I can't speak for other browsers, but keeping a GitLab tab open in a background tab consistently causes my CPU dedicate 15-25% of its cycles to Firefox.

I do file bugs against GitLab every now and then, but I haven't done so for this one because I assume that there's an existing item on file for this (and I don't care to look for it) and that this is all part of the the longstanding, "Yeah, we really need to work on our frontend story, especially for mobile."

The end result is that just don't keep GitLab tabs open. Which is a little obnoxious, given the well-known issues with how slow GitLab is to complete requests.

Mmm, I found https://gitlab.com/gitlab-org/gitlab-ce/issues/521 but that was a Firefox bug. It is probably something in GitLab but I'm not sure someone filed a bug report for it. We have complete UX and frontend teams and at this moment all the views should work great, also on mobile. GitLab self-hosted should be fast but we're working on the speed of GitLab.com.

Here's another issue about Firefox using 100% CPU https://gitlab.com/gitlab-org/gitlab-ce/issues/342

Searching an issue number in the list page's text input should take me to the issue. Sounds small but it's a frequent irritation; I will be 200% happier with Gitlab once it's solved, and my teammates even more so.

Hi throwanem. This is an excellent idea. I've created an issue about it in GitLab's issue tracker so we can move forward with it: https://gitlab.com/gitlab-org/gitlab-ce/issues/25771

Thanks!

If you have Alfred, you can set up a custom web search since issue numbers are in the URL. I have tons of these set up to get to different repos, my global list of PRs, etc. It's that much faster because you don't even need to open a browser or navigate to the page to type in a search field.

Interesting idea, but it doesn't really fit my use case, because I always have a tab open on the issues list page for each project on which I'm working at a given time. When I want to open tabs for issues of present interest, as when reviewing for deployment or preparing a changelog update, it'll be a lot more convenient to do so by searching issue numbers and middle-clicking their entries in the result, rather than the current method of copying a URL to an issue page, then doing C-TAB C-l C-v <end> <backspace> <backspace> <backspace> <backspace> 1 2 3 RET with as many issue numbers as I have in hand to deal with.