Show HN: Swift Mail. Fastmail's modern mail standard delivered natively on macOS

This looks good, but I'm not clear why it's a subscription. I'm happy to pay for upgrades when a new version with new features is offered; and I'm happy to not pay if the new version isn't compelling enough. (Sometimes support for major OS upgrades is enough of a feature.)

I'm increasingly disinclined to feed the subscription-based software economy, especially for the personal tools I use.

Same! I like Sketch's and other app's approach. Keep the version forever, 1 year free upgrades.

Great work! Will def. try it out

Unfortunately that’s not supported by Apple on the App Store. Luckily Mac apps aren’t super reliant on the App Store but it’s still a huge issue.

I've purchased "Due" via the App Store, and it tells me I'm entitled to free feature upgrades for 1 year, shows a date when my unlock period expires, and offers a longer-term subscription. So it must be possible.

After the expiration are you still able to access the older version?

Yes. They seem to gate on a feature-by-feature basis, know when you bought the license, and you don't get access to the new features after 1 year (unless you're on their subscription plan). You do continue to get the latest app, which includes bug fixes, but the features released past expiration aren't available to you. https://dueapp.zendesk.com/hc/en-us/articles/360053244591-Wh...

That is so much more work than simply not getting the latest update. Imagine if sth like IntelliJ put everything behind feature gates instead of just telling you to download an older version…

I don't think it's their fault, it's the way that Apple have decided to force everyone to develop.

Apple want that reoccurring revenue from other people's work.

I believe Reeder is doing kind of this feature. They release every year or every other year a new version with new version name like Reeder 5. So, if you want an upgrade, you just purchase Reeder 5, but you can always download Reeder 4 if you have purchased it before.

I'm a fan of Sketch's approach as well, along with Panic's approach to Nova's licensing, which I believe is similar.

Unfortunately, as a solo developer on macOS, some of the more complicated approaches don't seem feasible from a maintenance and support standpoint—at least not initially. I hope to stick with the macOS App Store as well, which I assume adds additional complexity since that type of offering is not supported natively by the framework (to my knowledge anyway).

I would like to explore additional options in the future, but I don't see any major overhauls happening in year 1.

Yeah -- I was confused too. Is the subscription just for the client, or does it come with an email account?

The subscription is for the client.

While Swift Mail is a Fastmail Platform Partner, it is not affiliated with Fastmail (or others) and does not offer email services directly.

Most macOS email apps do subscription pricing these days. Airmail, Spark, Mimestream… I wonder why that is.

Lots of money

Apple encourages it heavily and even hosts workshops about how to do subscription apps.

Apple charge you a subscription just to develop for their platform.

> I wonder why that is.

Money.

Apple gets 30% off your app purchase and/or in-app purchases, while it gets 30% from your subscription for the first year and 15% for the subsequent years.

So, if you think your app is worth 10$, you lose 3$ and the ability to make more money and to support its development (if you do, they have to be in-app purchases, like extensions, etc. and again 30% is gone).

On the other hand, if you sell it with a 0.83$/month subscription (= 10$ / 12months, just to be "fair"), the first year you lose 3$ but you gain later 1.5$. And btw, I don't even think 0.83$ is allowed (AFAIK the price points start with 0.10$ so you can't have weird numbers like 0.23$ or things like that), so you round it up to 0.89$/month or 0.99$/month and you gain even more money, which will be used also to keep the development going.

If you really want to make $10 from a user, you just charge them $12.99. No need for Subscriptions.

If you want to make 10$ you need to sell it for 14.29$ because the 30% is taken from the total. (30% of 14.29 = 4.29).

Anyway, that's beyond the point. You need to find a way to support your development also for the next years.

Wanted to add that Apple's fee is pre-tax, and Apple doesn't pay taxes (duh).

Additionally, the refund policy of Apple is so bad that they keep the 30% and the developer has to pay for refunds 100%, meaning it's a net loss and a lot of indie game devs got doxxed into insolvency because of this.

Wow I didn't know that. Thanks for pointing it out.

That's brutal.

It isn’t true.

https://news.ycombinator.com/item?id=23995750

> Additionally, the refund policy of Apple is so bad that they keep the 30% and the developer has to pay for refunds 100%, meaning it's a net loss and a lot of indie game devs got doxxed into insolvency because of this.

