Show HN: What do you think of my new social app? (oncircles.com)
First time posting on Hacker News in about 2 years! The reason I’m posting is that I thought I’d write about the product I’ve been building, in hopes it resonates with people.
Why build yet another social app? Because I deleted almost all social media around 3/4 years ago as it was just net negative on my life. There was very little that was truly interesting, I didn’t care about looking good to other people and because of the ‘media’ and ads, the apps were all designed to make me spend too much time for what I was getting. I hated it and what it meant for people’s behaviour, including mine and my friends.
I also noticed that the vast majority of my friends were pretty much passive on the services, even if they had an account. Turns out that most people felt uncomfortable sharing to people they didn’t know too well, which inevitably happens as you meet new people, add them, and often don’t develop the relationship much further. I had a sense that there could be a better way.
In that, I remembered the days of Path and Google+ which had the model of focusing on particular people in your network, both through the feed and in how you shared. Path in particular was a ‘real life’ social network, something that despite being brought to market in 2010 or whatever, seemed to be more relevant today.
So problem found, problem solved. I took the journey of learning how to code when COVID hit (I was working in Architecture and Design and was about to start my masters degree at Harvard GSD in Boston), then one thing led to another, my prototype garnered some investment interest, and Circles was born.
The idea is very much like the name suggests, it’s about adding your contacts (synced through your phonebook) into Circles that define what the relationship is. This means that you only see things from people you have tagged (rather than everyone) and when you share, you choose exactly who it goes to (rather than all your friends or followers).
It also takes cues from other privacy focused social products. Posts are encrypted, reactions and comments are only shown to people who are contacts with eachother, and user profiles only show that which has been explicitly shared to the user viewing the profile.
As they were mentioned briefly before, it’s essentially a crossover between Path and Google+ with a wrap around layer of privacy, so the potential revenue has to eventually come through paid features rather than ads.
We’re in the app store and google play store (yay, cross-platform JS frameworks), the invite code is ‘FULLCIRCLE’ and download links are below:
Landing: https://oncircles.com
iOS: https://apps.apple.com/gb/app/circles-share-more-with-less/i...
Android: https://play.google.com/store/apps/details?id=com.oncircles
I would love to hear your feedback and suggestions on the idea. Hit or miss?
PS: If you want to chat about the idea or are interested in working together I’d love to chat. Always interesting to meet people on HN and we have the funds, ideas and drive to continue making cool things that can solve big problems. Send me an email at james@oncircles.com
Thanks!
143 comments
[ 5.4 ms ] story [ 203 ms ] threadI guess the question we're trying to answer now is whether we narrow in on creating a more intimate social network for regular use or create a type of social-email hybrid for this exact use case.
Many people says they "hate" social media, in reality everyone is still using it and it's very hard to even convince people (even close family) to use another messenger even though the barrier is way lower than a social network.
So in that the hurdle has been overcome because people are able to share content and interact in ways that they 'can't' do elsewhere. There's also a group of people (like myself) who avoid typical social media, so downloading a different solution is less of a hurdle because it's designed for them.
However, you point about people 'hating' social media yet still using it is well and true. It's almost like fruit vs mcdonalds, everyone would prefer to be an avid fruit eater and have control of their diet, but few do in the end. In that we believe the only moat for social networks is providing utility that other networks can't - YC wrote a really good blog post on this in relation to WeChat (https://www.ycombinator.com/library/3t-how-wechat-grew-to-be...). We're obviously not there yet but it's a start.
Also, if looking for a solution to that messaging problem, Beeper and Texts.com are creating something exciting there. Perhaps the social media equivalent of these services will be enabled by interoperability by DeSo, ETH or the like. Interesting to think about.
Update:
On maybe the fifth try, it worked. But now it's requiring my contact list to proceed. The app itself on the home screen is just called "rn-fetch-blob" (i.e. a default react native project). I'm not going to give contact list access to an app that didn't even bother to give itself a non-default name/icon on the home screen. I think the app needs a little more time in the oven before you have other people try it out.
