Our plans for Thunderbird on Android
blog.thunderbird.net
blog.thunderbird.net
Thunderbird isn't acquiring K-9 to kill it or merge with an additional product, they're buying it so they can increase its development speed and then put their name on it. That seems to me like about the best acquisition goal that they could have -- basically saying, "what exists is good, instead of building our own thing let's just pour resources into what exists."
It doesn't seem that conceptually different from what companies do with products like Blender, where sponsoring more development on an existing product is better than rolling an in-house version. The main difference being that K-9 is not independent now, the companies that do this with Blender don't end up owning Blender.
So with any acquisition there's some cause for caution, but overall I am inclined to be pretty optimistic about this. I use K-9 today, and I would like to see K-9 get more resources and to get more parity with other clients. I'm a little bit hesitant about my primary email client on my phone entering an experimental phase where its interface is going to change a lot. Again, concerns, but overall this seems like actually pretty good news.
THunderbird is buying or acquiring K-9? Isn't K-9 FOSS? The OP says only, K-9 Mail officially joins the Thunderbird family (unless I overlooked something).
> And cketti [lead dev] has already joined the full-time Thunderbird staff, bringing along his valuable expertise and experience with mobile platforms. [...] To accomplish that, we’ll devote finances and development time to continually improving K-9 Mail.
Acqui-hires generally discard the acquired organization's product; where does it say Thunderbird is doing that? It looks like they are embracing it.
There's no indication of IP transfer so it seems like this is not technically an acqui-hire, but not for the reason you're stating.
I personally see this as a way for cketti to get acqui-hired and maybe get away from K-9 Mail. The amount of hate directed towards him from the userbase (mostly justified, sorry to say) has been substantial and I can't imagine he really wants to continue dealing with that.
Hopefully Thunderbird will do another UI redesign and fix the mess that was made. I ditched K-9 after the new design came out two years ago.
Given that KMail is really the only true open-source non-janky Android email client it's incredible to me that people can be so uncharitable.
With the UX changes came long-desired fixes. KMail may be the only O/S mail client at this point with a fully working IMAP search (after Apple Mail screwed it up years ago). Plus they fixed a super-irritating power mgmt issue.
I recall the K-9 developer needing additional funding in the past, so this is great news for that project [0]. In addition, I'd much rather use "thunderbird" on mobile as an extension of K9, since it has thus far been my favourite mobile client. Work in the most recent years has been wonderful, and it's a pleasure to use (although I've vastly cut down on mail on mobile because hey, a keyboard is better).
Overall, this seems like it makes a lot of sense. Hopefully this will mean that the shift to supporting a mobile client won't detract from the good stuff desktop Thunderbird is doing as well.
Good news because it should increase the chances that the two products will live a longer life.
Possibly bad news because I'm one of the crowd that went back to K-9 5.600 because the new version destroyed the UI that was the primary reason why we picked K-9 over other apps. 100% self selection bias here. There was maybe an ongoing effort to offer the old interface as an option. I'm not sure Mozilla's going to invest into that. On the other side, the way I'm using Thunderbird is very close to the old K-9.
A wish: an IMAP server backed by Thundebird's local storage and K-9 as client.
thunderbird.net states at the bottom: Thunderbird is now part of MZLA Technologies Corporation, a wholly owned subsidiary of Mozilla Foundation.
> There was a time when Thunderbird’s future was uncertain, and it was unclear what was going to happen to the project after it was decided Mozilla Corporation would no longer support it. But in recent years donations from Thunderbird users have allowed the project to grow and flourish organically within the Mozilla Foundation. Now, to ensure future operational success, following months of planning, we are forging a new path forward. Moving to MZLA Technologies Corporation will not only allow the Thunderbird project more flexibility and agility, but will also allow us to explore offering our users products and services that were not possible under the Mozilla Foundation
> Note that Mozilla has nothing to do with Thunderbird anymore.
which fails to claim anything about the Mozilla Foundation or Corporation in particular.
> ... MZLA Technologies Corporation, a wholly owned subsidiary of Mozilla Foundation.
Mozilla the non-profit owns the for-profit corp who owns Thunderbird.
In their own words from the footer of https://www.thunderbird.net:
> Thunderbird is now part of MZLA Technologies Corporation, a wholly owned subsidiary of Mozilla Foundation.
> Who makes Thunderbird?
> Thunderbird is developed, tested, translated and supported largely by group of dedicated volunteers, plus paid staff. Thunderbird is an independent, community driven project. Therefore its paid staff, budget and fundraising are entirely managed and overseen by the Thunderbird Council, which is elected by the Thunderbird Community. Thunderbird development is made possible by funds donated by the Thunderbird community. (Mozilla Corporation, the makers of Firefox, and Mozilla Messaging no longer develop Thunderbird. But Mozilla still supports Thunderbird by hosting many of the Thunderbird resources.)
[1] https://support.mozilla.org/en-US/kb/thunderbird-faq#w_who-m...
Having decided to just look while writing this comment, apparently it was a couple million last year? Which doesn't seem bad.
I would love to donate to the browser, while my small donation won't make it independent from Google's money it might be a small step towards it.
I don't especially mind their political agenda, however that's not what I want to donate to, I have other political orgs I prefer.