That is not true.

https://news.ycombinator.com/item?id=23995750

Over all, I like how it looks, but the price seems high. $3 a month is almost as much as I pay for email service.

Plus it seems weird to double drip. I’m already subscribed to you for the email service and you want me to almost double that price for a client to use it?

Will the existing Fastmail app be deprecated to force folks to the new subscription app?

The price is a deal-breaker for me. It's borderline insulting, having to pay a subscription fee for an already-paid email service. Were it a value-add, it'd be compelling.

-- edit --

Turns out it's not made by Fastmail, it just uses "Fastmail's standard" of JMAP. I'm still not going to pay a monthly subscription for it; a one-time fee would be much more palatable IMO.

this isn't a first-party app for Fastmail, at least afaict. I had the same initial reaction before realizing this.

(I still have the same reaction to the pricing... but at least it's not double dipping by Fastmail themselves)

Could be worse, it's half the price of Mimestream...

Which is a fraction of Superhuman.

Is this actually made by Fastmail? It doesn't look like it.

I honestly think removing Fastmail from the title would be waaay better for everyone.

Some people even wondered if an email account is included in the price... (because of, ... Fastmail in the title).

Anyway, they only refer to JMAP which was created apparently by Fastmail. It has nothing to do with the company, besides using their open email API protocol.

$3/month? Seriously?

Does nobody here see the irony of being a community of people who make their income working in software that refuses to pay even $3/month for software? Ditto for the universal hatred of ad-based business models.

This is why the consumer market for software is basically non-existent. Consumers value their time at $0 so they get free tools subsidized by intrusive ads or expensive hardware.

Businesses make more rational decisions with their time and resources. To the developer of this cool product, I recommend targeting B2B instead.

Subscription models are cancer. Sell me a license for a flat cost and stop wasting my time.

Where do you work?

How does the business that pays your salary make its money?

Do they pay you for your efforts on a subscription based model or a one time-payment + ownership model?

I bought my house. I bought my car. I bought my laptop. I bought my servers. I don’t need to pay for them every month.

In broad terms I am tired of having to make a paycheck every month and having a long list of companies wanting to take a slice from it for “making my life magically better”. Hidden price hikes, dark patterns to cancel, etc.

You haven’t addressed any of my questions, but I’ll play along.

Im also tired of having to pay my employees a slice of my revenue every month for “making my business magically better.” Hidden salary demands, dark patterns to fire them, etc.

I own my car. I own my laptop. Why can’t I own you?

I guess what I’m saying is, why do you think the code you write for your employer deserves to be sold on a subscription model (salary) —- yet you don’t afford the same opportunity to indie developers?

Wait this doesn’t make sense. They own the code I wrote for them. The ‘subscription’ is for future updates of that code. They can keep the version and never update. If I quit they still have what I wrote.

And they pay you 10,000X the cost of $3/month for the ability to own that code.

But it seems you still aren’t following the hypocrisy inherent here.

If this one time payment + ownership model is so great for everyone as you say, why aren’t you working exclusively as a project-based freelancer? Why hasn’t your employer fired you the minute you finish every project?

Is the company that hired you stupid? Are you taking advantage of them?

Or does the subscription model for software only make sense when you’re the one selling it?

Your house and car also have monthly subscriptions that come with them. Electricity, gas, etc.

Probably not a good comparison.

Gas is not a subscription. It's a recurring consumable, but when you buy a gallon of gas, or a cheeseburger, you bought an object and you own that object. You didn't rent the momentary use of it. Electricity has only a token subscription component.

The subscription aspect of a house is the property tax, which does blur the definition of ownership when it comes to land, if the government always has the ultimate claim.

Still they are perfectly apt comparisons. Consumables, even recurring ones, are not subscriptions unless you happen to subscribe to a service to provide those consumables. Even then you still actually own the delivered product and the only thing that stops when you stop paying is the service. They don't take the cheeseburger back, you just don't get a new one.

HP printers are now a subscription. They do retain control over the object and even the already-delivered ink stops working.

And it's an outrage, not reasonable. Artificially turning the seat heater in a car into a subscription is an outrage, not reasonable.

For most software, there is no reason it must be a subscription.

