00:14 <georgeowell> rocketchat has been super problematic for me deploying it as part of my buisness

00:15 <georgeowell> on the so called stable branch they change the UI in a really negative way. They actually removed the ability to see all the channels on the server.

00:15 <georgeowell> you have to know what a channel is called in order to find it

00:16 <georgeowell> they just updated to re-add that feature but added a whole new set of defualt behaviours

00:17 <georgeowell> Considering it is meant to be an app for organisations/buisness this is really a very bad way to run a project

00:19 <georgeowell> Updating the Wordpress would be really nice as a lot of the themes don't work with the older version currently in Sandstorm

00:20 <georgeowell> I would also really like an update to Framadate as the current version is actually usable

01:09 <xet7> Somebody asked about security disclosure scope, so I added it and all known security details about Wekan to https://github.com/wekan/wekan/wiki/Security-Disclosure

01:17 <georgeowell> looks good

01:17 <georgeowell> do you use pgp?

01:18 <georgeowell> you could include your fingerprint in there

01:48 <xet7> Yes I do have PGP key also

05:35 <knightcat> hello, does someone know where i can find the time limit for an error message to expire and resend the token email to a user?

05:36 <knightcat> a user tried to log in but he keeps seeing "it looks like we sent a log in email to this address not long ago...."

22:43 <ocdtrekkie> kentonv: Can you look at maybe doing a vagrant-spk release? The release script has a couple dependencies, but looks like it does all of the work, including Windows support.

22:44 <ocdtrekkie> I would include the #211 PR that isn't merged yet.

22:49 <kentonv> ocdtrekkie, I can try running the script but someone else will have to tell me if it worked...

22:50 <ocdtrekkie> It looks like it needs a github-release tool, and if all is right with the world, at the end you should have a release on GitHub tagged with the current Sandstorm version.

22:53 <ocdtrekkie> Wow, it looks like the script even downloads the Windows setup tool off their website if you don't have it. o_o

22:53 <kentonv> ugh, there are multiple tools on the internet called github-release, none of which seems particularly popular

22:54 <ocdtrekkie> Asheesh does not kid around when he writes scripts though.

22:57 <kentonv> I've found at least three github-release tools with no apparent relation to each other

22:57 <kentonv> the only one which looks like it might match the use in the script is https://github.com/aktau/github-release

22:57 <kentonv> just written by some random guy...

23:11 <ocdtrekkie> It has 919 stars!

23:12 <ocdtrekkie> That's not bad for some random guy. :P

23:28 <kentonv> gah I have to give this thing a token that gives it read/write access to all repositories because there's no way to create a token for just one repository

23:28 <kentonv> you can create tokens that are restricted in what actions they can perform but not in which repositories they can access. Why does anyone think that's useful?

23:30 <ocdtrekkie> That seems poorly designed. :/

23:31 <ocdtrekkie> I guess the workaround would be to create a user with only commit writes to the one repo, and use that user account?

23:31 <ocdtrekkie> rights*

23:32 <kentonv> that really shouldn't be necessary. :/

23:32 <ocdtrekkie> Seems like a good feature request for GitHub.

23:33 <ocdtrekkie> hot take: A good feature request for GitHub to ignore.

23:33 <kentonv> my experience is GitHub doesn't care about these things very much

23:34 <kentonv> meanwhile, dammit, the Sandstorm release script was running in the background, and whenever it gets to the point of doing GPG signing, it pops up this full-screen password prompt and if you merely click anywhere then it cancels the prompt and the whole script fails right at the almost-last step.

23:34 <kentonv> and I was clicking on a web site right when it hit

23:35 <kentonv> I wish UI widgets would ignore input events that they receive within a second of the widget appearing.

23:36 <kentonv> just, universally

23:36 <ocdtrekkie> I wonder if I could find a GitHub employee who cared. (I just this week started a conversation with probably "the right person" about my Windows Update woes. He is interested in getting it looked into further, which is... kinda huge.

23:37 <ocdtrekkie> kentonv: If your UI widgets did that, at least one person would complain about having to wait a second to do something.

23:40 <kentonv> it would need some additional logic. Maybe if the UI element appeared / moved in response to user input, then there's no pause. But if it appears/moves as a result of background processing then it's locked for a second.