So a couple of thoughts: 1) Allow me to sign in and post texts without permissions or sharing contacts. 2) Give the option to ask me for photo/file access permissions when I push the photo upload button. 3) When adding contacts, a prompt for contact permission is okay then but should be optional i.e. I should be able to put a phone number or email address in manually. I don't know if you had that or not I stopped at the file permission.
But let's assume for the moment that you aren't running a contact harvesting scam... I hardly use Facebook and I'm very particular about who I add on there. I actually go back and remove people I don't interact with or who create noise --I want to be able to log in once a week and see what people are doing, not scroll through memes or other crappy shared content.
My contact list on my phone, on the other hand, has friends, ex-friends, family, current and former coworkers, etc. Most of them I don't want to be connected with on a social network.
So I guess what I'm saying is that from my perspective, you aren't solving any problems, and are just a potential security risk when it comes to sucking up contact information.
Thanks for the grace there, and that's a problem we are solving. You might have X contacts on Circles but you're only going to see from those you have placed in a circle. So only content from people you actually want to see from.
In terms of syncing contacts, we don't store any contact book information on our servers. We pass through the contact book to see if a user exists there, we return them back as a connection. If not we return them as someone you can invite.
But... Again, you expect me to trust you. And if I don't trust Facebook or other well-known contacts to use my contacts appropriately, why would I trust what appears to be two random individuals in the UK to not retain that information? Are we supposed to forget the lessons of the past with all of the apps that have used our data inappropriately?
Do you log who someone has looked up in Circles when they join? If you do then you could trivially reconstruct someone's contact book.
That's what Facebook were found to be doing, except they were actually creating 'shadow profiles' for the contacts.
In the end for me the question is: How can I trust you to not leak my private posts? Why should I not continue using Signal groups? I realize that I’m probably not your main target group.
You can trust us because we say we are going to respect your privacy and we need to build trust and do what we say we will in order to build a great brand. I understand this often isn't enough and trust has been abused by other companies, and for that we'll keep pushing on the privacy front of the product.
I don't want to sync all my phone contacts because many are confidential; do you provide a workaround?
I loved the circles concept on Google+, but many people said circles were too much overhead to create and maintain; how will your app be better?
Your landing page and video emphasize one person's personal connections, whereas Instagram and similar social networks are seeing major usage in B2C accounts; what can you say about your ideas for this kind of usage?
Good point and something we haven't experienced before. There definitely needs to be a workaround. We had manual search and add in a previous version so
The answer here is UX. Google+ UX was clunky at best, where their circles feature was essentially just an add on to the Facebook experience (which facebook then replicated through lists). On Circles, the whole app runs through Circles and we've focused the app entirely on the benefit of adding people to Circles and creating more intimate connections with people.
Great point. I think for influencer, B2C type accounts, for now people should definitely continue to use Instagram as its not something we're entertaining for now. However, the way WeChat (I lived in China for a while) splits personal and business connection is fantastic and something I think about a lot.
In time we'd love to explore the route of a hybrid interoperable chat (beeper.com and texts.com) and social app, where B2C accounts are relegated to a more minor part of the experience (90% real life connection, 10% other). This works amazingly in WeChat because the little communication they have incentivises B2C accounts to communicate with signal and also makes WeChat monetise beyond advertisements.
I think this is a great idea. What are your strategies to make users to bring their friends?
Most bug feedback I've gotten since going live has been at the verification page for your phone number. If this is the case, the solution has been to go back a page and submit your number again - happy to support here over email if you'd like!
Our strategy revolves around the intimacy of the sharing that the app has created so far. Despite being simple (and buggy it seems!), you really only need 2-3 family/close friends to make the app joyful - successful or not we've aimed set up the app so that you get that experience as soon as possible.