None of the value of updates and the treadmill of maintaining compatibility with other software, and the developer's wish to collect recurring income changes the fact that the software itself, if it functions today, can function exactly the same forever.

Maybe it will need to be run in a vm to provide an entire preserved world to go around it, maybe no one else will accept the files it produces any more after a while, and maybe you don't want to do any of that, but those are separate issues.

You could address those issues some other way like by writing a converter or something. If that's impractical, then you will surely buy new versions voluntarily, and there is no excuse for forcing you by breaking your existing things and essentially holding your life hostage all day every day.

The guy making the mail client is not being retained as a full-time employee for his expertise by the people buying his mail client. It's not at all a similar relationship.

But to answer your question - I work on a variety of internal projects for a company that moves freight. Every year I have to justify my salary by either being budgeted to continue supporting a product or by moving to a new product undergoing active development. I'm hardly writing one product and then receiving a unending revenue stream from it.

Quite simplistic view of the world. Of course when the company or guy selling the stuff has a rough sales period, he exits and your software that you love is over.

Or you can pay subscription, stable revenues, everyone is happy.

Or, the company closes and you keep using the version of the software that you bought X years ago for $30 and were perfectly happy with. If you want to upgrade it's on the company to continue improving the product with new features that are worthwhile investing in, rather than adding stuff no one cares about, with forced upgrades, or worse, doesn't improve the product at all, and you're stuck paying everything month for the same thing.

Which scenario would you rather have?

A: $3/mo

B: $72 with a year of free upgrades

The latter, easily.

So you would pay what equates to $6/mo to have it all as a single charge once a year instead of monthly?

Its about "owning" and you can use the client for longer than this time.

This is why i love jetbrain, ITS a subscribtion, but you have a fallback licence.

Most things are flat payment for the thing plus a support window (aka warranty period).

Just let me own it as is, and decide if I want to pay for new features/updates later. Plenty of software is sold under this licensing model, and it works.

It feels like this product isn't a good fit for people who have a free mail client that they are happy with. The probably describes a large portion of HN readers. But there are people who will be happy to pay $3 per month for something that does exactly what they need.

> Consumers value their time at $0 so they get free tools subsidized by intrusive ads

I mean, I just use Thunderbird, but I'm sure there are ways to torture myself if I wanted to.

100% reliable contacts and calendars and email for the last 20 years.

So much criticism.

People, have fun developing stuff for free or for little to no money while having to support a shitload of OS versions and possibly multi-device, multiple framework versions, bugs, people sinking your app on the App Store if you don't react or whatever.

3 dollars per month are nothing IF the app is useful. Email is literally the thing you use the most. Why 3 and not 1 or 5? Maybe 3 dollars are just enough to keep people who just criticize you out of your customer base, which is maybe more worth, considering that nobody installs an app with 1 or 2 stars.

That hits a little close to home. A couple months ago I built a tool to transfer my playlists from Spotify to YouTube Music, and I put it on github. Since then it's really taken off, I usually get a few stars a day, and I've put a bunch of work in both fixing issues that other people had and also just doing support for users having problems.

The thing that is funding the development of it at this point is my slow-burning rage at Spotify; the more people I can help get out of Spotify the more wrapped in wings of vengeance I feel.

Well, I personally am happy that you found what motivates the development of your app :)

However, some people still want to make that ugly thing called ... money... :)

When that's the case, you really need to do your best NOT to let your app sink on the App Store - when that happens, you can say goodbye to your income (be it passive or active), unless someone really gives it a try because there is no other solution, etc. etc.

I'm agreeing with you that there is so much effort that goes into software beyond just the code. Just going from "something I can use" to "something everyone can use" has been a huge deal with my little Spotify-to-YTMusic app. I wrote it in an evening and transferred all my music. Since then I've probably put in 30 hours refining it, fixing it for other use cases, and supporting people trying to use it.

I am, thankfully, in a position where I can do that in my spare time.

If someone gives me an email client that can handle gigabytes of email with a breeze and provide near instant full text search, I'd dish out $3/mo almost instantly. The current state of most email clients is sad to say the least.

And if paid annually it's just $25.

Does this work offline? Fastmail’s first party iOS app has such an amazing lack of offline functionality that I’m seriously considering canceling my Fastmail subscription. (Also the lack of threading is getting old.)

