Comment by danfang | Hacker News Reader

Comment by danfang | Hacker News Reader

Hey HN,

I'm Daniel, founder of Thread.

I started Thread two months ago because I believe current social media and social networks are built in a way that harms our ability to meaningfully connect with one another. They're highly addictive, superficial, and socially isolating. They also run a reputation of violating user privacy and selling it to third parties.

I strongly believe in ad-free, private social networking as an alternative. It's all about building and maintaining relationships with the people closest to you. If this sounds interesting, please try out the app! It's still very early, beta software.

On pricing: I think advertising is out of the question. It's a strong misalignment of incentives in trying to deliver long term value for end users. I have some ideas here, but the most obvious thing that comes to mind is a subscription based service (say ~$10/year).

I would love to hear your feedback (questions, comments, suggestions) on what I'm building.

Reach out to me personally at daniel@thread-app.com.

https://get.thread-app.com/privacy-policy/

We may disclose personal information to: ... third parties, including agents or sub-contractors, who assist us in providing information, products, services or direct marketing to you.

There's no marketing in iMessage, certainly not based on the things I type into iMessage.

Also, to be clear, "marketing" is advertising. You say "ad-free" but then you give yourself the right to advertise.

There's also nothing I can find about end-to-end encryption, which is one of the best features of iMessage. I almost exclusively use iMessage because I know Apple isn't spying on me. When friends message me on Instagram, I reply in iMessage.

Founder here.

This is great feedback. My interpretation of the clause is that we can use services like Mailchimp to send you marketing materials for when we release new products and features. These would be strictly opt-in.

I'm not a legal expert, but I will try to come up with a policy that more clearly expresses that you will NOT receive marketing about third-party services and our only usage of these services would be to update you all on our new product offerings.

We're considering offering E2E encryption, but we believe that we can deliver a ton of value by allowing users to search through their messages and for us to efficiently organize and categorize your content. I think there is a middle ground where we can encrypt non-pertinent data, but still preserve metadata so we can provide some intelligence on top of the content you choose to share on the platform.

> I'm not a legal expert, but I will try to come up with a policy that more clearly expresses that you will NOT receive marketing about third-party services and our only usage of these services would be to update you all on our new product offerings.

You do not need to be a legal expert to state this in your policy in plain language.

Examples:

"We will collect your contact information for the purpose of updating you about this product."

"We will not collect analytic data."

"If this policy changes, you will be notified 60 days before data collection begins."

You do not need to predicate openness and honesty on legal expertise.

Fixed.

It now reads: "We may disclose personal information to: third parties, including agents or sub-contractors, who assist us in sending direct marketing about our new products and features"

What I meant about legal expertise is that we don't want to be in violation of our privacy policy if we decide to use tools to understand our users better, or if we decide to email users about new products and features. Which is why I left some room in the privacy policy to enable such features.

still leaves wiggle room. Suppose a "direct marketing sub-agent" opens up shop, helps you send a couple of emails, so technically they are assisting you in sending direct marketing, but then that allows you to maybe sell all of your users' personal info to that third party who just set up shop out of a vacuum.

Even if you have good intentions and don't sell them the information, what's the guarantee that these third parties won't use the information you provided them for personal gain anyway?

Two things.

1. We'll be transparent about the services that we use. That way you'll know who your data is being shared to, and what data will be shared.

2. We'll try to take an Apple-like approach of anonymizing data as much as possible and only sharing the few pieces of information we need to deliver a better product.

I'm sorry there's no fool-proof way to guarantee perfect data privacy - at some point you'll have to trust us to do the right thing with your data. We hope that our incentives will . be aligned. Our service is predicated on keeping your information private and that's what you'll be paying for.

> We may disclose personal information

This alone will be enough of a deal breaker for privacy nuts, such as myself. If I'm going to use your service, you just champion privacy. This means no marketing gimmicks.

Sorry. For me this comes no way close to being an iMessage alternative. Even if you become a privacy champion, you still need to expect us to trust your servers, which is a tricky thing without end to end encryption.