Postbox is based on Thunderbird with a more modern UI, but its Windows and macOS only (does work in Wine).
I stick with FairEmail on Android for now. Its completely FOSS and in active development.
If you keep your desktop on 24/7 as many people do, then you might keep your email client (thunderbird) open in the background, checking for mail every 5 minutes so you know when you get new email. Since your storage quota is only 250mb, you delete email from the server automatically on download. Now suppose you want to take an email with you to a community meeting. You can print it out, or you can email it to yourself at a different address. Or if you were able to link up your phone with your computer's storage, you can just search up the email on demand, find related messages, and make changes to drafts for your community mailing list.
If you could connect to thunderbird with an IMAP client, then this would be possible.
However I see two ways to let my phone read old mail
1. Set up my own IMAP server at home and deliver email to it, then program some filter system to mimic what I'm doing in Thunderbird.
2. Access Thunderbird directory on my laptop over IMAP. That IMAP server would be on my laptop or (better) on some always on server at home that uses the same file format as Thunderbird. I would rsync it with my laptop.
I admit that number 2 is weird. Number 1 is the proper way but I have so little incentive to do it that I'll probably stick with my current setup forever.
I don't feel like setting up my mail client again but looking at the current design I'm pleasantly surprised. I'll probably give it a go as soon as the Thunderbird team and the K9 team have joined forces for an actual release.
Personally, it was the transition from Holo to Material that drove me off Android entirely. There was no such thing as dark mode at the time Android 5 dropped; you had to set brightness very low to stand a chance at reading anything in the dark, and with many (almost all?) apps still using the dark Holo, looking at and using the device was jarring. I switched to Jolla, and later to iOS.
I'm happy Android users didn't have to suffer through another platform-wide redesign in the meantime, but I personally don't find Material appealing.
The problem is that very few developers seem to read the Material Design guidelines on offsets and such and the offsets often look cobbled together. Not even Google themselves can consistently follow their own design guidelines.
To be fair, as long as I can keep basic features (i.e. an app drawer and placing my icons and widgets wherever I'd like) I'd be fine with a Cupertino styles phone as well, as long as the damn thing is consistent. Back in the Holo days everybody but the game devs followed Holo because it was quite easy to follow, there weren't a lot of subtle details if you used existing components. Looking back at screenshot it looked a bit clunky, but every app looking nearly platform native was a huge plus that we don't seem to get on Android anymore.
It actually made KMail usable with e.g. 5000 mail entries in a folder, whereas before 500 could cause it to lag on scrolling and be unusable at 1000.
Or does IMAP cover that sufficiently well, e.g. not requiring duplicating a ton of data? I haven't dug into the spec before, I've just always used it with "download everything all the time" configs, but I know it has remote searching and a few other things that'd fit reasonably well with a content provider.
Back in early android-days many email-apps actually added an android account for IMAP/POP accounts. Which felt quite awkward. And I suspect that it was trivial for any app to list your accounts and get all of your registered email-addresses.
In principle it... might make sense? A system-provided-and-presumably-more-secure login method isn't inherently a bad idea. But without a deeper commitment from them / a better way to actually trust it / more consistent updates from OEMs (lol, yeah right), apps really have ended up generally better and more trustworthy.
As far as "why": why not? Some apps do some things better than others. You can already do this with IMAP, it'd just be deduplicating the download cost / connections / etc, which is generally a good thing for mobile use.
However I recently needed an email app, I tried Fairmail but it has so many options, and its UI is very weird. I switched to K-9 and its perfect!, yes its not fully modern looking but its easy to use and simple to get started with. If K-9 stayed with the holo UI I probably wouldn't have used it.
Isn't Thunderbird using some weird file based system? Wouldn't an SQLite backed client be a better idea in general? What would the advantage of thunderbirds local storage have aside from maturity?
They do have a weird database format ("mork"), but that is for state tracking, address books, etc. and not the actual messages, I believe.
I guess it kind of depends on (sans acquisition) whether that old UI would have actually been offered as an option. If so, then yeah, this is probably bad news for you. If not, then... at least there should be some new features coming out, and maybe the eventual interface they go with won't be quite as generic as the new one.
But yeah, I sympathize, I'm just not sure that with any long-term OS project on Android that you'd be able to indefinitely get away with having no interface changes regardless of whether the project gets sold or not. Even though I do think it's reasonable to want to stick with the existing interface.
The justification given in the post is just non-information, because you could say this for almost any pair of software products if you define the terms fuzzily enough:
> The Thunderbird team had many discussions on how we might provide a great mobile experience for our users. In the end, we didn’t want to duplicate effort if we could combine forces with an existing open-source project that shared our values. Over years of discussing ways K-9 and Thunderbird could collaborate, we decided it would best serve our users to work together.
I don't want to say this is bad for K9, bit I still wonder what will happen to existing K9 installs.
If you install Android app X and, due to an acquisition, the app suddenly morphs into app Y which has a different branding, different UI, different features, a different backend and a different development team (but still has your data), isn't this exactly one of the things that the app store rules are supposed to protect against?
Umm, no? There's never been a rule about an app developer selling their app/company - that'd be pretty badly restrictive.