FWIW I’ve been using native iOS mail with Fastmail forever and it works perfectly for me.

Yeah good question. I do a lot of email on planes and travel where there is no internet. In fact I refuse to use any software that isn’t offline first these days.

I was a Fastmail customer for many years but moved to apple’s iCloud+ when they introduced custom domain support as that works entirely offline on all my devices. Also that gives iPhone/mac integrated contacts and calendars with zero hassle that actually syncs to everything properly. It’s cheaper per seat as well (£25 a month for that plus Apple Music for 6 people a month). No brainer.

Swift Mail currently offers partial offline support.

It doesn't provide full offline support by keeping a complete cached copy of all mail data on disk, as an IMAP client might. Any content received from the server will remain accessible offline and during subsequent app launches.

I would be happy to pay $10 for this but definitely not a monthly subscription when Mail.app is free.

Is JMAP the base for an email experience revolution? Wonder if we're ripe for one.

For a long time, The Bat! was one of the best e-mail client for its times in the 2000s. Loved it's ticker-style notification bar back then. Sad to see that this piece of software has deteriorated over time and not evolved to support Mac/Linux. Next, Thunderbird emerged as an open alternative that was stable for a time. When GMail took over, few innovations came for email with its own Inbox being a starting point, followed by Superhuman and Hey. Is there room for more? Or are folks too attached to the traditional form of email?

Apologies for the website bandwidth issue! The response caught me by surprise.

Working on this now & adding a few links below for ref in the meantime:

https://apps.apple.com/app/apple-store/id1595671863?pt=12263...

https://twitter.com/SwiftMail_io

https://mastodon.social/@SwiftMail

mailto:contact@swiftmail.io

The UI looks almost identical to https://mimestream.com/

It also looks nearly identical to Fastmail's web UI.

Because both are designed to look like Apple Mail (the MacOS native app look is a selling point)

I thought it was Mimestream until I looked deeper and realized it wasn't.

Is there a way to unlock the custom JMAP server option? Currently it says coming soon - is the feature actually missing or is just disabled due to being experimental, and if so is there a secret keyboard shortcut/etc to enable it? Been meaning to play with JMAP but the lack of native clients thus far has been a problem.

No, it cannot be enabled within the client.

It was disabled pending user feedback to allow private external testing of the feature while the app itself was open to public testing via website link.

This was to ensure a high quality, two way support channel could be established when enabling the feature for external testing. (It also ensured Test Flight data & support channels would not become polluted with failed attempts to sign in with gmail or similar)

A mail client would need to be very very good indeed to get me to stop using the native Mac client. It's fast, stable, and talks to both IMAP and Exchange.

Congrats. Also on not collecting data! No trials?

Thank you! 3 and 7 day trials are available.

I didn't see those lines somehow

It is very interesting, but there are plenty of stoppers for me:

1. Subscription model. I will be willing to pay even $250 for a good email client, that is superior to the native macOS client. If it really adds more features that I need. Meaning, I can use mostly all the email related features offered by Fastmail. But I am not willing to support subscription models.

2. No iOS client. I use email on macOS and iOS. Not having support for iOS is a dealbreaker. I want the same client.

3. No trial. As it seems like. I actually tried to download the app. There is a Sign-In page with a disabled Sign In button. I guess I need to subscribe to try it. I can spare $3, but I don't want to support subscription models. Seriously, give me an option to try it for 14–30 days, I will pay for it $100-$200, and you can ask me to pay you again for an upgrade in 3–4 years.

4. The main reason I don't want to try it, also because I am sure, there are probably some nice features, but at the same time, there are most likely so many issues with integrations between other apps. Like links from Reminders to the Mail client, or from Notes to the Mail client. And based on the roadmap [1] it seems like this is mostly MVP product.

--

[1] https://feedback.swiftmail.io/roadmap/roadmap

Actually, spoke too soon. There is a trial. 7 days trial for annual subscription, 3 days trial for monthly. But I was right about that being MVP/Alpha product:

1. No reply button to email. You can only compose a new mail message.

2. I have received new mail, got notification, but don't see it in the list (inbox). You need to restart the mail client to do so.

