HYOE - Host Your Own Email

Update 2016-04-18:

I just did a quick search for docker images and there are email stacks ready for use. I am so glad a few brave hard-working email geeks have posted these images - I'm eager to try them out.

Update 2015-04-22:

Email is dead. Yes. I've said it. Phones are dead. Yep, saying that too. These communication methods of yesterday are soon to pass away...

Mobile. Mobile is all we need, for what I envision as a Human-wide communication everything-to-everything mesh.

However, there is much work ahead to realize this vaguely-described vision.

Thus, I will not be hosting my own email. Instead, I will simply be taking 1 step away from central servers in the fastest and most minimal way possible, using off-the-shelf tools today. This includes keeping email - but turning over the hosting to an email provider, and hosting the rest myself.

I'm even going to short-cut the thinking, and follow a recommended suite from a tech author. I would like to name and credit him, but alas I choose not to give this away. If you ever read this dude, thanks.

Oh yeah, I have a new best idea. Coming up in a future post...


Today, It's Only My Best Idea

Please do not be fooled by the title - you will not find a production-ready solution in this article. What you will find is an idea - that's it - just an idea.

Email Hosting is Scary Hard

I think configuring a good-enough email hosting provider stack is hard.

It seems to me that ordinary people leave it to the experts. Email is such the target for hackers and spammers. That's the scary part - I'm an average programmer totally afraid of doing it wrong and becoming the next hacked-open-relay-spam-bot-enabling dork.

But hiring the experts these days still requires signing up with an email hosting provider that's going to charge $/maibox/month, and the fees for SPAM filtering, etc. add up very quickly.

Suddenly that desire to provide a small organization with a few email addresses becomes cost-prohibitive.

And yet, I still want to be able to make this happen for minimally capitalized entities like friends, family, micro businesses and non-profits.

Why Not Google Apps for Business?

Gmail is truly an awesome web application, and I use it right now today. The SPAM filtering is amazing - no one else even comes close. But in my humble opinion, Google has lost it's way.

Full disclosure: I have 2 domains on the formerly "free" Google Apps platform, while jamesmcfarland.com is a paid Google Apps customer - a choice I made prior to the time they eliminated free.

Google has been Zucked into the feverish pursuit of Big Data in order to monetize my every online action. To this end, I believe they continue to increasingly tilt the entire Gmail/Apps experience toward funneling everyone into Google+ - which I want no part of.

Although Gmail/Google Apps is an excellent product, I find the prospect of being totally owned by their central position wholly unappealing.1

I'm no longer feeling the "Don't Be Evil."

Looking in the Wrong Places?

I don't see much out there on the net in the way of how-to articles for standing up some reasonably bullet-proof SMTP/IMAP. This is unlike firewall tutorials and config tools, which seem to be readily available.

Further, I'm sure there is a wide variety of stacks and an infinite number of opinions as to what makes up a rock-solid email hosting stack. Perhaps every sysadmin out there does it in their own special way - so a manageable set of best practices has yet to emerge.

I am certain I'm looking in the wrong places, and/or I'm probably doing it wrong, but today I came to believe that the conditions are now just right for standardized email hosting stacks to be born.

For sure, some propers are due to Owen Winkler of Red Alt - who wrote an article 8 months ago with a very similar idea to what I present here.

Enter the Three Amigos

Three oragnizations I'm currently comfortable trusting, and standing upon the shoulders of: Ubuntu, Docker and Digital Ocean.

I'm not going to waste time on explaining why I trust these 3 orgs - I've already decided this is my best idea (note: I didn't say original idea).

Realizing HYOE

Here's my plan:

  1. Learn Ubuntu sysadmin.
  2. Learn Docker.
  3. Learn email stacks.
  4. Use it.
  5. Share it.
  6. Bask in the epic limelight (or, not - maybe just enjoy the satisfaction of having created something.)

From where I sit at present, I don't see anyting but the smallest chance I can actually execute on this plan. I will have to learn everything; I know almost nothing. I'm old; and I'm (relatively) poor. I lack self-confidence; I don't have a past record of success.

But this is my best idea right now. It's an itch I want to scratch right now. And right now, the world is rich with inspiration. There's Ubuntu people, Docker people, Digital Ocean people, and the list goes on from there.

However, I'm going to risk calling out a few people just in case you might become inspired too:

This post could be as far as I ever get - but I'm publishing it anyway.


1. Yes, Jaron Lanier is blowing my mind right now - bonus embedded book recommendation: Who Owns the Future [return to passage]