Maybe an alternate model is to get funding from big corporate sponsorship and have a distributed network that could be hosted on our own droplets (using open source, of course). You can then have a paid hosted saas for those not technically ready to self host. Until there's a solution out there that does just that, I'm very unlikely to bite.

Yeah - I don't see Thread as the right product for privacy nuts, as you refer to yourself. There are better services for that, like Signal.

I imagine Thread as a company that focuses more on delivering products that allow people to build more meaningful relationships and connections in a way that benefits their long term mental/social health. I believe we can do that in a way that delivers an 80/20 on privacy too, by not using targeted advertising, and having 0 incentives to sell your data.

If you're not willing to trust our servers, I don't think the product is right for you. Hopefully we build trust as a company that does right by our users in the long run, unlike what I'm seeing from companies today.

I'm definitely not your target audience here so please take this with a grain of salt (as I might misunderstand what people value about privacy), but isn't "using tools to understand our users better" exactly what privacy-first people _don't_ want? Gathering that kind of metadata about what users on the platform are doing, instead of making it end-to-end encrypted so you (as the provider) can't see into what people are doing?

> We're considering offering E2E encryption, but we believe that we can deliver a ton of value by allowing users to search through their messages and for us to efficiently organize and categorize your content.

Why must you organize the data? I’d like to be able to perform searches client-side, just like iMessage.

I think you're right that you can probably get some value out of pure client-side search and parsing. Here's my take:

1. I like being able to search through old archives and interactions that might have happened a long time ago. I believe there's value in preserving history and being able to easily look back on it.

2. We can provide value by doing basic message parsing. Right now, if your message includes a Youtube link or Spotify URL we'll give you a player to view that content in-line. If your message has a Yelp link, we'll pull the restaurant details such as where it is, how expensive it is, etc...

Our mission is to provide the best service to allow people to build meaningful, long-term relationships, because _that_ is a need that's not being addressed. A lot of the features we want to offer will require some centralized processing. If you want a purely anonymized, E2E encrypted service, I would recommend Signal or Telegram.

You probably will need to take advantage of network effect in some way. I.e. if I have a $10/year paid account and I create a group, the person joining my group can join it for free, however if they are trying to join another group on their own then it would get them to pay. Or free for X period of time, because getting a person on the platform to start is hard.

I registered but I don't know anyone that has it so I have to convince people to set it up. Maybe allow public groups? Or maybe not.

Looks rather cool!

I created a group and wanted to update the name, I was confused by how it worked, I thought I'd type it in and click the back arrow seeing no button. Realized later I have to tap "Done" on my keyboard. You might want to consider adding a green "Done" button below or near the "Delete Group" button even though it's slightly redundant, it's less confusing.

I'm trying it on Android. Not sure how I can convince friends to try this out though, have a hard time getting people to use Signal. How's this compare to iMessage in regards to encryption btw? I know encryption isn't entirely trivial to cover this early on.

Edit: Also when I click "Contacts" and I deny permission on Android it crashes the app. Might want to handle that case.

These will both be addressed in the next release. Thanks for the due diligence.

No problem, it looks great otherwise!

I'd say charge for certain easy to develop features. like number of circles or extended message history, or creating ephemeral messages, or viewing ephemeral messages twice, extended emotions to add to someone else's message (instead of just a heart or a laughing face, more wider range)

and maybe less easy to develop features like animated stickers

Yeah, I'm definitely considering a freemium model. My own struggle is that I believe deep, interpersonal communication should be for everyone - it feels wrong to "exclude" people from certain mediums of expression just to make a profit. I have some more 'experimental' ideas that I think would allow Thread to remain free at its core, but still charge users for additional, advanced services.

This is super cool. How long did it take you to build it.

I started building out the product in early July, so it's been about two full months of work.

I recently quit my full time job as a software engineer because I really believe this problem requires my full engagement and devotion!

I apologize for not really having anything to enhance the discussion, but I like what you're saying and the ideologies you express; I thought you should know.

Reply on news.ycombinator.com