In any case, I've had installed apps have the name change as a part of rebranding before. It's a little bit jarring sometimes, but name changes happen. They're usually small name changes, but I don't thing taking "K-9 mail is continuing development with a new name" as "app becoming something completely different" is really fair. Unless the functionality and UX completely changes, I don't think it's fair to read a name change as something so negative that needs to be "protected against".
It'd be nice if there actually was a rule against something like that…
[1] The key features for why I didn't just use a simple file sync was that it supported bi-directional syncing of play counts, ratings and for podcasts also the playback position.
The problem is that while I don't need the podcasts syncing for regular podcasts (for which a separate podcasts app on my phone would be perfectly fine, and besides I don't listen all that much to podcasts that way anyway), I'm also managing my collection of radio comedies I've scrounged together from all sorts of places as podcasts in iTunes (by simply manually setting the media type to "Podcast" – luckily on Windows, where iTunes hasn't been split up into its components this is still possible), so that
a) they're all together in one place and not cluttering up the music part of my library
b) especially to get the nice at-a-glance listened/not listened/partially listened display, and
c) while iTunes allows setting the "Remember my playback position" for any kind of file, Rocket Player apparently supports remembering the playback position only for podcasts
I've got no idea how GoneMAD and MusicBee's file syncing behave in that regard, but a quick test installation of MusicBee showed that while I could get used to it as an iTunes-replacement for my general music library, it doesn't allow manually importing files as podcasts, meaning I'd have to resort either manually hacking the database, or setting up some sort of fake Podcast running on a local HTTP server in order to import new episodes... :-/
At least for now the old versions will keep working, and if some future Android version does break things, I guess I'll decide what to do about it in terms of a replacement at that time and not now…
Though I guess it does mean that my personal main bug in iSyncr [2] won't ever get fixed now, so maybe I need to resort to APK hacking after all…
[1] Which is how I found out about iSyncr + Rocket Player having been sold in the first place.
[2] After a file has been synced, it won't be synced again if any subsequent changes that don't change either of title/artist/album (or possibly file size, but due to metadata padding small metadata changes won't necessarily change the total file size) happen within 24 hours [3] of the original sync.
[3] I think this is trying to work around the fact that Android reports the file timestamps in local time, so due to either travel or even just summer/winter time a file might suddenly appear as outdated as compared to the last modified timestamp on the desktop. As my desktop with my music library, is in a fixed location, for me 24 hours is excessive, though, and 1 hour to cater for summer/winter time would be enough.
https://www.theverge.com/2022/6/1/23149832/google-meet-duo-c...
> Pretty soon, the Duo app will get an update that brings an onslaught of Meet features into the platform; later this year, the Duo app will be renamed Google Meet. The current Meet app will be called “Meet Original,” and eventually deprecated.
The way you present it sounds devious. But we're just truly trying to work together in the open source ecosystem the best we can and put our resources to their best use.
I’m missing a story about mobile Linux. Has this been discussed; is this something on the roadmap?
There are already excellent e-mail options for iOS. Right now there is almost nothing outside of the cli that’s usable on mobile Linux. Thunderbird would have a chance of being the main choice while helping adoption of mobile Linux in the medium-to long-term.
I understand it’s not realistic to expect anything anytime soon, but I do hope this is being discussed and that we will see a strategy for it.
It's the most viable and promising so far I think. I have too many directories for it to be usable without some new features, though (namely highlight dirs with new mail and more filtering capabilities). It's practically unusable, but granted I may have an unusual setup. If you just have a handful of active folders it might be great.
Didn't notice any surprising syncing issues so far.
I bought one a few months ago to support the project but never ended up using it. It's still sitting in a drawer.
Realistically, there's no hope I'll ever do anything with it myself. ;)
Nothing in this world is purely altruistic. If it didn't make business sense, this wouldn't be happening. Please surprise me and stay true to your word on this.
> The way you present it sounds devious. But we're just truly trying to work together in the open source ecosystem the best we can and put our resources to their best use.
We are all from the internet here, you know exactly why we are this way. Even if true, please understand why your words are perceived this way.
You're reasoning is that the Internet made you do it, therefore it's true?
However this particular deal sounds a lot like white labeling, don't you think ? :)
So sayeth every acquiring company about the acquired. I can think of very few that it held true. Maybe this can be added to the list, but only time will tell. I wouldn't suggest people holding their breath though
These aren't companies. It's a communty project and a FOSS project merging resources.
History
>These aren't companies. It's a communty project and a FOSS project merging resources.
Exactly how other situations historically have started. Lot's of FOSS projects "brought into the fold" to "shepard" along only to not get the resources promised.
> History
That's all you have? They are baseless then.
Bearing in mind it's not a 100% successful predictor, but is useful for figuring out how things are _likely_ to play out.
While yes Mozilla could theoretically enforce its trademarks and appoint its own team to start making more direct decisions, we've seen how that goes with various ex-GNU projects like libreboot or ASF projects where they don't have the community like OpenOffice
> If you install Android app X and, due to an acquisition, the app suddenly morphs into app Y which has a different branding, different UI, different features, a different backend and a different development team (but still has your data), isn't this exactly one of the things that the app store rules are supposed to protect against?
