We need to stop using Fastmail as the go-to example of email done right.

They recycle your email address once you stop paying. A great black hat technique is to gather up all Fastmail addresses you can find in the wild and poll them to see when they're available again. Then launch social engineering attacks from them to impersonate the previous user. I found this out the hard way.

Amateur hour. At least DHH has said hey.com locks your email address for eternity, the only right way to do it.

In my book "email done right" is using email addresses on your own domain.

Using @fastmail.com makes no sense to me. If being locked into somebody else's domain is acceptable, then Gmail is a better deal.

And if you still insist on using somebody else's domain, then all bets are off after you stop using that account. First of all because the service provider can always change their policies. Just like how when Yahoo announced in 2013 out of the blue that they'll start recycling usernames.

Fastmail is a paid service, you own that username for as long as you keep paying. The notion that you can take up resources, for free, for all eternity, is only sustainable for big companies like Google. And given how crowded @gmail.com is, I wouldn't put it past them to start recycling in a couple of years.

Using your own domain has the same problem. If you ever stop paying for that domain, or lose it for some other reason, someone else could purchase it and start impersonating your email.

How would you prevent that, everyone only uses a well known email provider domain that doesn't recycle addresses?

If someone takes over a domain they can also get https certificates, reset all their passwords and do all kinds of other things that are related to it if there's no second factor. That's just what you have to accept if you don't renew a domain.

Fairly recently there was a story about somebody getting locked out of their gsuite account because their domain dns was hacked (IIRC) so a custom domain also introduces an additional attack vector.

It's not the same thing, because by paying for it you have a binding contract and can switch providers too. If you don't pay for it, then you don't have a claim on that email address, as the terms of use can change at a moment's notice.

If paying for it is a problem, then pay it in advance for 10 years and setup automatic renewal. Also leave a digital will. You probably want to do that anyway.

How is that not the same thing as paying for an email account with fastmail?

In terms of the binding agreement, it isn't any different, except you can can move to different providers whereas if you don't own that domain, the service provider has you by the balls.

I use my own domains but if those expire someone else can gain access to my email? Don't forget that domain names can be valuable and may need to get sold if someone bids high enough.

So you mean to say "email done your own uncompromising way"?

You can use a custom domain with FastMail, and they aren’t reading your mail to target ads at you.

Yes, I am aware, I'm a Fastmail customer.

> they aren’t reading your mail to target ads at you

Neither does Google ;-)

> Neither does Google ;-)

This is true since 2017 : https://variety.com/2017/digital/news/google-gmail-ads-email...

Ok, so compare to protonmail or tutanota. The point is the price is high compared to other paid email providers. If you want to revolutionize email, you're not going to do it with a proprietary system that costs > $8 a month.

Oh hello no. Any source on this? What's the timeframe?

I don't get how they can complain about Google stifling innovation in an open system since 2004 and then close off their counter-offering by not offering IMAP.

I'm gonna stick with running my own mail server to get around Gmail. It's $5 a month, has a calendar, I can use any mail app I want to, I could import old emails if I cared to. I'm also sure I could get the "tracking pixel name-n-shame" feature they're so excited about if I took some time to write some in-depth filters.

The feature set would be extremely difficult to clone via IMAP + SMTP without being confusing. One of the major features is that you can change the subject that is displayed on an incoming email, and that it won't be changed when you respond (so if someone sent you an email with "no subject" as the subject, you can change it to "foo bar foo". They will still see "no subject" when you reply, and conversations won't be broken in Outlook). That'd be very hard to clone in an arbitrary email client that didn't know how to fix it. Another one is that every unique sender gets put into a screening folder until you tell hey where to put it (inbox/paper trail/feed), and that functionality can't be cloned in a email client (because they don't know how to force you to answer that question).

Even if they found a way to make it work now, every time they had a new feature they wanted to add to their web client, they'd have to worry about making it not break Outlook/Thunderbird/$RandomEmacsMailClient, because if it does people are going to miss emails and leave the service. And since their target market is people who hate the way email works today anyway, the target market probably won't ever want to touch IMAP again if they can help it. If they want to go back, all their mail can be exported and new mail will be forwarded to their next address. It seems fine to me.

> every unique sender gets put into a screening folder

I just set up a gmail filter to have everything skip the inbox. Adding exceptions is not so convenient but it works.

> I'm also sure I could get the "tracking pixel name-n-shame" feature they're so excited about if I took some time to write some in-depth filters.

Sure, _you_ could, but how many other potential customers can?

I'll admit, the lack of IMAP is a bit of an issue for me as well, but the other features may overcome it.