3. No images are displayed in the message body.

4. The interface is very not-intuitive in some parts yet. Like in the Threads - to expand the message you need to click a tiny button on the right, clicking on the message or on the body does not do anything.

Anyway, I am excited about this email client. But I don't see how I can justify paying for it yet. It is good maybe for read-only mode, but if I don't replace mail app, I don't see how it is much different from me just opening fast mail in the browser.

A few of these are not correct.

1. Reply button is below the sender's profile symbol in the thread message header.

2. Thank you for the feedback on activating the notification. This should work as anticipated but it's been quite some time since I received feedback related to notification behaviors so I'll certainly take a look.

3. Images and remote content are blocked by default (similar to Mail.app); clicking the 'photo' button (next to the Flag button) in the message header will reveal these images.

4. Expanding threads by clicking in thread message header white space would be a welcome addition and feature request. I'll add this to the planned feature list. Expansion via body content click will require a bit more investigation, however.

Appreciate the detailed feedback!

These are just a few issues discovered opening the app for 5 minutes. I wish you and the product all the best. And I will be happy to try it again later.

But at this point, I just don't see myself paying a subscription for an MVP/alpha product, with so many visible bugs, and not working (or not intuitive) basic features.

- Clicking a button for every message to see images is a weird decision, to be honest. I understand some people are more paranoid. But both Mail.app and Fastmail web client provide features for privacy to hide your IP for loading images. There are I guess some WebKit context menus, which suggest that I can download the image or open the image in a new window, and when I click it - nothing happens.

- You are right, there is a reply button. Just not where I expected it, considering I was expecting the same behavior/view as Mail.app. But even that - another bug / unexpected behavior - the reply email (From) chosen first from the list, but not the same email account that received the email. Say I received an email on support@example.com, and when I click Reply - the From account is selected to myaccount@fastmail.fm.

- I don't see any ability to attach inline images to messages. Drag and drop of the image from Finder to the Compose does not work. I see that I can attach documents to the messages, but don't see the ability to create inline messages.

So at this point. I already wasted a trial period to see that the state of the app in a very early alpha/beta. I would call it alpha because I believe there are many basic features are still missing. And it is definitely beta because some basic features don't even work, like lists don't update when I receive new emails.

I am sorry. It is just too early. I can't try again later in a few months, assuming you will release some updates soon. Which is very disappointing. I will give it another try, when I see it again, maybe mentioned somewhere else.

Really wish you actually had a TestFlight first to build the app to v1 and collect feedback. I do feel like there is a good potential of the app. And despite even that I really hate the subscription models, I went ahead and gave your app a try.

2. For 3rd party clients you cannot have a mail client without an intermediate server.

Consider a mail service that only has IMAP-IDLE [1]. iOS's mail does not support that.

So ANY 3rd party to be able to deliver notifications you either have to go through Apple's push notification service (or ask for Local push connectivity [2]). But regardless, it implies that in order for you to trigger the notification (either via Apple to the device, or directly) you must thus know that a new email has popped up. The only way to keep track of this is to use an intermediate server that maintains connection with IMAP-IDLE, and thus needs to know credentials.

[1]: https://en.wikipedia.org/wiki/IMAP_IDLE

[2]: https://developer.apple.com/documentation/networkextension/l...

Can you just use the background refresh, and fetch for new emails like every 5-10 minutes or something? Email is not messaging, don't really care if I receive some email 5–10 minutes later, unless I am expecting email (like email verification), than I will be opening the mail app by myself.

Problem with that is that

    earliestBeginDate

means that you can request a no-earlier-than begin date. iOS does not guarantee it will run ON that date, or within x time-units. It merely provides a minimum delay until it MAY get scheduled.

Sure. I understand. But that is something I can leave with. Even just once a day pull the email for me and notify that I have new emails. If I expect anything right now, I will open my email app and refresh it.

I don’t get notifications when my real mail arrives, I just go every other day and check.

Thanks for following up below to update some these points. Just want to touch on a few, if I may:

2. An iOS client is planned going forward. The first few iterations of Swift Mail were actually deployed to both iOS & macOS, so there's quite a bit of shared platform code in place already when the time is right. I have a visionOS build up and running already as well, which I think bodes well for that effort.

