[mahmoudhossam]

I'm surprised there's no work being done on the UI on the roadmap, K-9 looks like an app from the "Ice Cream Sandwich" days of Android.

With that being said, K-9 is the best Android app I've seen for mail that isn't made by Google.

edit: it seems the work has already been done, thanks for the correction.

[BubuIIC]

Have you looked at the beta versions yet? The UI was overhauled over basically the last 1.5 years and the points mentioned in the blog post are the bits missing to release that as a stable version (IMAP IDLE mostly): https://github.com/k9mail/k-9/releases

I've been running the betas since more than a year without any problems whatsoever. (If you don't need push obviously)

[y04nn]

Thanks for the tip, the GitHub release makes a real difference compared to the outdated version on the Play Store. It's sad that there is not many reliable email client alternatives to GMail that works with custom server.

[JohnTHaller]

You can subscribe to the beta channel on the play store, too

[out_of_protocol]

Is UI any different from current stable version?

[cketti]

Yes. This blog post contains a few screenshots:

https://k9mail.app/2020/06/01/Whats-Up-With-K-9-Mail

[alvarlagerlof]

I tried it from the Github releases page. It has definitely been updated a lot from the Play Store version.

[richardwhiuk]

I struggle to understand why there hasn't been a release for 1.5 years, especially with a ~full time dev.

I definitely struggle to udnerstand why IMAP IDLE is critical for a release.

Edit: Oh, probably a minSDK force - i.e. the cost of standing still in mobile development.

[cketti]

> I definitely struggle to understand why IMAP IDLE is critical for a release.

The functionality is present in the latest stable version, but needs to be reimplemented in the new version. As a user I would be upset if the app was updated and that feature taken away, even if it was only temporary.

[richardwhiuk]

Which leads to the question about why it was removed...

[rlpb]

Something to do with issues with newer Android no longer permitting the background connection to remain open for battery saving purposes, I think?

[zlynx]

I am not using the beta version of K-9. I was able to make it work mostly correctly on Android 11 by setting it to "Not Optimized" in the battery power permissions. That will allow it to run in the background. It seems to be able to use IDLE and to also run its periodic polling that way. Otherwise I would open K-9 and it would be hours out of date.

[Aachen]

What's a minsdk force? I know what minsdk is but the sentence doesn't compute for me

Edit: oh, randomly spotted this:

> A major factor was the API level requirement by Google Play. It required us to make changes to internals of the app in order to be able to publish updates via the Play Store.

I don't get the problem. Why not do a release? If google doesn't want the hard work and updates yet, fine, you can still tag a new release and people can use f-droid, download the apk, whatever right? (I'm using it via f-droid, thought most people would be doing that since it's an open source client with graphics from the 2009, the kind of thing you only use as foss fan).

[richardwhiuk]

In order to release a new version to the appstores, both Google and Apple require a certain minimum version for the app to be compiled against.

Historically, Apple gives you about a year before you have to compile against the latest SDK, and Google gave you a lot longer.

However, recently, both have become more aggressive about requiring new versions to be compiled against the latest SDKs.

Thus you can't do a "proper" release without moving up to the latest SDK.

Compiling against the latest SDK often breaks things, especially for an app which does a lot of things in the background.

This is part of the "cost of staying still" in mobile app development, which is both hard to predict, and can be very expensive.

[Aachen]

Does this also mean older devices are forced into obsolescence? Since this is minsdk and not maxsdk it sounds like it, but there's still many apps that run on my 2018 device with Android 7 (and even 2012 device with 4.4) so that can't really be.

[distances]

No it doesn't. It's about targetSdk, not minSdk.

[swanswan]

Well anyway older devices are indeed forced into obsolescence, at least OS-wise.

[rjzzleep]

I mostly agree, but I'm quite liking Fairemail [1]

[1] https://github.com/M66B/FairEmail

[Krasnol]

Tried to install it from F-Droid, didn't work.

Installed it from Github, it wanted me to pay for putting a colour on my account(??) and it will remind me of something something.

Nah man...don't do that if you actually want to sell a pro version. Just state it upfront and I will consider it. putting it in while I'm already adding an account and letting me read some long list of features I won't have if I continue right in the middle of the configuration is a dark pattern to me.

They lost a potential customer here.

[g_p]

I sympathise a lot with the developer here.

> Tried to install it from F-Droid, didn't work.

Developers get very little say in what version F-Droid builds and ships [1]. I think there are some very minor differences between the F-Droid build and the regular (open source) build when it comes to OAuth API keys - centralised mail providers often don't agree to the keys being used in those builds made by third parties.

> Nah man...don't do that if you actually want to sell a pro version. Just state it upfront and I will consider it.

Not really sure how much more the developer could do here to make it clear - the app description [2] states it contains in-app purchases (at the top). The 5th line of the description says "Almost all features are free to use, but to maintain and support the app in the long term, not every feature can be for free. See below for a list of pro features.", and there's a full list of pro feaatures lower down the description.

