fosstodon.org is one of the many independent Mastodon servers you can use to participate in the fediverse.
Fosstodon is an invite only Mastodon instance that is open to those who are interested in technology; particularly free & open source software. If you wish to join, contact us for an invite.

Administered by:

Server stats:

11K
active users

re-upping this question about changing a masto handle within-instance: neuromatch.social/@jonny/11360

has anyone tried to do a handle-change within-instance preserving post/interaction history, at least locally? it seems like we should be able to just manually make the changes in our DB and then make a manual redirect for all the old URIs to the new URIs. It would be great to keep everything and sync across the network but keeping bookmarks, posts, (our representation of) boosts from the person's perspective, even if it's out of sync elsewhere would be fine.

Neuromatch Socialjonny (good kind) (@jonny@neuromatch.social)How can someone change their handle on mastodon if e.g. it includes their deadname? Is there a way to migrate an account within-instance?

Looking at the masto status and account update services, and the account move service, it's not entirely clear to me why it should be forbidden to update the uri and username fields - that seems like it's implicitly allowed by the ActivityStreams Move action, and AS requires that all objects have an absolute URI as an id but doesn't specify that they can never change. Neither does ActivityPub. ActivityPub actually specifies that an Update should completely replace the old object and so it should be possible to replace the URI/id.

www.w3.orgActivity Vocabulary

now and as always i remind everyone tempted to tell me that "cool URIs don't change" that i disagree and here is one very obvious reason why - people change.

wizzwizz4

@jonny I would say that ActivityStreams's Move action is within the spirit of Cool URIs Don't Change.

However, I've just re-read the original article (w3.org/Provider/Style/URI), and… hm. This is not nuanced enough. The rest of Tim BL's 1998 style guide is outdated, wrong, or overly-simplistic, but my sentimental attachment to this particular article was preventing me from seeing its flaws.

Do you know of a better canonical reference for how URIs should work?

www.w3.orgHypertext Style: Cool URIs don't change.

@wizzwizz4 not offhand, except for me generally seeing it as being shortsighted and a particularly ideological belief about the way the web should work that never really survived contact with reality. Imo the ideal behavior would be for websites to work like the read-write web: when someone clicks a link in a page and it takes them to a 301, the browser should politely tell the page that its link is up to date, it checks for itself and then updates the page. but that's a very different web. I mostly think that one just needs to embrace the reality that "all things must change and it's better to be able to heal around that change rather than insisting that change should not happen" which means you design your pages with changes in mind (e.g. periodically checking your outlinks and updating them, etc.)