Mozilla Thunderbird Old Version
Caterina Haggins
Thunderbird version 128.0 is only offered as direct download fromthunderbird.net and not as an upgrade from Thunderbird version 115 orearlier. A future release will provide updates from earlier versions.
I have just had to install thunderbird on a new PC, and the email setup system is causing serious crash errors (red boxes with lots of text and Stack Overflow programming bugs (the actual thing not the website) ). I want to try installing an older version of Thunderbird, specifically 78.13.0
This is sort of what I was looking for but is hard for anyone not tech savvy to use. How would I know about this page from the Thunderbird official website? THAT is my issue; that releases are so hard to find
You wouldn't know about the archive site from the official site, and I think that is a deliberate measure to steer users to the current release, especially now that moving between old and new releases requires some tech savvy (profile per installation etc.). In my view, a more serious flaw is posting the new version on the official site long before it is released as an automatic update to the previous version. Most users should see 78.13 on the official site until 91 bugs are resolved to the point where auto-update is justified. Links to 91 downloads should be accompanied by a disclaimer that it's only intended at the moment for experienced users who can run it without affecting their current setup (separate installations and profiles).
The latest upgrade is nothing short of abysmal, and almost unfit for fit. Who signed off on this release really does need to take a long hard look at the concept of 'usable software'. Under the previous version, everything worked fine, under this one, I have so many problems its difficult to know where to start.
MailBoxes
My 'Inbox' currently has 404 messages in it, when I go into the inbox, it automatically opens half-way down the list - not with the first or last, just half way down, regardless of the order column or A-Z, Z-A order. So the first unnecessary action every time the mailbox is changed, is to scroll to the bottom to see new emails.
In a morning, when I first start up and it checks for messages, the number of messages will generally exceed the amount that can be displayed, so you scroll up and down to see the others, thats fine, standard practice. But if ordered latest at the bottom, and you start with the oldest new message, once you hit reply or delete, the list shoots to the bottom (ie latest) so you cannot see the current or next message in the list.
The search facility is now a joke. 50% of the time having typed in a search term (usually a name or email address) it opens a new window, does not do a search and blanks off the search - the new window shows no emails or search terms.
You also cannot move the cursor back or forth (with cursor keys or the mouse) to correct anything. If you notice a typo, the only thing you can do is delete back to the point and retype. Very infuriating, especially if you are searching on a term you have pasted in from the clipboard.
When the search does work, and it finds several emails you click on the email you want to look at. Previously this opened the email, now it opens a new search window containing just that email, which you then have to go and click to open the email. When your searching through 10-15 emails to find the one you want this is time consuming and frustrating.
Should you click on an email that has a discussion style reply, previously, it showed all the discussions and you could easily click between the various emails, which opened in the bottom panel, to read the discussion quickly and easily with the up and down arrow keys. Now only the latest email in the discussion comes up in the top of the new window, you have to click it to open in the bottom panel, but it also opens every discussion email in a new tab - which is horrendous if you have an email with 30-40 replies, as you then have 30-40 tabs to go close.
The whole system has become borderline unusable since this latest update, and somebody from Mozilla needs to get a hold of things and sort it. I intend to try going back to the previous version (although I suspect the email database will have been updated which would remove that as an option) and turning auto update off.
It used to be a nice, simple, quick and easy to use piece of software, one of the best for managing emails and multiple accounts. But this latest version is atrocious and just makes what was a simple task, now complicated and difficult.
Don't forget that you also can't add the Delete button to message windows. I always had a Delete button in the Mail Toolbar. The installer took it away, and it's not in the available icons to add back. Why?
Exactly this. Somehow 115 is missing the ease of use and intuitive use, 102 and previous had.
Sure, new versions of software always have the "getting to know it" factor in it. But with 115 this has gone too far. That ease of use, the logical UI, are gone or at least hidden deep
As a dedicated / suffering Firefox user and with a strong dislike for webmail, I've been looking for dedicated email clients forever, having suffered the coma ( Eudora '8' ) and life support ( 'Hermes' ) status of Eudora.
