Pop founder here. We’re a team of 5, and folks from our team have built products like Screenhero, Slack Calls, and most recently, Screen.so.
Pop lets you have a video call in 3D, with positional audio. It works in a browser, without any download. The experience is surprisingly fresh, and the social dynamics are quite fun.
You can feel like you’re part of a group, you can see who’s looking at you, you can go to the side with someone and have a separate conversation, without interrupting the current conversation. You can share a screen that floats in space (and always faces the viewer). You can even bow to say hello!
We use Mediasoup for the WebRTC backend, Ionic/Angular for the frontend, and BabylonJS as our 3D framework.
Also, as an experiment, we’re hosting a virtual launch party at [EDIT: 6 hours later, and we're done! Thanks to everyone who came by!].
Happy to answer any questions here too, of course!
Your homepage doesn't explain anything about what I should do when I land on it.
What's a space? What's going to happen if I create or join one?
Why would I want to do this?
Or is this intentional and you're soft-launching or just doing some testing?
edit: I clicked create a space and you immediately ask for access to my camera.. but there's no explanation of why you're asking for it, or what I'm getting out of it.
I feel like there are huge assumptions about users understanding what's about to happen..
And before I grant access to a remote site to my camera, I want a better idea of why I'm doing it, and what you're doing with the feed.
I think you mean, as long as it explains why, what and how in the shortest possible number of words it’s a good video. Spending time trying to be funny or not actually explaining would be bad, if it takes a while to explain more thought and structure may be needed.
Sorry, I didn't mean to come across that way. I understand how you would like it, but personally I'm happy to play a video with sounds if it explains things quickly and clearly.
I didn't take it as a negative. I just wanted to clarify. I'm fierce on this topic as I can sense video becoming more pervasive and I think it's usually the wrong choice - as well as personally hating it most of the time.
I watch plenty of YouTube videos but that's when I'm in a different mode. For understanding stuff I want to skim read short clear text with illustrations. Animation and video can enhance that but it shouldn't replace it.
If you check out the text of the link at the top of the page, it has some helpful context. Specifically, that the linked website is for "video calls, in 3D"
But in any other context you'll likely not know what it is. Even with the title, I have very very little idea of what it means. Video calls with one other person? Multiple? Are they recorded? Define "3D". There needs to be at least some high level explanation on the page (or a link to one).
I gave access to my camera and everything, but I didn't understand what this product does until I saw this screenshot in another thread: https://i.imgur.com/VVt0zCU.jpg
I think it would be really helpful if you could see your own avatar.
This is a great point. Connection process should be an array of the founder's faces with chat blurbs welcoming them to connect their camera with your avatar in the middle with a preview of your camera!
In short, you move around in the virtual space and your avatar's head is your webcam feed. There is also positional audio, so your position changes who you hear. User can also apparently spawn screen share windows in the sky.
Give it a try, it works and feel better than it looks. It is very similar to what you'd find in a VR chat application but it works surprisingly well from the browser with just the mouse.
(Completely unaffiliated, simply gave the demo a spin.)
We held our office Christmas party on gather.town. Really neat - a 2D 90s computer game meets Zoom. Haven‘t tried them for productivity yet, but as a social experience they are significantly better than conventional video chats.
Have you done any measurements of capture-to-display latency?
I've been interested in why Discord feels more natural than Zoom and so far I think it comes down to latency. As far as I know Zoom targets 150 ms, but audio-only platforms can get closer to 25-50, which can make the difference between what feels like using walkie talkies versus a real face-to-face conversation.
Any plans on how you will monetize this if it takes off?
We haven't yet measured audio latency, but our main product (https://screen.so) is all about low latency screen sharing, so we're very interested in reducing lag of all sorts. Performance is something we'll focus on in 2021 in a big way.
Monetization: very similar to what everyone else in this space does. Free for social users, paid for business users, and finding a good way to segment the two.
I’ve used Mibo for a bunch of department meetings at it was very well received! One of the great things is that it allows you to have serendipitous meetings with people you haven’t spoken to for a while just by running into them on the island.
