Archived
request: consider changing user settings storage to be server-side #54
Assignees
jcoffey-dev (John Coffey)
Clear assignees
jcoffey-dev (John Coffey)
No Assignees
No due date set.
Dependencies
No dependencies set.
Reference: coffey-labs/ihasmail-github-archive#54
Reference in New Issue
Block a user
The original description was not recovered.
Referenced by commits
Rebuilt from: notification email, session transcript.
Now that I have switched by Stalwart to 0.16 I am looking at making a lot of adjustments like this. There were several things that I wired in that are tied to the browser, which means they don't travel with the user. Thanks again for all the suggestions and call outs, you have been a great help.
Date not recovered.
Thank you so so much, both for agreeing & implementing it so quickly. I've read through your description in #55 and fully agree with your decision to keep some settings stored locally, and with the set you chose. Those are all specific to the specific machine/situation; some might have a multi 27" monitor setup, preferring to run every window maximized, while others only run a single 32" monitor with the window tiled to one side — resulting in actual narrower viewport than on the 27" monitor.
Right back at you, like I said above: for implementing all the fixes & changes so quickly & for providing ihasmail in the first place.
A bit of background on my situation (warning: this is becoming quite long; feel free not to engage 😁. You've read me mentioning my work a couple of times before, and that's the main driver. We're a small-ish IT shop (around 50 people). I'm the CTO, but also manage our own IT & do a lot of Linux/automation work for customers. Usually we service a high number of smaller customers (often 30—80 employees) but also have several schools and school districts as customers running up to 22.000 email accounts in one case.
Over the past five years we've migrated several customers onto a mail stack backed by Open-Xchange (OX), which builds on Dovecot+Postfix but with its own calendar/contacts/files module, and most importantly its own web UI. That stack speaks the usual protocols, and that was fine, if not great. It was one way to run affordable or even free full-stack groupware, especially for those aforementioned schools & school districts which are criminally underfunded here in Germany.
The big, big problem that arose recently is Open-Xchange-the-company deciding to completely re-do their Open-Xchange-the-software stack to only run in Kubernetes from now on. As you may guess from what I've written so far, neither we nor our customers are at a size where managing real Kubernetes stacks (everything beyond the all-in-one-binary-on-one-machine test cases) is completely out of the question. We do not have the expertise, the time to learn it, the procedures in place etc., and the schools absolutely do not have those either together with not having the funds to pay external people to do it for them (all for the biggest school districts).
Therefore we are now in the position of having to find a new Open Source, hostable-on-premise, full groupware stack that we can use for ourselves & our customers — instead of having to migrate them to the ever-present Microsoft 365, which we simply don't vibe with from so many view points. Hosting with other providers is often out of the question, too, both for privacy reasons (which is a pretty big deal here in Germany) & for cost reasons.
We're currently considering:
One additional problem we see with Dovecot is that it also belongs to Open-Xchange-the-company. They have the Dovecot v3 line which is currently commercial/proprietary & used with the current OX Kubernetes stack, and we do not see that much evolution in the Dovecot Open Source versions the other stacks mentioned above use. We're concerned about Dovecot's OSS future, to be blunt.
Before discovering ihasmail we leaned slightly in favor of the Nextcloud-based stack, but with a heavy heart. It was the best of the fine but not great solutions. Now I'm already seriously considering to make the executive decision to go for Stalwart+ihasmail.
For all of those reasons the fact that you posted about releasing v2 on Mastodon of all places, the only Social Media outlet I use, was serendipitous to me personally & to us as a company.
At the moment I'm the only one doing all the Stalwart tests, exploring & testing ihasmail. I still have one or two things that I know I will likely report (mostly around calendar scheduling with multiple people). Soon-ish I will also get several of my tech-savvy colleagues to join in on the testing fun, and we'll all give ihasmail a serious go. Expect some more reports then.
I hope I don't sound too demanding with what I wrote. I'm an OSS author myself, and I fully respect authors' rights to just say "no" for whatever reason, or "patches welcome" or "yeah, but low priority, don't know when".
Again: thank you, thank you, thank you.
Ha, I actually LOLed at that 😂 Yeah at that scale (and much smaller scale) Kubernetes absolutely makes sense and is an impressive piece of work, no doubt. Just not for our kind of scale at all; we're at the diametrically opposed end of the simplicity-complexity scale.
I've seen that, thanks. I currently follow the whole repository & get notified of everything. Will probably not keep that up forever, but for the time being I'll keep it like that.