I periodically retry Thunderbird, which admittedly leaves me always setting it up again so waiting forever for IMAP sync, which Thunderbird seems very poor at (by default) but is also probably due to inherent issue with the protocol. I never used Eudora in IMAP mode which probably says a lot about why it was so usable.
I appreciate the developers have an extremely difficult job with Thunderbird. It's barely been maintained and wouldn't be a buildable, releasable product if not for all the development hours it inherits from Gecko (Firefox code base). It's hard to know how much work that inheritance also adds though. Continually triaging patches for their impact in a mail client context; having to pass on unpopular policies like the move to Web Extensions that broke so made so many developer's efforts redundant, requiring complete re-writes; the move away from XUL; the integration of a 60 FPS graphics engine target for the likes of WebGL and other APIs that are essentially irrelevant to a mail client, unless you think it's important to play 3D games natively in your Thunderbird.
The list goes on and on and on whilst the developers have been open about the extreme ball ache they've had to confront to merely convert the UX (user experience, user-facing windows, tabs, dialogs, sidebars, ... you name it) into something they can work on improving.
Whilst many users may have loyally, tirelessly stuck with Thunderbird over the better part of the last two decades, that continuity has never been supported by continuous development. Integrating PGP and calendaring were nothing short of incredible with the limited developers they had available. Neither of those features would have, I'm presuming, been implemented with almost any leg ups from Gecko APIs. Probably OpenPGP libraries would have provided core functionality, but they would still have had to implement, integrate and make the overall experience of using encrypted email as simplistic as possible. Something no open standards email client has ever done before. Hence the plethora of closed-standard implementations of crypto email that have proliferated since Snowden.
I personally don't care about calendaring and scrambling email with crypto isn't a priority. But the Thunderbird people doubtless inherits the ethics of Mozilla, such as enabling dissidents and whistleblowers to risk their lives to make the world a better place. Hence crypto support.
If you use Proton, for example, you need to invite non-Proton correspondents to view your crypto-scrambled email via a download link on a website that likely doesn't integrate with whatever email scenario they have. Corresponding between the walled garden of Proton users, crypto-scrambled style? No worries! All happens within their web app client. The same applies to some other post-Snowden attempts to tighten up email.
I won't be continuing to use Thunderbird this time around either. It's been a somewhat handy tool for trying to migrate mail from one gmail account to another. I'm thankful for that although it's currently confusing the crap out of me by redownloading messages back into All Mail from the source account I'm trying to retire, after I've told it to move the messages to the destination account ... as it happily did for other mail. It's performance is atrocious despite the advantages that would surely be available, but likely they've not (yet?) the developer resources to use, from WebAssembly and other hardware acceleration such as WebGPU that is turned on by default, at least sometimes I presume, but does not appear to be well utilized.
Front end UX is critical, but it's only ever enabled by the speed of the core tasks like mail handling. Otherwise developers have to go with dodgy compromises like only downloading headers by default; lazy loading mailboxes and so on.
Thunderbird developers need to publish performance targets and how they're going progressing towards them. How long, on X hardware, does Thunderbird take to present a zippy version of a folder/label with Y emails totalling about Z size? That sort of thing. They collect 'telemetry' stats. Have done for years. They need to make performance at least the equal highest priority amongst more behavioral tweaks such as keeping the mailbox positioned (scrolled status, which mail was last selected, sort mode) as desired, consistently.
I also feel that, whilst being reluctant to just target chunks of functionality I do not use, it could be important to investigate how many people actually use the non-mail features such as Matrix integration, Chat in general, feed reading (RSS). If they are not overly common, would the extremely minimal developer resources be better spent on the (arguably) core functionality behind all of the features, perhaps disintegrating some features into Add Ons? Bound to be a suggestion that creates a lot of argument unless the disintegration can be done without any noticeable disruption to users. But it is likely a debate that is sadly necessary due to limited, but much better than 5 years ago, developer resources.
c80f0f1006