An (non mom-test) example here is in how it improved my close relationships. My aunt passed away from cancer a few months ago, but before that we were able to share moments on Circles that otherwise would have not made it into chat and/or the typical social apps. Beyond holding the experience dearly, the possibilities of intimate sharing outside of typical social media made her invite her friends.
We can of course bake a bunch of viral marketing features into the product (and have started doing so) but getting this single thing right is by far top of the list.
As someone who also abhors social media, how is this different from those experiences? How are the circles managed? How are my posts and feed managed?
I used Google+ back in the day, and the circles concept was introduced with that. The circles were somewhat tedious to maintain, and became difficult at scale. How is this easier or better than that?
Also, without a web option, it's a no go for me for a social media app.
Also a great point, as mentioned in the OP, I've been off social media for some time because I hated them. So bar the standard stuff of onboarding, posting a piece of content etc. We've pushed to do the exact opposite of a normal social media app. Some examples:
- You must add someone to a Circle to see them in your feed, and share to them. - Reactions and comments are only shown between mutual contacts. So if you're viewing a post - Names and Posts are encrypted at rest, and we don't ask/take any data we don't need. We want to push this further in time but our intent is that we actively eschew collecting specific data about individuals, so don't have incentive to make the app driven by ads/addictive features.
Google+ was interesting but ultimately was just Facebook with more friction (eg. connect with as many things and people as possible). We're using the Circles concept to target the inner parts of your network, rather than everyone. Still a WIP, though!
Also good feedback, we focused on mobile because of limited resource but a web option would be great. Is there a specific reason it's a no go for social apps?
Personally I'm quite conservative about installing apps due to concerns about security, battery usage etc - the only social media apps I have on my phone are WhatsApp for family stuff and Twitter for casual scrolling. I'm happy to just jump into a web app and try it out though.
Web-app seems to be the consensus here so will definitely keep that in mind. Thanks for the feedback!
These two statements don't go with one another.
Also I think WA / Telegram groups basically replaced what Path / Google+ / FB Lists set out to do. It's just more convenient.
Right now Instagram does all of what I would want from a "public" social network while I have whatsapp groups for close friends.
True most networks are passive, and the public networks are catering for this. I would also agree that the chat apps is a great solution for close connections and often use this as their primary social network for this reason.
From experience, because chat is a linear thread, conversation focused and notification heavy, it's really not good for sharing things that don't directly concern the person/group you're talking to.
So Circles is all about creating a space for sharing your life with those who matter most, sharing things that you don't necessarily need to share but would like to away from the spotlight of social media. For example I've been sharing my travels in Portugal to my close friends and family circles (around 20 people who are mainly not interconnected). If I did this on chat I would be opening/maintaining 10+ conversations about myself! This is the problem we're solving currently.
That said, overall I agree that there's an overlap there that takes away from the value we offer to people right now. Gotta start somewhere though!
But I did use Google+ and it was just no one else really using those features.
So nope I'm not seeing the benefit of syncing my contacts with your service.
WhatsApp won in our family for sharing.
That's true, same with Facebook links and in some ways, close friends on Instagram. Our argument here is that rather than having it is as one-of-many features, we're focused on making a joyful experience around this single feature, creating a different space for sharing.
Feedback noted though!
Trying to create interested-based circles was also hard, as the cardinality explodes quite fast (and you're back to square one with not knowing who you share what with).
I guess this is where Snapchat is onto something with giving a very rapid selection of who you want to share with. I usually have a small group of people in mind when I want to share something.
Defining that group, fast, almost on a per share level is the key challenge.
I think the "circles"-approach could be a way, but I think it must be much more dynamic and alive than how Google+ solved it. It must continuously evolve.
Couldn't you make circles of circles in google plus? I remember something like that, I had circles and then grouped them based on interest.
On that point we also offer a quick/simple mechanism for sharing to specific contacts. So you're able to share to circles and/or specific contacts.
