was always disappointed that they got aquirehired - their app died on the vine.

Agree. It could be very useful at events and a good addition to Twitter's nearby feature, because of the intention to actually meet people.

hey thanks a lot! the app was inspired because i needed to scalp tickets at a concert so events would be key

Creates a chat room with all the people around in it or it just randomly connects you to a single person?

Yeah that was the question I initially had - I'm assuming it's a chat room, which I like a lot better.

Assuming it's a one mile radius for everyone, making it a chat room the app would either need to intersect or union everyone's area... This would make the novelty of the app unusable or pointless, respectively.

Yeah, I wonder how that works actually. If radius A intersects with B and C but B and C don't intersect with each other, do B and C see A chatting to someone who they don't see?

would really like to know more about this as well.

That's how mobile radios have always worked... CB, FRS, GMRS, etc. Definitely not a new problem. Repeaters can mitigate the problem somewhat.

That's a really great question, my curiosity is also piqued.

Looks like there are "areas" to unlock.

A previous app (now defunct) named ChatSquare would group people together based on the nearest Foursquare location.

There is an app like this (I think - as I can't download this one) called WeChat which has a similar functionality. Not sure about the popularity in the US but in Asia it is huge and the "people nearby" feature is very fun.

I just got the username "James". This never happens.

If you can solve the "AB—BC" problem, then you'll have a really powerful application that'll easily get some investments, you just need to market it right.

what's the AB-BC problem?

As described here, https://news.ycombinator.com/item?id=6642919

Is this available on Android?

Yes, there are 100x android devices in places like India compared to iOS devices.

Screenshot has a typo "...with people within in a 1-mile..."

ahh, crap, thanks for pointing that out

Revenue possibilities for this app:

1) Give it to participants at conferences. You could partner with conference promoters and advertisers

2) Partner with baseball and football stadiums and "teams"

3) Partner with bands and gigs

Some things I've noticed:

- Push notifications are much needed.

- I logged out, but now I can't log back in.

- I'm unable to send more than one message at a time, but it's complaining that I need to wait 2 seconds.

This looks really nice! I was very impressed at the download speed, so good job on that one.

All I'm really missing is something telling me if anyone is nearby. That's probably not easy.

Were you guys at the YC startup school? I remember talking with someone about a similar product, although the target market was slightly different...

The app is pretty bare bones, but it has promise.

Localization request: for anywhere but en_US, replace "1 mile" with "1.6 km" =)

We still use miles in the UK too.

I use en_UK, dollars, miles and Fahrenheit (and Rankine) degrees.

There is also http://geospot.meteor.com with auto purge.

Well, we had this on the roadmap for www.wobbleapp.me. Will be interesting to see the reception in advance.

Well, I would have tried it, but upon installation, it says: requires iOS7.

Sorry, not going to happen :)

Great idea! Be careful about nefarious use (aka sex offenders) if that's possible.

I tried to search in App Store on my iPad, it showed me no result ?

waiting for my location to be unlocked, i.e. midtown Manhattan

so, a few thoughts now that we have a room created for midtown NYC:

1. no push notifications? seems like a no brainer to have this when a new user enters the "room" (i'm calling these chat spaces that) and when a new message appears. suspect you'll get more re-engagement

2. why do I have a ribbon next to my username? see: https://twitter.com/daveambrose/status/395630523961577473

3. photos! would be good to share media in addition to just text. maybe for a v2

On top of that, @replies would be awesome as well. We already have a nice Back Bay/Cambridge/Boston chatroom going that is going along fine, but we have already naturally used @replies since we are so used to them.

yea definitely sounds necessary. im just wondering - lets say you leave the 1 mile area, somebody @ replies you, when you open the app you'll be in a totally new spot and the message will be at the old spot.

1. you're absolutely right. just pushed an update for push notifications and coding in some logic as we speak

2. you get a ribbon for being an explorer (being one of the people to unlock an area)

3. i want to add media , but im scared it might get crowded with a lot of nudes. maybe if we add karma and let users with a certain amount post, it would work

I wonder what the architecture for the messaging looks like.

it would be great to see closest online person to chat if there is no one in 1-mile. (Of course chat will be inactivated with that person)

black status bar text on dark red background, this needs to change. But a very neat idea! Already found some people to chat with in Tokyo

CB Radio

Reply on news.ycombinator.com