On my phone currently, this applies to the following apps:
* Element
* Free Now (European taxi app bought out by BMW/Daimler, formerly mytaxi, formerly hailo)
* Google Hangouts (Was installed as Talk)
* WhatsApp (now Facebook owned)
* A local cinema chain bought out by a UK + Ireland chain
* Pocket Casts (independent -> NPR -> Automattic)
I never understood why they dropped vector as a name. It wasn't a worse name than element. Riot was a terrible name choice however. Sooner or later it would have got caught up being used by people coordinating a protest or something and the press would have had a field day with it.
But your data has never left the matrix network or your own home server if you chose to have one.
But it did go from Vector -> Riot -> Element and various UI and branding changes along the way yes :) Part of this is also that the app was not really "finished" back in the Vector days. It only really came into its own during the Riot age. For example E2E encryption didn't yet exist at the start, and the underlying support in Matrix was missing as well. It's still a protocol and toolset that's evolving rapidly.
However the benefit of Matrix is that you can choose any client you want, there are several for mobiles.
Will that help? Not sure. Mozilla seems to have weird goals nowadays.
Anyway, I'm so fed up with empty terms like "experiences" and "excitement" all around in IT. But guess that's what happens when you let marketing dictate way too many things in software development.
What evidence is there for your version of events?
I currently use Thunderbird on desktop, but use a combination of Gmail and Outlook on Android. If Thunderbird picks up K-9 mail and turns it into a first-class mobile client, I'll probably use Thunderbird instead of the Outlook app. People like consistency.
Every few months Mozilla tries to overhaul it's UI and succeeds only to try again later. What are they trying to do? What's the real reason behind their slow descent into irrelevance?
Last time I tried Firefox on Android, it was a mess. Crashing every few minutes, unpredictable. I hear its a lot better now which is great! But why did they ditch the old one?
[0] https://caniuse.com/?compare=firefox+101,chrome+102&compareC...
edit: added link to caniuse.com
There was a time when Chrome was faster by an ample margin, but it's no longer enough to be an important distinction. Between the privacy respecting features, and plugin support on mobile - Firefox is the best choice.
If anything, it shows the damage of a browser-tech monopoly which Edge and Brave aren't helping with.
This is bad for Firefox because developers will eventually (and some still do) will simply stop supporting Firefox due it being so slow to catch up. Chrome already has such a huge market that a lot of web devs already don't care (e.g. Microsoft Teams).
What does this signify? It's the death of Firefox unless Firefox pulls a miracle here. Chrome is leading, all the devs are following, and they leading by leaps and bounds. Firefox is forced to catch up which causes frustration among users.
How long before Firefox gets tired of playing catch up and simply becomes irrelevant?
I am not happy about this. This is clearly bad but inevitable. The only other real competitor is WebKit which is restricted to Apple only. A huge market still. But Google's monopoly over the browsing market will only cause harm in the long run. Even though it will make devs jobs a lot easier.
Coincidentally, Brave believes they are privacy-invasive and don't support those features either (last paragraph in the article)[0].
Firefox, as with all other browsers aside from Safari, will remain with less than 5% market share for the foreseeable future. But I'd much rather continue supporting different browser engines than use a skinned Chromium to uphold the status quo.
[0] https://www.cnet.com/tech/computing/google-plan-for-chrome-c...
Mozilla, Apple, and Brave all disagree (as do I).
The overarching theme of your writing on this topic is that Chrome can just spit out Web 'features' at random and unless Firefox implements all of them immediately it is dying. I hope you can try to understand that not only is this not an accurate assessment, but maybe also that Firefox has value beyond being a copy of Chrome.
Mozilla needs to step up and find ways to expose such APIs though, because they are useful. Just saying "no" without proposing a solution is a sure way to lose. There's a bit of hope recently with the implementation of WebMIDI in Gecko.
It's telling that the only browser vendor that feels this way has the logo of one of the two most privacy invading companies in existence.
There's also Kiwi. It runs pretty much any extension available for Desktop Chromium.
But 20 > 0, all the same.
So why would anybody bother writing an extension for Firefox mobile?
I reluctantly agree. I think that the continuing relevance of email is largely a consequence of open standards and interoperability - things that Google et al are getting less and less keen on. As well as Thunderbird allowing me to have a copy of my emails on my desktop/laptop, rather than in someone's cloud, I think its good to have an open-standard success story to point to.
But in reality I use the Fastmail web client on desktop and Fastmail's app on my Android devices. Despite using Thunderbird for over a decade, nowadays I really only use it to sync a local email archive. I wish it were otherwise.
The category of FOSS email apps may not be dying, but it is small, and never far away from death. Having a large, established project with decent resources backing K-9 is definitely a good thing for the health of the category IMO. Personally I use FairEmail and hope that it will continue to thrive as well.
Wary of future problems? Sure.
But I don't think I could imagine a better set of news for my two favorite email clients. Well. Unless some big moneybags company started investing in DeltaChat. But let's do one pipe dream at a time!