Great team as well! They helped us set it up in a way that also worked for the corporate firewall.
I believe they use Jitsi as their backend, combined with some cool engineering to make it all work on k8s.
Yeah, I'm pretty curious because it must've cost in $10k-100k range right? Isn't this the type of vanity purchase that usually happens after you validate your idea with 10,000 customers and you got a round of investment to go big? I would expect a newly launched startup to use getpopapp.info, or used a made-up company name with an acquirable domain.
My experience is from 5+ years ago (when google counted keywords in domains), but a 3 letter domain would go for at least 6 figures, possibly up into 7.
And pop.com would be in the high end of that would be my guess, given it's better that most 3 letter domains as it spells a very marketing-friendly word
Yeah it’s nice but IMO I’m not sure that this is the product I would put on it. I mean how many more apps do I need that insist on turning on my camera? Zoom fatigue and privacy needs are real things. And now this apparently detects who I’m standing next to in virtual space. Why? Anyway I’m certain it feels awkward and uncomfortable and that’s not something you want with a new concept.
I would put a bubble bobble clone on the domain. This is what the internet is lacking and perfect for pop.com a major tour de force of classic arcade game remakes!
I think the players here have already validated many other ideas and have the cash to get this without too much problem. If it doesn't work out, I'm sure they can find a buyer...
> Isn't this the type of vanity purchase that usually happens after you validate your idea with 10,000 customers and you got a round of investment to go big?
- The founder previously flipped his startup Screenhero to Slack and got some cash out of that I suppose.
- Interest is negative, that money is just costing money.
- Short dictionary domains are a pretty low-risk investment, I don't know the market well but I suppose they tend to increase in price over time.
- So if the company fails he can sell it. Meanwhile it's an asset on the company's books so it's part of the valuation etc.
Sounds like this is the "second startup" version of rich-ish parents buying an apartment near college for their kid to live in, hoping to flip it for more when they graduate. Doesn't seem all too "vanity" to me, it's just combining pragmatic usefulness with an "good enough" investment opportunity.
Yeah, I'm incredibly curious as well! I think it's a very cool project, but ball-parking development costs, my guess is the domain probably cost more than the product itself, assuming they didn't go down NIH-type rabbit holes.
Edit: And to be clear, I'm not even necessarily knocking that decision. Maybe an expensive TLD could make life easier for a project like this, since it could promote trust / buy-in, increasing usage at an early stage. ¯\_(ツ)_/¯
Since our company is personally bootstrapped by me, answering this will divulge more of my personal financial situation than I would like to share.
However, I can say that I worked with https://news.ycombinator.com/user?id=ted0 who was AMAZING at helping find and acquire the domain. I can strongly recommend him as an amazing person to work with for anyone interested in buying a high-value domain.
Author co-founded screenhero which was valued at $1.1B after less than a year and later acquired by slack. 1 round of funding of $1.8MM from YC and founders club.
So a few million for a domain is probably the equivalent of me spending ~$500 for a domain. Honestly not crazy at all with that perspective.
We bought this domain a few months ago, and you're probably right, since Gmail seems to mark our outgoing emails as spam. I'm not very well-versed in how to undo the reputation the previous owners seem to have created for the domain, so if you have any recommendations, please let me know!
That's a long-term manual process. You need to go around to every spam list you can find and open up direct communication with the person/s that control it or go through their removal process (some offer a simple process, some are obnoxious, as you'd expect).
That said, I checked your site against blacklists. It appears clear -
One trick for upping your reputation with Gmail is to have as many of those recipients as possible (who need to dig into their spam folder sometimes) click reply and write back to you naturally. Gmail considers that behavior to be highly indicative of "not spam" well beyond simply using the "not spam" button.
Cool! I think this kind of avatar embodiment is the future. I'm working on one that doesn't use the camera, but audio lip syncing to provide a good experience. https://jel.app
Also if you want an OSS alternative to any of these commercial tools, check out https://hubs.mozilla.com
+1 for hubs, I only found it recently and was positively surprised when trying it out the first time. It is missing the webcam-video-on-avatar function though, or it is well hidden?!
nope, never implemented it. you can share your camera but they are not avatar-attached. the theory being that we didn't want to implement that unless absolutely necessary, since it has a lot of tradeoffs.