Not really sure if I would agree this is a dark pattern as such - I've seen far worse store listings that don't call out "pro" features until you start using the app, and discover 90% of the advertised features are paid.

Maybe I have more sympathy for open source developers, but I think it's getting increasingly difficult for them to compete with the commercial apps that are "free" (with data mining), or are paid. Looking at how regularly updated FE is, and how responsive the developer is, I imagine it could be a monthly paid subscription, based on how much time many people spend using email, and it would probably be more than worth it. Certainly looking at SV SaaS pricing, it feels that FairEmail is probably very much under-priced. And open-source too, so people really are just paying for convenience.

Comparing it with hey's email service + client at 99 USD/year, it's not easy for open source apps to compete, but it's in our collective interest for there to be good, credible open source options available.

[1] https://forum.xda-developers.com/t/app-5-0-fairemail-fully-f... - commenting on not knowing when F-Droid will ship that version.

[2] https://play.google.com/store/apps/details?id=eu.faircode.em...

[freebuju]

Fairmail is not meant to be a free-to-use product. I very well remember checking their readME and almost immediately recognizing this fact.

OTOH the "free-tier" is what attracts your interest as bait, probably accounts for the number of app installs visible on the store too etc.

[jorams]

The README has a single section about Pro features, and it starts with "All pro features are convenience or advanced features".

That definitely looks like the non-pro version is meant to be a free-to-use product.

[mgbmtl]

I use the free version and happy with it.

[berkes]

I've moved to SimpleEmail, which is a fork of FairEmail. Donated to both though. SimpleEmail ripped out the donation part in their fork, which is their right under GPL, but not a nice move IMO. Main reason for me is the fast pace with which FairEmail was iterating. Often with breaking changes. I like my email app to be boring. Which FairEmail is, thanks.

K9, to me, is just too much options and features. A lot of which needed tweaking and tuning, and were not something one could simply ignore.

[g_p]

FairEmail has (at least for me) reached a point of relative stability now - it's mostly iterative improvements and minor edge case/bug fixes.

Good example, a recent release [1] fixes a couple of bugs that look to have originated from upstream Android projects, and added some new support for removal of "tracking" parameters from Facebook URLs.

I'd agree with your comment that stripping out donations is "allowed", but "not nice" - I can only sympathize with the developer of FairEmail - it is clear he puts a huge amount of time into an app that has no "covert monetisation" like most apps.

I was never particularly happy with other email apps and privacy (especially not the commercial closed source ones which receive your emails on their own server back-end, but keep this part quiet in their description, just to make push easier) - for me, at least, FairEmail delivers the same or better, but all entirely on-device. There's even some basic learning-based on-device support for sorting mail automatically into folders (spam, FYI stuff, etc). That's for me the spirit of FairEmail - doing what others do server-side, on-device, without spying.

But it is clearly a challenge to make money from this, and I think (based on FAQs) that the developer has a struggle with those who think that everything should be free.

No relation to the app, just a happy user that likes to pay for open source apps rather than become the "product".

[1] https://github.com/M66B/FairEmail/releases/tag/1.1488

[da_big_ghey]

I don't know, the UI is good. I like the "Ice Cream Sandwich" look, honestly. I wonder if the time of an open-source project is best served by re-doing the UI just to keep up with trends.

[kgwxd]

I think it's rarely worth the effort, but eventually they're forced into it. I really hate the current mobile software ecosystem. Hope we get something like the xfce of mobile environments at some point where things don't really change for decades.

[genewitch]

It really is too bad that the Nokia N800/810/900 series of devices were too little too late, because having what is basically an ubuntu desktop on a phone-screen sized tablet was awesome. Maemo was the original "OS" and was replaced by "Mer" and if i am not mistaken, at least as of 2019 both projects still exist.

If i ever pick up a librephone, pinephone, or something that can be switched to another OS, i'd like to try it again, even though i don't like using phones in landscape mode anymore. Maemo raised the bar so high on landscape orientation user interfaces that everything else is a joke.

[gkya]

This is a very benign and positive UI change tbf. I say that as someone who heavily uses this app since many years and who is _very_ conservative w.r.t. software. I've switched to the beta and am really happy with it. Not just changed, really improved.

[daptaq]

> I'm surprised there's no work being done on the UI on the roadmap, K-9 looks like an app from the "Ice Cream Sandwich" days of Android.

TBH, that's actually a good thing. Material Design was a mistake.

[genewitch]

What is Material Design and why was it a mistake? It's mentioned as being a part of a "fork" called pep in another comment, and viewing the screenshots it looks like google hangouts does/did.

[jerome-jh]

No, do not touch the UI. If the UI is changed I don't pay! (half joking ;)

[kgwxd]

Too late :) I just updated to the beta after I read about it in this thread. Seems like more clicks to jump around accounts, I have 3. I don't really care though, it's still better than any alternative.