Pricing 101: Price has nothing to do with the cost of providing the service, and everything to do with the value the customer puts on what they're buying.
The value you put on email is $10/year, so Fastmail doesn't represent good value. Other people obviously think its worth more. Fastmail are providing their service to those people.
I need to respectfully disagree with you here. In a free market, pricing does have something to do with the cost of providing a service because companies that overcharge would quickly lose market share to competitors, especially in a wildly competitive space like email.
Also, ten bucks a year has nothing to do with how much I'd personally be willing to pay for email. It is just my estimation of the amount you could charge people while still keeping healthy profit margins in this industry, however I don't have much domain knowledge regarding email and that figure could be very wrong. I'd love to hear from an expert.
...companies that overcharge would quickly lose market share to competitors
This is only true for companies offering equivalent products, and that is never the case in tech because there are so many intangible factors.
In this case, one of Fastmail's intangible benefits is that they're not Google. Maybe Fastmail's customers think that's worth $50 ... but it has zero impact on Fastmail's costs.
Bzzt. Notice how everyone recommended fastmail and nobody else? It's not a highly competitive free market. It's "hi, can anyone recommend a MacBook cheaper than the one apple make" time.
Free markets reductionism not useful with no serious competitive pressure
Email seems like more of a free market than most things considering the incredibly low barrier to entry. You can even host your own email server, you can't reasonably do that with a lot of the software that you use every day.
Just because a particular provider is good and cheap and has a lot of satisfied customers doesn't make it any less of a free market.
It just feels like a truism. Absent a formal regulation oversight any market is free in these terms. Were not comparing Japanese gold plated apples and trader Joe floor dropping oranges with mould on, were looking for competition in the sense BMW competes with Mercedes and Lexus. Google is giving away some thing of huge value, which is highly reliable spam free email. They "charge" your privacy which is beyond price. If you seek equivalent functionality the software cost alone for the one I regard as comparable is 350 dollars which as a purchase amortised out is $10 a month over three years but I have to spend another $10 a month to host it, and I still have a single point of failure. So price wise fastmail may be enormously good value or expensive depending how you feel: there are not a heap of competent competitive alternates. Maybe five? Sure, that's a competitive market in Herfindahl Hirschman Index maybe? But in practice it's not far off a duopoly. Google or o365, fastmail or proton. It's a pair of competing entities not one market.
As long as you don't want deliverability to Gmail or Outlook addresses. Though you can pay for Outlook deliverability it seems. They're very pinickety about allowing mail through from self-hosted or even low-traffic shared hosting IME [albeit limited].
Unused IP, set PTR (seems important) and MX, SPF and DKIM, although I'm not sure how relevant the last one was.
I can imagine that Gmail will block you if you try to send mail from an IP that was previusly used to spam.
Couldn't really work out why Gmail wanted to block me, yes obvs using SPF/DKIM/etc. - long trusted domain (15y), never any spam, whitelisted on client side. MS blocked it according to their third-party system because it was associated with another IP address [ie other IP was once used by same ISP] that had once had spam on it (wasn't currently blacklisted in any SBL).
There's no way to check before you build out the system, so unless you have an easy way to change IP and ISP then I'd be cautious.
Email is a highly competitive market with a low barrier to entry and thousands of companies offering similar services. It is nothing like hardware, so your comparison is bizarre.
By the way, I was asking for the point of view of an expert. If you are not an expert on email services, then I have no interest in hearing you speak condescendingly to me.
So, there are two separate-but-related products here.
The first is e-mail.
The second is an extremely hard-to-validate promise of future work, such that (a) cutting-edge security and privacy protections are maintained; (b) properly designed regular backups and redundant infrastructure ensure your e-mails won't be lost; (c) staff ethics, training and oversight will protect against insider threats; (d) protection against spam and phishing will be effective - but will not block legitimate e-mails; (e) e-mails you send will get delivered and not marked as spam or silently discarded, even when sending to the likes of gmail; (f) no e-mails sent to you will be silently discarded or otherwise lost; (g) if issues do arise, that they will respond promptly and competently; and (h) as webmail technology advances and market conditions change, they will keep up-to-date and stay in business.
There's no way for me to validate most of these things. Instead, we rely on a supplier's "reputation" - an amalgamation of their past performance, their visibility and their marketing.
There are many e-mail providers. There aren't many e-mail providers widely known as reputable.
In other words, Fastmail doesn't face much price competition within the market of "e-mail services recommended by users of Hacker News" even if they do within the market of e-mail services generically.
Once you add the words like Gmail or replace Gmail or instead of Gmail then no, it's not thousands of companies offering similar services. Running a service in Port 457 is the least of the offerings. It has to have spam detection as good as Postini, sieve, IMAP, utf-8 support, tls certificate management, a web portal, DNS and related domain management.
The killer is five nines availability. That's pricey.
I used to run mailsystems and maintain mailsystems code in the 1980s. I don't do it now because it has too many moving parts. I got a grandfathered domain into hosted by Google and it's worth a damn sight more than the apparent market force bottom price.
Hardware or software or service your competitors have to be comparable or at least bearable to apply price pressure. I do not judge fastmail as price competitive because they (like apple) are competing in quality. Not price.
The value you put on email is $10/year, so Fastmail doesn't represent good value. Other people obviously think its worth more. Fastmail are providing their service to those people.