The purchase price for that three letter domain must have cost you something serious. Pop.com was that old DotCom project from Dreamworks and Ron Howard that went belly up.
Mozilla Hubs is awesome for this too, you can set up the room with objects like presentations, or just various tables and it also lets you have a little window for your webcam too. The audio is also spatially aware so you can have little groups of conversation too.
We bought this domain a few months ago, and you're probably right, since Gmail seems to mark our outgoing emails as spam. I'm not very well-versed in how to undo the reputation the previous owners seem to have created for the domain, so if you have any recommendations, please let me know!
It's a cool idea but after trying it, I feel it needlessly incorporates all the constraints of the real world into the virtual world.
Why should I waste time clicking my mouse to move to a certain point in virtual space when no-one is really there anyways? Isn't the whole point of virtual meetings that you don't have to physically be there to attend? Walking around in the void to hear others only makes my personal experience slower and less natural.
I feel like maybe the solution here is just to have Zoom + private "rooms" for separate conversations.
(Sorry if this sounds too negative, because I always enjoy seeing new ideas and there was obviously hard work put behind this but that's my 2c)
I agree that it doesn't seem like the most efficient way to have a conversation, so if that's the goal, I don't see it working. It also doesn't make much sense for having a group call with one or two people.
However, I could see some value in making public chat rooms that people could join. This way 50 people join a room to talk about a particular subject and naturally split off into separate groups within the room. It would be sort of like a being at a party, where you could walk around and mingle and network with different groups of people. It would be a fun way to interact with new people, especially in current times when everyone is locked inside.
So, I see it less as a business tool, and more like VRChat (although without all the crazy avatars and games) where strangers go to socialize.
I can see the appeal in being able to drift in and out of conversations. The private room dynamic is kinda rigid, I basically have to explicitly secede from one conversation to start another.
This got me thinking, what is most affordable way to create a videoconferencing website nowadays? Is there some cheap streaming service, or is it better to deploy your own server in the cloud?
If you want to support < 5 person calls without things like hd screensharing, you can just do peer to peer webrtc connections.
Once you get above that, there’s a variety of FOSS you can self host, like jitsi and mediasoup (what pop.com uses.) There’s also some relatively cheap paid APIs like twilio and opentok.
This is really cool, I joined the launch chat and people were emulating in real life behaviors like standing in a circle and moving their focus to the speaker. I also noticed that multiple conversations were happening at the same time, without interference since the audio is spatially modified.
Love the fact that it's entirely in the browser without an app too.
What you are describing sounds a lot like AltspaceVR
I am visiting a weekly XR Meetup on Thursdays and it feels a bit like going to a conference. You have someone presenting their work, with displays and people gathering in groups chatting about things.
I can't get it working. I've tried in vanilla (no content/script blocking) Firefox and Chrome. I've accepted the microphone permission, but it seems there's a hard requirement on having a camera attached? I'm getting:
"DOMException: The object can not be found here." - FF
and
"DOMException: Requested device not found" - Chrome
Since you state you can turn both the camera and microphone off, I think it'd be much better UX to not force camera and mic access up front.
Also, Firefox asked for permission before I clicked on the "Request Permissions" button.
- the idea someone just mentioned in the "launch party" about having a visualisation of an avatars audio "range" (how far they project sound) would be useful
- a visualisation that someone is speaking could also be useful (little speaker next to their name)
- definitely need some kind of password protection for joining, gate crashing would be pretty easy (unless of course you rate limit attempting to join a space to stop people brute forcing the space ID)
- ability to mute someone would also be useful in a more public setting (and the mutee should probably get a visualisation that an avatar cannot hear them)
- being able to agree with another person to go into a completely new space would also be useful way to take a conversation private
congratulations, looks cool!
I'm one of the persons behind https://laptopsinspace.de and it would probably be interesting to share some experiences.
Our tech stack is similar, yet different.
