Conversation:
Notices
-
@mcscx @simsa0 Why would a remote instance, knowing only of User A on my server, accept a message from User B? Imho it only increases CPU and bandwidth usage. Another word for sending messages which have not been requested is: spam.
-
@mcscx @zash I believe @manuel@lamatriz.org has started working on this. It is probably the most proper solution which requires least hacks.
-
@mmn @mcscx @zash @manuel@lamatriz.org Yes! we are working on this. We want to make able subscriptions to a conversation feeds but not form panoptican approach, it means, we want to subscribe them for show them in other applications, in our case, a plugin we are developing for wordpress !wpgnusocial We are not trying to resolve what someones are calling «missing dents» or «fede…
-
@simsa01 @mmn @mcscx @zash If your are subscribed to me and your are not receiving some of my publications, it is a technical issue that have to be checked.
But this is not a reason for proposing a panoptican Twitter-line social model and get away form a federation social model based on people connections.
This is a interesting debate and we find this is paramount.
!gscamp !gs !gnusocial
-
@simsa01 @mmn @mcscx @zash Great to know this!
I think all this conversations on #federation have been very productive. We have identified a technical issue - you are not receiving all publications from people your are subscribe to - and now we have a solid base of how federations works on #gnusocial and how it is based on people connections #important
!gs !gnusocial !gscamp
-
@simsa01 If both authors are subscribed to, both authors' notices will be shown. If they are just "known", that doesn't mean anything and there is no subscription link setup and no reason why I'd want to host a bunch of crap this "known" author writes which I have expressed no interest in (and am not a direct recipient of). The "remote retrieval" (i.e. fetching a miss…
-
@simsa0 In the latest GNU social, conversations are stitched together assuming proper conversation metadata (URI) is provided. This metadata is _not_ provided with StatusNet (which makes up totally new conversation URIs when receiving notices and then continues to distribute this incorrect URI when pushing out replies on the topic).
-
@simsa0 I think everyone are capable of understanding pros, cons and peculiar behaviour duribg the development phase. It would be bad if "newbs" actively avoided software which admits to being actively improved.
-
@simsa0 You're welcome to give us all your money so we can improve these things. But it will still take time after that. Until then, I think it's a good thing that the network is used, bugs are reported (can't report bugs without actually testing!) and users don't have to stick with proprietary spy-networks to do what they do here. And people do come here. People do s…
-
@lnxw48 They _are_ desirable features in the technical _specification_ of the protocol. There is absolutely no reason for anyone to have an implementation which accepts spam. The way to solve this - which is being worked on - is to have another type of subscription (i.e. conversation thread, which is already an RSS/Atom feed - just no mechanism to subscribe). Not an u…
-
@lnxw48 If you're thinking about the Las Indias(?) article, I didn't interpret it quite like that. More that it's not always a bad thing - in many circumstances it is actually a desired feature not to receive notices (which would bump up the thread in current implementations) from _every_ party who gets involved in the conversation. I think everyone agrees that there…
-
I don't think this attitude is getting anyone anywhere. I prefer constructive criticism and helpful attitudes. You've just made it to my block list.
-
Also, you're welcome to say I don't do anything to the commit timeline here: https://openhub.net/p/gnusocial/commits/summary
I bet it will tell you that work is being done, but without guaranteed time nothing it is hard to perform miracles. (I don't get money for this, so I can't devote time away from my job or too much free time). Thankfully, there are also other contributors who I greatly appreciate the work of.