I recently came across a product by a company called Helm. It’s essentially a mail server designed to be as plug and play as possible for home use. The argument is that this gives you all the convenience of something like Gmail, but gives you full control of your data. All of it comes in a slick little package, and more than a few caveats.
Image Credit: Helm
The Promise
It’s a move that seems to be a little out of Drobo’s original playbook. Take something that a lot of prosumer or somewhat technical consumers (DAS, NAS, a server), wrap it in a simplified UI that abstracts away intimidating terminals, and spend some money on interesting industrial design.
Helm claims that out of the box, you can get this up in under three minutes. Here’s the process:
At launch, Helm is focusing on making this a mail server first. It offers email, calendar, and contacts hosting, including setting up a custom domain. They even handle creating that pesky static IP for users. Everything is encrypted across the board.
For all this simplicity, I see a number of issues.
The Problems
Perhaps the most basic problem is the pitch. Helm claims to help you own your data and gives your privacy back. But being essentially a managed service, it gives quite a bit of access to Helm to do setup, configuration, roll out updates, and back up critical data. You get a certain degree of data locality, which is an interesting option, but at the end of the day, you’re still trusting Helm to keep a lot of your information secure. Is this that much of a benefit over webmail offerings? I’m sure Helm has many good arguments why it is, but it’s not obvious to consumers, and that makes it a tough sell.
Another issue is cost and value. The Helm server costs $499, and requires a $99 subscription fee for services after the first year. This might be more palatable if Helm was subsidizing the hardware costs with the subscription and was able to deliver something impressive. As it stands, you’re getting an underwhelming hardware package, albeit one that looks pretty cool.
This is what you get for a $500 server. The ARM CPU is a little underwhelming, essentially it’s a mid-range smartphone chip from 2016 with less cores and frequency. It’s nice to see ECC RAM onboard, and 2GB should be enough for a basic mail server, but I’m skeptical how it will perform as more services are added. 128GB NVMe is…fine. Nice to have fast storage, but I’d rather have a smaller system SSD with some added capacity on the side. It at leasts provides for adding a secondary drive if that’s something you need.
It’s clear that Helm is using a server grade ARM platform, encryption is baked in from the start, and you don’t see too many ARM platforms with ECC support. Given the reliability of those components, I wonder why the company got stingy with a 1 year warranty. This device essentially requires an ongoing subscription to work, it would be nice if I wasn’t on the hook for hardware failures as well.
The last little red flag is how many features this doesn’t have available at launch. For a privacy-focused device, you’d think getting a VPN would be table stakes. It’s listed as “coming soon”, but who knows when that will happen. To not launch with such a basic feature makes me question the priorities of the company. File sync, password management, and messaging are also on the list of future features. Hopefully some of these will be available before the first subscription payment is due. Otherwise the value of that annual fee is dubious at best.
Is Helm Worth It?
If you’re looking for hardware value, you’d be much better off with a low-end Intel NUC running ownCloud on Linux. That would give you a similarly low-power server, with a lot more capability. However, you’d be saddled with configuration headaches. One thing I can guarantee is that you won’t have a mail server up and running within three minutes with that configuration.
Every few years or so, I start thinking that I should use a Raspberry Pi or an old laptop and spin up my own personal mail server. It’s one of those flights of fancy like owning a vintage car. At first you see all the benefits, but the barest research reveals all the headaches you’ll be taking on as part of the project. That’s what attracted me to the idea of Helm in the first place.
When I first read about Helm, I thought of it as kind of a successor to Drobo. Both feature friendlier designs, have easy to understand dashboards, and seek to limit technical exposure. But after some consideration, the two really stand apart. Helm must be a consumer device, because it seems far too underpowered and limited for ever the smallest of offices. There’s no local data redundancy, and I’m doubtful how many accounts the server can handle at once.
Helm is an interesting idea, but one that seems born incomplete. A configured mail server is nice, but given the lack of features beside that at launch, I’m skeptical there will be much reception. I wonder if the company would have been better suited marketing an even lower power server as a VPN appliance, then backing into a broader server offering. That has more of a direct privacy appeal. Still focusing on privacy isn’t a bad thing. I’m curious to see how much Helm can innovate on this hardware platform going forward.