- Full encryption integrated into the client for all data (e.g., I don't want Windows Search able to index the mail so someone has access when the client is closed/unencrypted; I also remember testing Thunderbird years ago and was able to go into the individual unencrypted .eml files [I think that's what they were] and read the messages without having Thunderbird opened)
- Full local sync with Android/iPhone (i.e., home WiFi, Bluetooth, or USB cable); it still amazes me that Thunderbird still doesn't have this built-in
- Xobni-like functionality (e.g., showing all emails and attachments to/from sender when clicking on an email, keyword searches); yes, I know the plugin is still available but it doesn't work properly with current Outlook versions (and no plugin like this exists for Thunderbird)
But if the upgraded K-9 also got other integrated Thunderbird functionality (e.g., calendar, contacts, tasks), that would be especially amazing as I could then move away from Google (which I'd also LOVE to do). For example, the Notes field in Google's base Contacts app is limited to 1000 characters. That means I can only sync Outlook to Android one-way (else I lose longer notes when they sync back).
To be able to move away from Microsoft and Google to open source... now that's a future worth dreaming about.
I was pleasantly surprised by the experience of installing and configuring Thunderbird on two new computers I bought for my parents. Pretty much seamless, and I kept waiting for it to fail it furiously downloaded all 15,000+ messages in each of their AOL In-boxes. Nope. It worked fine, unlike Outlook and whatever the atrocious POS client Microsoft is including with Windows these days.
Surely there is a space for minimalist office suites and email clients, written starting with a browser as a jumping off point, rather than a compiler? I guess I'm arguing for an Electron-based docs/sheets/email/calendar using modern software best practices, and great components. An email client should be a webview + sql-lite, no? And (for office docs) maybe with a more thoughtful file format, like a simple html subset. Does this exist? And if so where do I get it? (And maybe there are better "jumping off points", like VSCode, which is itself a specialization of Electron).
(I like electron fwiw, it's enabled me to have a desktop experience with applications that would've otherwise never come to linux)
Thank you for recognizing this. I admit I had a moment of doubt before pressing "submit". :)
I don't use thunderbird currently, but have for /very/ short amounts of time in the past. What seems bloated about it? For reading and writing e-mail it worked fine and other features seemed to stay out of the way in menus I didn't need to use.
I think NextCloud [0] has a few apps that do G Suite-type stuff, but I don't have experience using them. Etherpad [1] also seems really good, I've used it in a few Zoom calls and the experience was smooth. Personally, I'm looking for a Google Photos alternative that's stable enough for my parents to use.
Does it? There was just an article about claws mail on HN. Or on the office side, gnumeric and abiword seem decent enough.
I will say I do have a massive, single gripe about Thunderbird, and while it's a trivial thing it's also - and in part because it would be so easy to fix - massively infuriating: the default keybindings for archiving, deleting, marking read emails etc. are, astonishingly, single characters. I've lost count how many times I didn't correctly highlight the filter textbox and started to type a search query, which instead had the effect of randomly archiving, deleting and marking read the top N emails in my inbox before I notice. This is literally why we have control characters, so I find it incredible that this is not only the default behaviour but unchangeable without a plugin [1]. I still find the tool very useful in every other way but I can't believe something so obviously wrong has not yet been fixed after what must be at least 20 years of this software tool being available.
[1] https://addons.thunderbird.net/en-GB/thunderbird/addon/tbkey...
And I went back to webmail for 6 weeks because when tbird upgraded itself, the new OAuth (I think) stuff needed cookies, which I didn't allow so just couldn't connect and just kept failing without any explanation. That took half a day of chasing down.
And if I get a calendar invite it just tells me "all calendars are currently disabled. Enable an existing calendar or add a new one..." without 1) saying why or 2) saying how (and why would it not just use the one it's literally showing me while saying they're disabled).
And it doesn't recognise domains correctly, and it strips out inline attachments (yes, really), and fucks up formatting (turns html mails into messed up plain text), and it's caused me so much grief since its install, can anyone suggest a competent, reliable email client that has a focus on quality and not a focus on stupid UI crap that makes it feel like someone's personal side-project. I'd gladly pay .
As a user, I'd prefer a minimalist office suite that still doesn't have the browser as a starting point.
I think a browser is actually not a good UI for an office programs, because it really awkwardly competes with your own document model: HTML at its core still wants to be a document, except probably not the document you want to have. So you have to reimplement document APIs on top of already existing document APIs.
E.g., tracking and working with selected text is extremely important in office programs, but a browser already has built-in management of text selections: Except, the built-in management doesn't know which part is your chrome and which is your document, making you to either implement your own selection mechanism in addition to the browser's or handle all kinds of really weird special cases.
And so on...
That's thunderbird for ya.
Really the only web-browsery thing a mail client needs is rendering of the email itself (if it's not just plain text, HTML email is an extension!), for which, sure, it's a better idea to use an existing webview. But for the rest of the interface and logic? No.
FWIW, my email life improved massively when I left the likes of Thunderbird and KMail behind for the simplicity of mu/mu4e [1]. I hear similarly stellar things about Notmuch [2]. I'm never going back to an email client that even thinks about itself in relation to "office suites".
(For casual reading, and very rare composing, on a phone, k9 works passably well for me [3])
I strongly prefer Maildir. https://en.wikipedia.org/wiki/Maildir
As far as email clients go, Thunderbird is literally what you are suggesting. Thunderbird is essentially the Firefox web browser, with all of the browser front-end code stripped off, and an email client dropped in its place.