We also use mediasoup for the video, Svelte for the frontend and ThreeJS as the 3D framework!
This feels like a fun lightweight version of one of our products, vTime XR https://vtime.net/vtimexr which lets you chat with people both in VR and on mobile. It is amazing how much better chat is when you can face people and speak to them. When you throw in VR it gets pretty exciting too. Thanks for sharing.
139 comments
[ 2.9 ms ] story [ 151 ms ] threadPop lets you have a video call in 3D, with positional audio. It works in a browser, without any download. The experience is surprisingly fresh, and the social dynamics are quite fun.
You can feel like you’re part of a group, you can see who’s looking at you, you can go to the side with someone and have a separate conversation, without interrupting the current conversation. You can share a screen that floats in space (and always faces the viewer). You can even bow to say hello!
We use Mediasoup for the WebRTC backend, Ionic/Angular for the frontend, and BabylonJS as our 3D framework.
Also, as an experiment, we’re hosting a virtual launch party at [EDIT: 6 hours later, and we're done! Thanks to everyone who came by!].
Happy to answer any questions here too, of course!
Edit: there are more screenshots here: https://www.producthunt.com/posts/pop-9
What's a space? What's going to happen if I create or join one?
Why would I want to do this?
Or is this intentional and you're soft-launching or just doing some testing?
edit: I clicked create a space and you immediately ask for access to my camera.. but there's no explanation of why you're asking for it, or what I'm getting out of it.
I feel like there are huge assumptions about users understanding what's about to happen..
And before I grant access to a remote site to my camera, I want a better idea of why I'm doing it, and what you're doing with the feed.
I won't watch a longer video just to understand what random product does.
A simple animation, a voice over, and maybe some video captures of the software would be great.
Audio is unnecessary and means I have to stop everything else I'm doing. And so does anyone else in the same room unless I can find headphones.
And 3 or 4 seconds is ample time to demonstrate that specific aspect that is hard to demonstrate with the written word.
Off topic, but your both Andy.
I watch plenty of YouTube videos but that's when I'm in a different mode. For understanding stuff I want to skim read short clear text with illustrations. Animation and video can enhance that but it shouldn't replace it.
And I never, ever want spoken voiceovers.
You only know this because you saw the HN headline.
I think it would be really helpful if you could see your own avatar.
This is not some universal rule that cannot be broken. For most startups, landing page is useful. But not always.
I like the simplicity of Pop's page.
or an even earlier version of Google, "Search Stanford" "Search the Web" https://i.imgur.com/EKiQnJX.png
They both provide more information than Pop does.
In short, you move around in the virtual space and your avatar's head is your webcam feed. There is also positional audio, so your position changes who you hear. User can also apparently spawn screen share windows in the sky.
Give it a try, it works and feel better than it looks. It is very similar to what you'd find in a VR chat application but it works surprisingly well from the browser with just the mouse.
(Completely unaffiliated, simply gave the demo a spin.)
I've been interested in why Discord feels more natural than Zoom and so far I think it comes down to latency. As far as I know Zoom targets 150 ms, but audio-only platforms can get closer to 25-50, which can make the difference between what feels like using walkie talkies versus a real face-to-face conversation.
Any plans on how you will monetize this if it takes off?
Monetization: very similar to what everyone else in this space does. Free for social users, paid for business users, and finding a good way to segment the two.
Love it
Great team as well! They helped us set it up in a way that also worked for the corporate firewall.
I believe they use Jitsi as their backend, combined with some cool engineering to make it all work on k8s.
I would put a bubble bobble clone on the domain. This is what the internet is lacking and perfect for pop.com a major tour de force of classic arcade game remakes!
If I had the money or was running a big corp looking to launch a new product, I'd pay $5-10m for it. I'll be very surprised if it was less than $3m.
- The founder previously flipped his startup Screenhero to Slack and got some cash out of that I suppose.
- Interest is negative, that money is just costing money.
- Short dictionary domains are a pretty low-risk investment, I don't know the market well but I suppose they tend to increase in price over time.
- So if the company fails he can sell it. Meanwhile it's an asset on the company's books so it's part of the valuation etc.