If you try the app, I would be keen to hear your feedback. Shoot me an email if that is the case.
One idea might be to go the Pateron way but with a free tier. But then its something akin to youtube/tiktok not really a social network.
We plan to monetise through paid premium features, none of which have been included in the first public version of the product. So previously didn't think it was relevant to talk monetisation!
One point against my argument might be reddit gold kind of stuff. They are more of a status signal. You can monetize status signals.
In terms of monetising utility, Beeper/Texts.com seem to be getting a lot of attention, where they charge 10$ a month for their interoperable chat service. People seem to be happy to pay to have a centralised client for all their chats.
Otherwise as you've mentioned you have status signals, which is the case for Twitter Blue too. There's also the dating apps.
it is?
Would you pay for a subscription to a social network without ads? Doesn't seem too far from an actuality
The problem is, there are way too many free options for us to organize. And this doesn't bring anything to the table which is spectacular.
For the desires you've stated above, is there any product that comes close?
After writing all this out I feel like what I want is ok cupid but for finding friends. The problem with that seems to be every time someone tries to do that they have one leg in the dating world (and I understand why, its lucrative). For once can someone actually put people discovery at the center of their product?
There is a cool app that I've been using called Lunchclub (https://lunchclub.com/), it links you with another person one time a week. Might be worth trying if you haven't already. Otherwise I'd really been keen to chat on how we could solve that problem!
Completely true about the people discovery apps (probably why most of them end up as dating or a hybrid dating/networking), so in my experience it seems that the only thing that overcomes this is strong interest based products focused on groups.
https://oncircles.com
https://play.google.com/store/apps/details?id=com.oncircles
https://apps.apple.com/gb/app/circles-share-more-with-less/i...
- circles from Google
- combined pictures ( haven't seen this one and is a must!)
I'm downloading.
Edit 1: Required contacts ( Ugh). But - Say hello to @God ( God, i really hope i can change the handle later on) = No forbidden username list probably.
Edit 2: Ugh. Invite only lol. Disallowed contacts & storage again.
The app is invite only but I've posted an invite code in the OP (FULLCIRCLE), also will remove the invite proxy in the onboarding in the next update. Hopefully you'll still try!
The front end is built with React Native (JS) as it provides a way for us to more or less maintain two OS with one codebase. It also is somewhat simple for us to convert out code for web with react-native-web. We do use a series of community build packages (more than I would like) but that's how its went for us early stage.
The back end is built with Rails as a simple API layer, keeps it simple to make models and controllers and build lightly with the focus on the architecture. Phone verification integrated through Twilio and we use Lockbox for at-rest encryption.
Hosted serverless on AWS EB (if going back I would stick to Heroku at least until this point and beyond).
As an experiment we also were building on top of the Matrix protocol (https://matrix.org/) with their Ruby SDK, but the cons outweighed the benefits in the end. Would love to figure out a solution here, and love the idea of interoperability championed by products like beeper.com.
Overall, I think that React/React Native and Rails is a good stack for hacking MVPs and would recommend.
Still, super impressed at what you've built.
Looking forward to syncing up soon!
The main reason was because we were building with the Ruby SDK and it didn't have bridging (future chat...) nor a direct E2EE solution included. In that, we had a bunch of business and product solutions we wanted to understand and Matrix was clouding our ability to do that.
Would still be very keen to see a consumer grade social network built on Matrix, though.
Chat soon!
I really like your idea. We're building something similar with a focus on photo sharing.
Have you considered making the contact book upload an opt-in? We also thought about connecting our users via their phone book, but we chose a different approach as we're concerned about privacy.
If you make the contact book upload opt-in, you at least give the user the chance to try out the app before giving away the contact details of their whole phone book (without their permission).
Super similar concept. A few apps I've seen pop up in the space are Lalo (https://www.lalo.app/) and Waldo (https://waldophotos.com/).
Yes, we initially had it this way and it seems from this HN launch that's what users prefer! The feature is flagged so we can bring it back and will likely do so (having an opt-in as you've mentioned).