Nope. Totally not interested in anything Electron...
It might be interesting to try and port/rewrite siag/scheme in a grid to racket (or Julia...)?
It's the only "light-weight" office suite that comes to mind...
The fact that Thunderbird does stuff like this to me really weirds me out, despite mostly loving the program.
I feel like we've lost some of the point of open source. My rose-tinted glasses are probably in play, but I used to be able to use software, build it and modify it.
Today, the idea of being a "casual" contributor seems to be lost. Things need a huge build process, and some kind of management organisation. The emphasis used to be on contributing patches, not funding.
I think what those have in common is being corporate codebases that were "dumped" as open source. If you look at the equivalents that were built as open-source from day 1 - KMail, Konqueror, and KOffice - you find something that while still C++ is much cleaner and better structured, since they were always meant to be maintained by a distributed collaboration rather than a small team of l33t h4x0rs.
> Surely there is a space for minimalist office suites and email clients, written starting with a browser as a jumping off point, rather than a compiler? I guess I'm arguing for an Electron-based docs/sheets/email/calendar using modern software best practices, and great components. An email client should be a webview + sql-lite, no? And (for office docs) maybe with a more thoughtful file format, like a simple html subset. Does this exist?
I don't think browser-based is the right direction, and I actually think browser-based may detract from having good standard UI component libraries, since the browser itself tends to fall into the trap you mentioned - browser-native components are simultaneously not great, but imposing and complex enough to suck the oxygen away from alternatives. But I agree there's a lot of value in using existing frameworks and libraries to develop your applications, and definitely something that contributes to both OpenOffice/LibreOffice and Mozilla-family applications being as complicated as they are is the fact that they include their own one-off GUI frameworks. (In fairness Mozilla did originally make a legitimate attempt to make their GUI framework an actual first-class GUI framework that could be used for other applications with XULRunner etc., but those days are long gone now). Again this is somewhere where I find KDE family applications shines in comparison - everything is properly layered between the libraries, frameworks, and applications, and the libraries actually see reuse across a wide range of applications.
Fairemail is much better in that regard, which is why I stick to it, but still quite dated.
Other than that, while it's generally just... a bit ugly, it functions fine for me.
I've recently started maining Seamonkey as a browser. It has a better UX than all of the big ones -- especially Chrome but even Firefox isn't so great anymore. The Mozilla spinoff projects have lots to recommend them, now that Mozilla is more an outreach organization than a software development organization.
Spam is an everyday issue. Every time I use Thunderbird I use "Mark as Junk" and "Delete Mail marked as Junk in Folder".
I created an issue asking for buttons for those often used actions instead of having to go to tools, mark, mark as junk. 3 clicks vs 1 button press. It was closed, essentially saying that I'm just one guy and there is no interest in this. This was 10 or more years ago.
So excuse me if I'm not very enthusiastic about this. I personally tried K9 but opted for Fairemail, even with it being trialware. A proper email client on Android would very much desired.
But then again I would love slypheed or claws. Sylpeed-claws was the best mail client back in the 2000s, much better than Thunderbird imho. We had Gnu/Linux now we have to deal with Android/Linux. Because Google. Why couldn't we just build on Gnu/Linux and have the best "of both worlds", because monetary interests. Bleh.
Now I can look forward to Mozilla's abysmal stewardship and a bunch of senseless UI changes that aren't good UX, just because they're new, and broken workflows. I think I'll pass. I'm sure the maintainer agreed to an acquihire it for good reasons, good on him, but I'm not interested in using Mozilla products anymore.
It has remained beautifully lightweight (quite an exception for apps nowadays) and the biggest (yet perfectly tolerable) annoyance has been long lines not wrapping automatically.
I'm 9/10 sceptic about this cooperation and I'll be positively surprised if the future versions do not introduce lots of unnecessary bloat.
> Account setup using Thunderbird account auto-configuration.
Perfectly unnecessary.
> Syncing between desktop and mobile Thunderbird.
Duh, what exactly does IMAP do?
*shakes head*
Then give an option to link it with the firefox account so as to link the Thunderbird desktop and mobile! Just like it works for firefox now. :mindblown: (Obviously Isolate the email app + calendar portion and the RSS + IRC client part.)
That would be a killer app!
Heard of Betterbird? Yeah, a Thunderbird port.
I use Aquamail on Android and love it. Great feature set, works great with multiple accounts, lots of configuration options. It's not open source or free and I honestly do not care.
Back in the day it was something like USD 4. Now it's something like USD 25. Still worth it, and I don't mind paying for software if it keeps the developers in business.
Super excited to see cketti's work get a major FOSS organization behind it. Contrary to some other comments, I have actually been really happy with the new interface updates and I think giving K-9 a light thunderbird skin would be fantastic.
https://github.com/k9mail/k-9/issues/655#issuecomment-113165...
At least I don't predict K-9 getting worse and has the distinct possibility of getting better and getting more features.
I have 15 GB of mail across 5 email accounts and Thunderbird is currently sitting at 350 MB RAM. Rarely crosses 500MB, I think.
I am encountering some freezes and occasional crashes which are annoying, but on linux there is nothing better.