Sounds like this is the "second startup" version of rich-ish parents buying an apartment near college for their kid to live in, hoping to flip it for more when they graduate. Doesn't seem all too "vanity" to me, it's just combining pragmatic usefulness with an "good enough" investment opportunity.
ps. wow this app works really well!
Edit: And to be clear, I'm not even necessarily knocking that decision. Maybe an expensive TLD could make life easier for a project like this, since it could promote trust / buy-in, increasing usage at an early stage. ¯\_(ツ)_/¯
However, I can say that I worked with https://news.ycombinator.com/user?id=ted0 who was AMAZING at helping find and acquire the domain. I can strongly recommend him as an amazing person to work with for anyone interested in buying a high-value domain.
So a few million for a domain is probably the equivalent of me spending ~$500 for a domain. Honestly not crazy at all with that perspective.
It's also on a bunch of domain blocking lists (don't have a complete list). I've submitted pop.com as a false-positive on OISD.
I wrote this elsewhere:
We bought this domain a few months ago, and you're probably right, since Gmail seems to mark our outgoing emails as spam. I'm not very well-versed in how to undo the reputation the previous owners seem to have created for the domain, so if you have any recommendations, please let me know!
That said, I checked your site against blacklists. It appears clear -
https://mxtoolbox.com/blacklists.aspx
Also if you want an OSS alternative to any of these commercial tools, check out https://hubs.mozilla.com
Our plan is to merge the two in the near future, and work on the intersection between 2D and 3D calls.
https://hubs.mozilla.com/
Wanted to get some input on what some must have features would be?
If anyone has an issue visiting the site double check your adblock.
make sure you get taken off freely available spamlists that adblockers sync with
then if I were you I'd talk to a Google exec to speed up the domain reputation cleansing process a bit (get some articles up in your blog asap)
Why should I waste time clicking my mouse to move to a certain point in virtual space when no-one is really there anyways? Isn't the whole point of virtual meetings that you don't have to physically be there to attend? Walking around in the void to hear others only makes my personal experience slower and less natural.
I feel like maybe the solution here is just to have Zoom + private "rooms" for separate conversations.
(Sorry if this sounds too negative, because I always enjoy seeing new ideas and there was obviously hard work put behind this but that's my 2c)
However, I could see some value in making public chat rooms that people could join. This way 50 people join a room to talk about a particular subject and naturally split off into separate groups within the room. It would be sort of like a being at a party, where you could walk around and mingle and network with different groups of people. It would be a fun way to interact with new people, especially in current times when everyone is locked inside.
So, I see it less as a business tool, and more like VRChat (although without all the crazy avatars and games) where strangers go to socialize.
Once you get above that, there’s a variety of FOSS you can self host, like jitsi and mediasoup (what pop.com uses.) There’s also some relatively cheap paid APIs like twilio and opentok.
Love the fact that it's entirely in the browser without an app too.
I am visiting a weekly XR Meetup on Thursdays and it feels a bit like going to a conference. You have someone presenting their work, with displays and people gathering in groups chatting about things.
"DOMException: The object can not be found here." - FF
and
"DOMException: Requested device not found" - Chrome
Since you state you can turn both the camera and microphone off, I think it'd be much better UX to not force camera and mic access up front.
Also, Firefox asked for permission before I clicked on the "Request Permissions" button.
Screenshots: https://imgur.com/a/BKY5bcE
I had a bunch of "phantom" avatars for some reason (https://imgur.com/a/ixBGuK8)
Some thoughts:
- the idea someone just mentioned in the "launch party" about having a visualisation of an avatars audio "range" (how far they project sound) would be useful
- a visualisation that someone is speaking could also be useful (little speaker next to their name)
- definitely need some kind of password protection for joining, gate crashing would be pretty easy (unless of course you rate limit attempting to join a space to stop people brute forcing the space ID)
- ability to mute someone would also be useful in a more public setting (and the mutee should probably get a visualisation that an avatar cannot hear them)
- being able to agree with another person to go into a completely new space would also be useful way to take a conversation private