gsreenivas

unfortunately DO does not seem to be as active as Amazon in maintaining the reputation of their IPs.

This is very consistent with what our experience has been with Elastic IPs combined with email authentication and blacklist monitoring.

We are using StrongSwan right now. We've taken a close look at WireGuard but have not yet completed our evaluation.

We automatically configure SPF, DKIM and DMARC for our customers. We are also investigating MTA-STS.

Device diagnostics are opt-in by default so they are not collected. Customer data like shipping and billing information is not opt-out unfortunately as we need to be able to process payments, ship the unit and track warranty coverage.

Appreciate the feedback on wanting more architectural details. This will be coming in a series of technical posts explaining how we designed and built the product. Stay tuned and thanks for your questions!

Amazon invests in ensuring their Elastic IPs are not on blacklists. Less than 2% of IPs we get through AWS are ever on a blacklist and when they are, we cycle through until we get one that isn't.

Two things: we will run the service in perpetuity as long as there are subscribers and we will be open sourcing what is required to run the service on your own.

Yes - we initiate a VPN connection first to the gateway, then inbound/outbound connections are over TLS. Over 92% of email traffic is over TLS and we will be exposing an option in the future where customers can require it or reject emails.

We don't believe it does. ISPs will only see encrypted traffic (a VPN tunnel to the gateway) so it's unclear how they will figure it's associated with a server.

Thanks balladeer - we will definitely be launching the developer program in less than one year from now.

Hey bootsz, when your connection is down, you'll still have a local cache of your messages, contacts and calendar events. Email has retry built in and sending servers typically retry for at least 48-72 hours. The hardware has no moving parts so it should last for a long time, but when there is an inevitable failure, there are encrypted backups that can be downloaded to a replacement device and decrypted using keys only you have access to.

yes - we do! and DMARC as well

Hi lvh - it's not a proxy. The EC2 instance can't see your emails. The Helm server initiates and receives TLS connections through the instance so it's really an extra hop on the internet. We don't see or keep data on deliverability of individual Helm servers. We use a dedicated EIP that we ensure are not on blacklists and configure reverse DNS as well.

Hey paraditedc - thanks for your feedback. We will be launching a developer program next year. We are very interested in a community of people like you hacking on Helm to build new services/features.

Hi TrueDuality - Giri Sreenivas, co-founder and CEO of Helm here.

We will be publishing more details to answer your questions in depth in a series of coming posts. I'll give you some quick answers right now.

We use Yocto to spin our own Linux build. Key projects we utilize: OpenLDAP Postfix Dovecot Darwin Calendar and Contacts Certbot OpenDMARC OpenDKIM OpenSSL Python SPF Policy Sieve SpamAssassin Duplicity StrongSwan Docker

We're using duplicity for backups. So you can use it to decrypt backups as you will have the keys as well. Backups are stored on a service we run using S3 and will always be accessible for the user even if something happens to the company.

We use duplicity for backups so there's nothing proprietary in our approach. There will be more transparency coming in a series of technical posts about how the product works, what open source software we use, etc. Appreciate the feedback - we'll keep this in mind as we move forward.

Thanks for your encouragement! I appreciate the feedback and keep tabs on our GitHub for what we open source down the road.

Thanks for the feedback!

Thanks for your comment! Yes - we are working on making the product available in Europe early next year. Please sign up for our mailing list on our website and we'll keep you posted!

We spin our own build of Linux using Yocto.

We are using an ARM-based SoC from NXP. We chose this to ensure that the device can only run signed, trusted code by implementing secure boot and signature verification of software updates. We will make a developer program available in the future.

I didn't use Protonet so I shouldn't speculate about what's similar or different about the products. I think we are in a time right now where people have a strong desire to own their data and are looking for a viable alternative to the cloud. They key, beyond building something very private and secure, is in nailing the experience and we're excited to share more about that.

Using a hardware root of trust, secure boot and a Secure Enclave for managing keys used for full disk encryption, it will be very difficult to extract decrypted data from a Helm server. The keys never leave the Secure Enclave, they aren't available to the application processor or memory.

Most cloud-based email services hold email in the clear - we believe this means you don't really own your data. Encrypted email services have challenges around search, access via proprietary protocols and the risks of running highly sensitive operations in client-side javascript.

Hey cwyers - this is a good question. Our offsite backups use keys that only a Helm customer has access to. They are created during the setup process and stored on a USB thumb drive we include in the box as well as your phone. We will be publishing how this is done using duplicity so people can see the details of how it works.

For now, a single domain per server. And yes, mirroring support is coming with 2 servers so you will have the fail over that you mentioned.

good point - I read it with the other connotation in mind. We plan to make a developer program available in the future. Stay tuned!

Sorry if that wasn't clear. We did not design our own ICs. We selected an SoC for its security capabilities and designed our solution to utilize it integrated with our firmware. We're using an SoC from NXP.

hey bachmeir - our Helm servers do not limit the number of users. For a family, for example, you'll typically come out ahead on cost with our product.

Hi keehun - thanks for posting this on HN! I'm the co-founder and CEO of Helm.

1 - First, we cross reference IP addresses we assign to gateway against known blacklists. This helps ensure emails will be delivered. We also fully support email authentication (DMARC, DKIM, SPF) and configure reverse DNS as well. Lastly, the IP address for a gateway stays fixed so the domain and IP will build reputation over time. Helm servers require the service to work to get around the residential internet connection challenges in the US (port blocking, dynamic IPs, untrusted IPs)

2 - We will be doing 2 things - first, we will publish as open source what is required for people to be able to run their own gateways with their own AWS account in the event Helm has to shut down. Second, the unit economics on the service are positive so as long as we have customers, my co-founder and I are dedicated to running the service. We take a page from Garry Tan and Posthaven in this regard.

3 - The way this works for sending emails, your devices that you compose emails on will connect directly with your Helm server over TLS. Your Helm server will then initiate a TLS session with the server hosting your recipient's email. So we as a company have no visibility to any of that data - at rest, or in transit. I hope this helps - I'm happy to explain this in more detail as needed.

You can request email sending support when you set up reverse DNS.

We will be making public the configuration for these instances as part of what we publish in open source. We haven't considered allowing customers remote access to the gateway but we will based on your suggestion. Thanks!

Hi tpetry - Giri Sreenivas, co-founder and CEO of Helm here. You raise an important issue with residential IP addresses not being trusted by email senders. This is why we use a static IP address assigned to a gateway for each customer. The Helm Server appears to have this IP when communicating with other email servers. This also allows us to define reverse DNS for a domain which is another marker of trust for email servers.

hi jstanley - I was answering the question about retry when there's a service outage. Did you interpret the question differently?

Hi apo - Giri Sreenivas, co-founder and CEO of Helm here. The gateway funnels packets to and from the Helm servers in people's homes. These are packets of encrypted TLS sessions so we can't do anything with this traffic. Happy to answer any further questions or concerns.