My fix for most annoyances was to copy mail locally, and run dovecot locally on the same box as TB (TB doesn't support standard maildir). I also added a wrapper script that does a VACUUM on all the sqlite dbs in the profile when starting TB.
With the above, TB has worked well for me.
Perhaps it wasn't DropBox but something else ... Perhaps I'm even making this up (although I don't think so). I've always wanted to know more about this promotion of a filesharing service in Thunderbird, so if anyone knows anything about this, please reply (:
AFAIR it was a preinstalled extension on Thunderbird from Debian.
So then I tried embedding the css in the email and same thing. Thunderbird rendered it perfectly and the others ignored and/or removed it. I'm at a loss as to why CSS in an email is ignored or stripped out by Gmail. Sure makes it ugly though.
Installed K9 from f-droid. Stick gmail credentials in. Fails to log on or function. I assume because gmail is actively blocking them.
No email notifications of the login attempt from gmail nor android popups. Just silent block.
I'm sure there is a reason that makes it all my fault or nothing else that google could possibly do other than that. So let's hear them.
https://github.com/thundernest/k-9/issues/6020 https://support.google.com/accounts/answer/6010255
Blocking 3rd party apps seems relevant to this discussion.
Blocking without notifiying as to WHY to make it look like another kind of failure is pretty anti-competitive at best. Bait, "yes we let you use your client with imap" becomes switch "we have enough market share to throw our weight around to hurt the competition"
/Me waves to the google employees who found the original post somehow not meeting the guildelines here.
Anyway K9 does not work with gmail, apparently since 2 weeks ago. I wonder if joining mozilla 2 weeks after that will change things.
It's possible that its interface isn't as objectively terrible as it seems, but it is gratuitously different; if you are going to be different, it should be to some purpose, but after 6 months of using it I still don't see any purpose. There are still some vestiges of its previous Eudora-like interface, but everywhere it is different from Eudora, it is to no purpose I can fathom.
1: The other two I could find are Evolution and davmail (which bridges EWS to imap). There may be more now; I last looked about a year ago. I can unreservedly say that using Evolution is better experience than Thunderbird, though its underlying architecture is a bit bizarre.
You are not actually saying what you don't like about Thunderbird so it's hard to oppose anything to your comment. Why Thunderbird should change its UI if it works okay? Gratuitously different from what by the way?
And to me, the UIs of Evolution and KMail don't seem that much different from Thunderbird's and from what I've seem from Outlook, same thing. There's a pane listing inboxes and folders, there's a pane showing the content of the selected folder and there's a pane showing the selected email's content. ?
Thunderbird is ugly, atrocious on 4K screens, does not group messages by conversation like mail client have in the past 20 years, and insists on listing emails in side-by-side columns instead of using a more compact and space-effective layout like many mail clients in the past 20 years. It's got a menubar, 3 level of toolbars and a burger menu; unnecessary tabs, cramped spacing, and two search fields. It offers XMPP chat and in-browser junk filtering when the former is pretty much dead, and the second is not needed anymore since GMail and its modern competition.
Someone mentioned LibreOffice elsewhere in the comment, and both projects seems stuck in the 2005 UI, UX and feature wise. I actively avoid using them even if I have to resort to using subpar web applications.
> does not group messages by conversation
Thunderbird can group messages by conversation. One has to ask for it specifically in the Sorting menu, which admittedly is annoying and should be the default. But it is there.
> [client-side spam filtering] is not needed anymore since GMail and its modern competition.
It kind of is, though. Not everyone is using GMail, and my idea of what does or does not count as Spam might differ significantly from what some email provider thinks.
I'm sure there are prettier mail clients around, and by all means, if you prefer one of those, use it. I used Apple's Mail.app when I using a Mac, because I needed to access an Exchange mailbox at the time, which sucks on Thunderbird. The UI looked shiny, I never got upset about it (UI or functionality), but I happily went back to Thunderbird when I got back to Linux.
It's not a unified conversation view like Gmail or Fastmail. There's a plugin that does that, but it's not very pleasant either.
I'm on Linux, all other email clients are pretty bad or similarly stuck in 2005 in appearance. I am very happy with the Fastmail webapp, but Mozilla is against PWAs so I have to resort to using Chromium to keep it around as a standalone app. I would like to use Thunderbird, but I have tried getting comfortable with the UI a dozen times in as many years, and I uninstall it in frustration 30 minutes later. These days it is also very sad to open its extension store and see that most of its plugins are unmaintained and haven't been updated in years. There's a true feeling of abandonware every time I give it another chance.
I have a hard time with complaining about free software. I either use it or don't, but I won't complain about it.
So yeah. For handling multiple e-mail accounts, Thunderbird works best.
Indeed, my reaction to Geary was "wow, they managed to make KMail look good!"
> Evolution assumes too many GNOME parts
I don't like this about Evolution either; I haven't run a GNOME desktop for at least a decade, and (depending on the distro) it can be a tiny bit fiddly getting Evolution working without a GNOME desktop. It "just works" on recent NixOS versions though. Once it's installed though, it's a joy to use, while Thunderbird feels like a drag. I'm not a UI expert, so I can't put my finger on what the difference is.