Would be great to connect and share learnings. If you'd like send me an email (james@oncircles.com).
- asking for contacts while saying nothing about how you monetize is a huge red flag to me
- since you are already using cross platform JS framework, how about making it available as a PWA? or maybe provide apk directly? google store doesn't cover a sizeable number of android devices
- your landing page still says 2021 in its footer
I guess they are talking about data-mining your at-rest data for personalised ads.
- Fair point. Noted.
- Both points here would be a relatively easy step, so will figure something out. Thanks!
- Noted. Thanks!
So to be clear, anything you share on Circles is encrypted in transit with SSL/TLS, then encrypted at rest on our servers, to which we currently (but hopefully not for long) hold the keys.
As others have asked, can you tell us more about your pricing strategy? I would rather that you start charging right away, so I know that a) this has a chance of surviving for a longer period of time and b) you won't fall back ot ads or the VC teat :-) Also, as you said, since it doesn't depend on dozens of my friends using it, I think it would be easier to convince a couple of my friends to start using and paying for it.
Another question is: I'm guessing your biggest competitor are WhatsApp/Signal/etc. group chats. Why is it better to use Circle instead of group chats (which are also often organized as circles of friends)?
It would have been a great idea to add a small price in the app store, as my core assumption was that people would not pay for a social networking app alone. Maybe I will change my assumption there. You don't need to answer this if uncomfortable doing so (or can send by email) but how much would you pay today to know the app is sustained ad-free?
Otherwise, we are fortunate to be funded by a single angel who gives us freedom to decide the path of the company. He invested with the recognition that the tides on privacy and social media are changing and that a more intimate, private social network would be a definite solution in the future (In his words, "the two most important parts of anyone's network are the closest circles and the fringes, so Circles focuses on the former").
Our next stop was to build a Circles designed interoperable chat that would allow you to send and receive messages to pretty much any major third party app. We would charge for this. However, right now we collect no individual data from our users and most user information is encrypted at rest, making it very hard for us to build any real ads mechanism.
In relation to group chats, there is a notable difference in how Circles are set up. In a group chat everything is shared to the group, and all the members are directly connected to each other because they are in that group. Circles in their current setup are closer to broadcast lists.
1. The vast majority of our privacy is in product experience. Sharing with less people instead of everyone, making reactions and comments only visible between mutual contacts. 2. Names, Posts, comments and replies are encrypted at rest. 3. We sync your phonebook for contact discovery and don't store it on our servers. Therefore it is not linked to your identity. This is a hard problem to solve as Signal wrote in this blog post (https://signal.org/blog/contact-discovery/)
That said, privacy is a super important issue and external testing and open-sourcing is something that we desire to do. So thanks for the checker here!
We are many that had trust broken one too many times.
IMO encryption/privacy means nothing if I can’t ensure it’s what it says myself. Ability to self-build is the only way to make that happen.
Often people don't use the list feature on Facebook because a core product assumption of Facebook is that everyone you add is your friend, and lists are not a priority in the Facebook experience. The Facebook sharing experience is also centred around engagement, where more numbers makes you look better. So users have the option of using lists on Facebook but there's little incentive placed there.
Lists also don't have much impact what you see. This means on Facebook the vast, vast majority of content still comes from people you don't really care about and random content accounts from businesses and influencers. So secondarily, Circles feed is simply chronological and only contains direct content from people you have added to a circle.
I recently stopped using Instagram because it was filling up with nonsense. Also they push features in your face that you don't really want to use. Facebook is also a sad story. But for some reason we keep returning. We're like little rats in an experiment, waiting for our next endorphin jolt.
Can relate to that, it was the same for me.
- People share privately in group chats. For example, in Whatsapp groups.