I've always been interested in running my own mailserver, but whenever I've tried, it's been surprisingly difficult (getting my IP accidentally blacklisted and stuff like that). What was your experience setting up your own server like?

I use runbox.com and am very happy with it tbh, basically has everything you seem to be looking for (but you don't have to self-host which I never want to do for email).

> It’s a closed system (no IMAP).

That's a huge bummer :(

Does Hey support custom domains? IMAP and custom domain support would make Hey very appealing as a Fastmail replacement after Australia shot its tech industry in the foot with new surveillance regulation.

Edit: Looks like custom domain support is planned but not yet available: https://hey.com/faqs/

Edit edit: timdorr beat me to it :D

Nope: https://hey.com/faqs/#can-i-use-my-own-custom-domain-with-he...

Or more significantly not yet. This is still in development and they've only sent out invites

Considering >50% of my email is just meeting invites I don’t see Hey as being an option right now. Throw in no IMAP support and I’m not sure if I would ever use this service.

Who is this intended for? Genuinely curious.

Plus the weird brag(?) about "we don't use artificial intelligence, we use human intelligence".

I dunno. Seems like a marketing-first ploy to get $99 from a lot of people who aren't looking too closely or who don't really use email seriously to begin with.

Or don't like current email clients

IMAP support is completely pointless for most people, but most people like privacy, and most people have to use email.

I use a calendar and email as two different applications. But I see the two have been tightly linked for decades. What is the workflow that demands this integration? Some clients will automatically create events from emails. But don't you get many more emails with dates in that you don't need to schedule for compared to the events you'll attend? Or there's the unified alerts when an important event is upcoming and when an important email arrives. But years ago we didn't have OS-level notification APIs like we do now. What am I overlooking that unified mail+calendar does that isn't available when they're separate?

It's convenient.

Example. I recently scheduled 35 podcast episodes with 35 different people.

Each one required me to propose various dates and times to folks, and lock in specific times that didn't overlap. This meant opening my calendar and inbox side by side to see what's available while writing an email to someone to propose a few times.

Then for locking in a specific time, gmail will intelligently auto-complete the calendar event notification email address of the person you have open so you can send them a calendar invite link quickly on their end.

Without this tight integration there would have been a silly amount of copy / pasting.

Sounds like you could really use something like https://calendly.com/.

That would require mucking around with another external tool. Google's calendar works quite well.

It's one of those things where everything just works with the least amount of set up on my end.

In this case, this particular tool saves a tremendous amount of time in doing what you’re describing. I just started using it last year and was reluctant to, and it really does act like a little scheduling assistant.

Companies with any significant amount of employees use email as the back end of calendar synchronization. Think managers with 40 meetings a week.

I was starting to get intrigued reading the article, but inability to import old mail (and inability to export backups via IMAP) I'd a definite deal breaker for me.

But you can export in mbox format.

Hadn't realized that, admittedly I kinda stopped reading after the no IMAP, no import. Thank you for correcting me.

I'm still struggling to understand the decision making here though: this is a premium product pricing itself above the various GSuite/Exchange/Online/Fastmail, and demanding a significant time investment from users (for the whitelisting). There is of course nothing wrong with that, and I really like the idea of a premium, well crafted email system.

However, the lack of archive import (and of sync'ing to alternate email addresses it seems?) seems puzzling here.

I'm not sure that the Venn diagram intersection of people who really care about email but do not require import of their current archives is a large enough market to go after.

I have a small-ish archive (15 years) and any service that wasn't able to work with is a nonstarter for me. I imagine while not the most common situation, we are talking about premium email here...

It also doesn't support custom domains, just @hey.com addresses:

> Can I use my own custom domain with HEY?

> Not yet, but eventually you’ll be able to set up [email protected]. Until then, everyone who uses HEY will get a @hey.com email address.

I think that some of the features they have are just not possible as a mail app sitting on top of another service.

I understand the frustration around no IMAP but by the sound of it the UI for Hey ("there's no Inbox, there's Imbox") implies that it might not make sense for this to be accessed in the way normal e-mail is.

Same, not impressed at all.

I tried sending a couple of photos to my Hey account and it rejected the email at around 20MB in attachments.

They claim 100GB of storage space, but what use is that if a 20MB attachment can't even make it through.

Have you used Basecamp for any period of time? I'm excited about Hey because I think it will bring the organization and clarity to email that I have enjoyed in Basecamp for a long time.

I love IMAP, I just wish I could disable IDLE in macOS Mail Mojave or later. I prefer no interruptions, but can’t turn them off now.

Maybe HEY will kick some of these ideas back to the forefront.

Fastmail is based in Australia. Some use cases might have issues with that.