> Mailspring is electron-based
I hadn't heard of mailspring previously, but I suspect we have similar views of what at typical electron-based app is like...
> TUIs are right out
I have no problem with TUIs, I use a TUI along side a gui mail client, but the latter is necessary when dealing with non text-only messages (particularly responding to).
> Claws is even more 90s than Thunderbird
Maybe I'm just old, but Claws is way more usable than me than Thunderbird. The only thing it lacks for me is an html e-mail editor, which is necessary for responding and quoting to non text-only e-mails. There are plenty of good TUIs that can handle text-only e-mails, so claws isn't really something I currently have a use for.
Over the weekend I migrated to a 100% Linux/open-source environment. So of course I went through a series of alternative apps. While you'd never find feature parity (eg. with MS Office and the Adobe Suite) existing alternatives are mostly good enough.
Thunderbird has been tooted to me as THE alternative to MS Outlook. What I found was a total mess of a software stack needing countless addons and customization to even get close to Outlook. Then I discovered Evolution. It was certainly different but worked pretty much flawlessly out of the box.
Congratulations and good luck for your migration!
It's good you found a solution in Evolution. I don't use many extensions in Thunderbird, just one to show different colors in the new mail window depending on the account used to send the mail. I'd be interested in knowing which specific features of Outlook you needed that required extensions in Thunderbird (and that Evolution has by default).
For traditional business email, you are pretty much stuck with two options - Outlook or Thunderbird. You can try and use webmail of various types - and there are some decent webmail clients out there - but it's just not quite the same as using a good Desktop client.
For the newest Outlook versus the newest Thunderbird, Outlook wins feature wise - it does a lot of things and all of it is pretty well integrated. But if you don't need all the bells and whistles, Thunderbird, in my opinion, is oftentimes better than Outlook.
For example, I find that Thunderbird has a better search than Outlook, and better conversation capabilities (threading and open message in conversation) than Outlook. It's tagging system and archiving system are better than Outlook's. It's add-on system is easier to use then Outlook's (partly because most everything is free). It also allows much more manipulation of the UI than Outlook does as far as I'm aware. All of these things add up, and Thunderbird becomes a much more efficient email client than Outlook for me, which is important if I need to deal with 100 or so emails a day.
On top of these things, in my opinion, Thunderbird is really good and not causing issues down the line. Part of my job is to manage workplace computers (probably no where near as many as some folk here on this site, but not a really small amount either), the amount of trouble Thunderbird gives me compared to the amount of trouble Outlook gives me... well, it makes me want to make everyone use Thunderbird. There are only two things I really have to worry about with thunderbird (they relate to the calendar and mass moving large amounts of emails), but with Outlook... well... I've had to repair people's email enough that I never want to do it again. It sucks. I hate repairing .pst files.
Mmm... I personally have never tried Evolution or davmail, so I can't speak to them, unfortunately. I will have to try them later probably. As for Exchange OAuth, yeah, I can see you having some troubles with it in Thunderbird. I was having Microsoft 365's IMAP implementation give us trouble (ended up dropping it). On the other hand, it was also giving older versions of Outlook trouble... so I'm not sure I can put the blame solely on Thunderbird for this one. I do hope you find something that does what you want it to though.
Thanks for that. If it wasn't clear from my original comment, I'm totally cool with people liking Thunderbird.
FWIW, It seems that the intersection of <people who like Evolution> and <people who like Thunderbird> is near zero, so perhaps it's a matter of taste there.
How do you do bulk actions on emails? Clicking a checkbox for every single one? Having to write an run a query for each set of emails you want to operate on in bulk?
Here's how I do it in thunderbird. Sort by sender (or other thing). Click first email. Shift click last email. Do whatever I want (usually archive or move to a folder).
My employer turned off IMAP access a few years ago and I've basically given up managing my github/jira/mailing list spam. Filters turn out to be surprisingly fragile.
This is pretty basic functionality for an email client to have. In GMail (Proton and Yandex too; probably any modern web client) works by clicking the checkbox. Click first email's checkbox. Shift click last email's checkbox.
I don't believe gmail supports this.
I don't think Thunderbird is that relevant today since that it's essentially abandoned and still no Android or iOS app for decades. I'd say that ship has already sailed and it would be a long way to go to support iOS.
It has gotten this bad that they needed something like K-9 to have Android support, and you would have thought that even with Google's money they would have already funded an effort like this on their own. Well now lots of Mozilla fans are going to realize that their 'donations' don't actually go to Firefox, or Thunderbird, etc but are really wasted on other schemes.
An opportunity thrown away for years and now they come back again with this announcement decades later and coming very late to the party after Thunderbird getting deprioritised.
A giant mess.
HN is for curious conversation.
(The first bit probably does qualify as snark though.)
I'm just wary of Mozilla products these days. I'm tired of important tools I use just losing functionality while blog posts pay lip service to power users and internet freedom.
As with all things Google, the only winning move is not to play. Long term I'm sure they'll axe this and tell you to use their app or bust.
I finally set up two factor auth in gmail using FreeOTP and have Fairemail (fdroid version) configured using an app password. (I run LineageOS with no Google services installed or account setup so the OAuth method isn't an option for me)
this is probably the better way in the long run, but freeOTP might do the trick for now!