3. (Trials do exist, as you mentioned below)

4. Swift Mail does provide support for mailto link handling, and it can be set as the default macOS mail client as well. I'm not familiar with linking from Reminders or Notes directly to the Mail app, however, so can't speak to that workflow at the moment.

Any and all feedback is appreciated & encouraged though so please don't hesitate to reach out as needed.

Note - that roadmap is not valid and was meant to be deactivated long ago after I cancelled my subscription during the beta (great experience but couldn't automate release notes with Xcode Cloud). A replacement roadmap and feedback portal should be back online very soon at that same link, however.

My number one requirement for any email interface is that it easily shows the "conversation history" of any contact I'm seeing. This saves me so much time, with the current software I use. I don't love my current software, though, so I'm always on the lookout for new options. Is this a planned feature?

Bandwidth Restricted

The page you have tried to access is not available because the owner of the file you are trying to access has exceeded our short term bandwidth limits. Please try again shortly. Details:

451 Actioning this file would cause "swiftmail.io//" to exceed the per-day file actions limit of 80000 actions, try again later

Thank you! The response caught me by surprise. Working on this now & adding a few links below for ref in the meantime:

https://apps.apple.com/app/apple-store/id1595671863?pt=12263...

https://twitter.com/SwiftMail_io

https://mastodon.social/@SwiftMail

mailto:contact@swiftmail.io

I like the look of it but I don't know what JMAP is and what the advantage is over say IMAP / SMTP.

It's a reworked email protocol, created by fastmail. Basically email over json (send + receive) instead of imap+smtp.

https://jmap.io/

Got it. Fuck that. JSON has no part in an email stack. Have we not learned from our failures as a species.

Edit: I've got 99 problems and 98 of them are JSON on a regular basis. Poor parsers, problems dealing with numeric values, terrible terrible schema support, poor encapsulation support, large things attached to it in base64 (difficult to stream on most parsers), difficult to read without external tools which actually tend to mung or fix the problem you are trying to see (jq does this), completely arbitrary and random metadata jammed in everywhere by everyone trying to make it self-describing, shitty enapsulated types i.e. ISO 8601 date in a text field rather than a principal type.

Urgh kill me. It's a hammer made of poo and no one knows any better any more. It's the PHP of wire formats. The COBOL of representation encapsulation.

Edit 2: the only positive is it's probably less bad than CalDAV but that's not even supposed to be part of an email stack is it? Everyone has outlook brain.

I do not concur. JSON might not be perfect for e-mail (it isn't?) but it's great for a huge range of use-cases.

It's a great achievement of our species.

Why is JSON inherently unfit for an email stack...? It's just a transfer format. If it's designed well, it shouldn't matter if it's JSON or CBOR or anything else.

It's actually pretty great (for submission and reading)

Yeah json is primitive, because it is (was, but still is) just JavaScript errr ECMAScript 3.

All your points are valid, but it’s a lot better than xml’s api, confusing whitespacing/attr/element, and huge complexity.

Simple is better, and json is def. lacking in some points, but in general it’s easier to work with than xml (or at least getting started with)

Hi,

thanks for this nice App. Will have a look at it.

There is another cool app coming up. It's called FMail2 http://fmail-app.fr

Maybe it's also interesting to compare it. Just to let you know about.

Getting a Bandwidth Restricted error:

Bandwidth Restricted

The page you have tried to access is not available because the owner of the file you are trying to access has exceeded our short term bandwidth limits. Please try again shortly.

Thanks! This should be resolved soon.

Looked at it and debated leaving Spark Mail but the price throws me off, and the lack of IMAP support to still be able to receive my junk Gmail is a deal breaker.

Aside from the obvious FOSS arguments, what are the big positives to JMAP for me as a user over using a client that supports Exchange?

It's HTTP-based and every request is stateless compared to IMAP, so it should tolerate unreliable connections better. IMAP in contrast is stateful and the process of establishing that state requires more round-trips, so connection establishment is slower.

Email snooze is, for me, a standard requirement for any email tool. I tried the free trial but cancelled.

This is way too expensive!

seems like you've exceeded the bandwidth quota of your cdn provider

Thank you! Should be back up soon.

Loving Swift Mail! Just installed last week and finally getting my email under control. Three email addresses from different services all together in the fastest email client I’ve ever used. Super smooth, clutter-free interface that I like opening every day.