- Circles were tried by Google, and got nowhere. One of the big lessons was that people don't want to organize friends into circles. In group chat (like in Whatsapp), one person creates/admins the group and all its members benefit. The grouping (chosen by one person, say "Tennis club") is more often than not equally relevent to all its members - sparing each member the burden of creating and maintaining circles individually.
- If you're talking about encryption, you'll need to be clearer about it
- They do. There's definitely overlap there although because groups are often centred around a topic (school, work, hobby), it's generally not the place to share things that aren't directly concerning the group/people in the group. Eg. I want to a great building I've just visited, I'm generally not going to share it to the group chat for my class in my masters, but I would like certain people to see about it. Circles here are closer to a broadcast list.
- Somewhat true, I think there are many points of failure in that case study. Our hypothesis is that by making a social network function entirely through circles, people will have a much more enriching and intimate experience. This has been proven true with current users, despite it not being for everyone.
- Noted. Thanks for this.
Understand that, to anyone who owns a home or is registered to vote, a phone is effectively useless. Very nearly every call I ever receive is a refi offer or vote canvassing or someone who wants to buy my house. The purpose of a contacts list is a way to allow me to know I'm receiving a call from a source I actually want, or need, to answer. But mostly this means businesses. The numbers are associated with an office, not a person.
And, as another commenter noted, the numbers that are associated with persons are effectively anyone I ever had a relationship with and never deleted the number. That includes women I went on one date with 20 years ago, people who are now dead, neighbors who haven't been neighbors in decades. I don't have their consent to share that information with you.
I understand it's a way to bootstrap a social graph without being able to export a friend's list from some other social app, but I would urge you to reconsider. For all of the other reasons I will never again use Facebook, at least one thing I think they got right was just showing you the friends list of existing friends who had opted into sharing that information with you. That was probably the best use I ever got out of it, since it allowed you to discover people you used to know and might want to reconnect with, but had never gotten or lost their contact info. It's much slower to require users to add friends one by one, but at least this requires a rough form of symmetrical consent.
Phone numbers are not private. I have never changed mine. Effectively, everyone I have ever called and told my name to may have me as a contact, and any one of them could have listed the number somewhere else. That is obviously how I get on all of these constant contact or whatever they are call sheets. Someone scraped it from public property records and now I'm a contact for thousands of people I don't know. I would not ever want to allow those people to auto-add me to a social graph. It's just inviting yet another avenue for spam. Even if you don't directly inject ads into your service, there is nothing stopping your users from doing it. You're inviting companies like LuLaRoe to leech off of you if you ever grow successful enough.
I already have a set of grouped contacts across different apps ( Circles, Private Interactions) all with encryption (Encrypted Posts) and it's trivial to switch across apps to share anything, especially via the phone, which stores any media I'm sharing ( The Vault).
The friction from downloading a specific app then also getting all of my different contacts on that app so I can replicate functionality that already exists makes this seem like a non-starter for me.
Is there some additional functionality or hook I'm missing? Not trying to be snarky, asking genuinely.
All that does is share content with nobody. To what? Practice pressing the share button in yet-another app? (I'm sure the workflow is common as the others)
That edge v. center distinction dictates the use patterns, and social media is therefore used differently than a chat app. With a chat app, the social media aspect is drowned out by the sheer volume of communication, largely text based and not media based.
But why is there a market? There is a growing movement to get away from Meta-based social media apps, because they're incredibly addictive and rob users of their attention / time. That being said, I still want to see what my friends are doing... but I'm sick of needing to choose between getting life updates from friends or having my brain back.
Build me a platform that isn't Meta-based and isn't addictive and lets me share fun things with my friends and I'll gladly pay a monthly fee for it.
Would love to chat about your current social usage and what you would expect from us in a way that you'd be happy to pay for. Can get me via email at james@oncircles.com if you'd be up for chatting!
In that, chat apps are great but are not designed for sharing individual updates, without running several different conversations & dealing with notifications.
Appreciate the feedback though & hopefully we'll develop into something that can gets you excited.