Closed pfrazee closed 6 years ago
hmm, okay what about "forget". I think "delete" is misleading, because people are used to delete actually you know, wiping the thing. but "forget" is like "forget I said that" and people intuitively understand that it's not really possible and that it's just polite never to mention this again.
I like that
On Tue, Oct 21, 2014 at 1:36 PM, Dominic Tarr notifications@github.com wrote:
hmm, okay what about "forget". I think "delete" is misleading, because people are used to delete actually you know, wiping the thing. but "forget" is like "forget I said that" and people intuitively understand that it's not really possible and that it's just polite never to mention this again.
— Reply to this email directly or view it on GitHub https://github.com/dominictarr/secure-scuttlebutt/issues/42#issuecomment-59975978 .
I like this idea of a generic delete type hmm, I'm assuming that you should only be able to delete your own messages though.
Yeah
On Tue, Oct 21, 2014 at 1:39 PM, Dominic Tarr notifications@github.com wrote:
I like this idea of a generic delete type hmm, I'm assuming that you should only be able to delete your own messages though.
— Reply to this email directly or view it on GitHub https://github.com/dominictarr/secure-scuttlebutt/issues/42#issuecomment-59976366 .
About the reltype, though, I'm trying to work out a scheme where X "somethings" Y.
For example, X replies-to Y
, X rebroadcasts Y
, X follows Y
. "Forget" or "forgets" wouldn't work great in that scheme. Maybe "hides"?
well hides has wrong intuition. does it have to be a verb? what about: pfraze "please-forget" i_farted
Maybe...
requests-you-forget
withdraws
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
We need a way for users/apps to remove messages from the indexes. A good example: the replication protocol collects the 'follow' links to decide who to fetch, and we need a way to remove links when the user 'unfollows.'
One solution would be to create an 'unfollow' reltype, but I think we can go ahead and generalize the solution with a single reltype. The mechanic would be, if a message is published with the 'retires' (or 'deletes' or whatever) reltype by the same author of the target message, SSB de-indexes the target message from the type index, and from the indexes created by links in the target msg.