Is there a way to preserve settings from the old version to the new?
Not right now. We've rewritten and reorganized the entire infrastructure that handles settings and configuration, and the new and old schemas are more or less alien. We've got an issue open to build a settings import/export tool which will allow things like easily migrating settings across platforms that should also make this possible.
It's looking like we'll have the public 2.0 TestFlight launched within the next couple of weeks, though the error bars on that timeline are pretty wide because development speed is dependent on our free time. You can also check our roadmap post for a feature-by-feature list of what's left; we're expecting to merge a couple of the outstanding items in the next few days.
It looks like the biggest domain count driver by far is image fetching for link posts, since rendering a website preview involves fetching both the preview image from the linked website and the favicon from Google (favicons account for all those t1.gstatic calls). Disabling website previews and the associated image fetching code cut the domain list down to just Lemmy instances. Mastodon appears to proxy image requests through the instance, which prevents that high domain count. Lemmy has recently added that feature, but right now only instances running the very latest Lemmy code perform image proxying.
We're looking into adding enhanced privacy features to 2.0; we'll include one to disable fetching favicons and image thumbnails, since favicons especially are a known tracking/fingerprinting vector.
If you want to keep using Mlem while the TestFlight is down, you can just download it from the App Store--the current release build is identical to the TestFlight, and should keep all your preferences.
TL;DR: Mlem v1 is not fast; Mlem 2.0 (announcements coming soon!) will be.
We’re aware of a number of performance issues with the current codebase, which all together result in the app not behaving as responsively as we’d like. Unfortunately these are largely due to design choices made during the hectic sprint to App Store release last summer, and so are infeasible to fix without rewriting the app from the ground up—which is why that’s precisely what we’re doing. Our 2.0 build should be significantly faster; we’ll have some announcements about that in the near future.
Sure! One of the big things we’re reworking for 2.0 is how we handle accounts, which includes changing the format of how we persist account information. The compatibility changes in this build add extra decoding logic to the account persistence system so that both v1 and v2 can load a saved account regardless of which version saved it.
This is not possible in the current version, but our 2.0 rewrite gives us the level of granular color control required to effectively implement such a feature, so we'll be sure to include it when that ships.
This is a known issue, and one that has us completely stumped--we don't have a single line of code that touches anything related to audio. We're monitoring it closely as we bring our 2.0 codebase online to try to figure out what's causing it, and aim to have it fixed in 2.0.
Gif support is still one of our major outstanding tasks. We are currently working on a 2.0 version of the app, and are hoping to ship it with full inline gif support.
Thanks for the feedback! We're aware that the iPad version isn't up to the standards of the iPhone app--we're a very small team, and so lean hard on Swift's automatic multi-platform features to make the iPad app even feasible within our development bandwidth. We're looking into ways we can improve the iPad experience without introducing unsustainable technical overhead as part of our ongoing 2.0 overhaul, and we'll be sure to take your suggestions into consideration.
This behavior can be customized under Settings -> Links -> Tappable Links. By default it uses "contextual," which will display the full link in large contexts and just the title in compact contexts (e.g., comments if you have comment display set to "compact"). Setting it to "large" will always display the full link regardless of context.
Thanks for the bug report! We're working on it.