liamcottle
[Improvement] Deleting messages should delete them on the other person's side as well
- When click the delete message button, it's deleted only in my window. It's not deleted in other contact's window(my interlocutor). (I have a Meshchat, and my other contact(my interlocutor) has a sideband.)
!The message should be deleted from both sides of the conversation! (I think that there should be an opportunity delete both messages - self messages and delete the messages of interlocutor. Be deleted from all places, from the program database, from correspondence, etc.)
- If I haven't changed the other contact(interlocutor) name on my side(in my meshchat), but other contact changes theirs, interlocutor other contact name(in my side) should also change. If I change the other contact name(interlocutor), at least once, and then other contact changes theirs to a new name, I should retain the contact name I chose.
- It's cool that the "delete message" button appears when click on a message, but it's inconvenient, and can accidentally click it and delete the message... It's better to show three dots near the message, with additional options, when click on the message or when the mouse cursor appears over the message, where can select "delete message." (like in Jami messenger jami.net)
- I use line breaks (new lines) in messages more often than I send them... So sending messages is Shift+Enter! And Enter is just a newline.
Overall, Jami's interface is more user-friendly! I'd love to see a similarly user-friendly interface! (Jami is open-source! Maybe can get something from there!) (my temp LXMF, connected to settings/(Show)"Community Interfaces": 02c9ba1e3fe290da65f59ab14c4f7dd6)
Deleting in both places isn't a function of LXMF as far as I'm aware. All Meshchat is doing is providing a UI for the underlying RNS/LXMF. Try posting on Marks GitHub or in the Matrix chat.
Deleting in both places isn't a function of LXMF as far as I'm aware. All Meshchat is doing is providing a UI for the underlying RNS/LXMF. Try posting on Marks GitHub or in the Matrix chat.
- I believe that deleting a message from all places is the main principle of decentralization and anonymity - we need to be able to delete our trace! :-)
- Unfortunately, I've been blocked from discussions at https://github.com/markqvist/Reticulum/discussions... (where I could bring my idea up for discussion, for example) And they won't tell me the reason :-(
- I'm not a programmer, and like many others, I spend my personal time improving programs by reporting bugs...
- I don't know what Marx is, etc., and unfortunately, because I have to work at another job, like everyone else on this planet, I can't study or delve into the topic (of Matrix, for example)...
- I've published my perspective on this issue as best I can, within my capabilities, because I really want Reticulum and its programs to improve. (And some bodies don't even do that.)
- Well, at the moment, I can't do much more myself – I'll be tracking this bug – maybe someone will take it up and need my help!
- I think this issue to be announced - I have announced it... And then the developers decide whether to leave my issue to deal with later or delete it, because it will never happen.
- If anyone else can post my issue to the matrix, or take on the task of correcting the protocol themselves, I'd be delighted! (I'd appreciate it if let me know, rather than just silently deleting my issues.) :-)
The message should be deleted from both sides of the conversation
Uh no, that is the stupidest thing I have heard in a long time.
How the heck could you guarantee that the message was actually deleted from the remote end, unless LXMF assumed complete control over the remote device, through some sort of verified boot procedure?
Something like that would only add confusion and give people a completely false sense of what you apparently call "decentralization and anonymity".
What you are proposing here is the exact opposite of actual decentralization and anonymity.
I will never in a thousand years add such stupidity to the LXMF protocol. Forget about it.
I've dropped my two cents, and now you know what my stance is on this. Just so you know, I will not provide further comments, answers, discussion or anything similar on this thread or discussion.
It is how it is, and it will stay that way. Anything else would be idiotic.
The message should be deleted from both sides of the conversation
Uh no, that is the stupidest thing I have heard in a long time.
How the heck could you guarantee that the message was actually deleted from the remote end, unless LXMF assumed complete control over the remote device, through some sort of verified boot procedure?
Something like that would only add confusion and give people a completely false sense of what you apparently call "decentralization and anonymity".
What you are proposing here is the exact opposite of actual decentralization and anonymity.
I will never in a thousand years add such stupidity to the LXMF protocol. Forget about it.
I've dropped my two cents, and now you know what my stance is on this. Just so you know, I will not provide further comments, answers, discussion or anything similar on this thread or discussion.
It is how it is, and it will stay that way. Anything else would be idiotic.
Unfortunately, you're very harsh in your statements. It's a shame you don't want to hear (my) thoughts in more detail! As I understand it, your solution is uncontested... And that's exactly what's wrong!
Words like "stupid" and "idiotic" are unacceptable. Unfortunately, by using such words you contradict yourself = I saw somewhere, in the descriptions of programs, that we need to be kinder to each other, not swear, and describe problems and suggestions without insults, etc. (I can’t immediately find a link to your own statements about this)
You may be a good programmer, but it seems you're unwilling to listen to others' opinions—that's bad. I hope this is a "professional deformation" (of the manager)... (that's also bad :-( ) (This means that you see the problem only from one side - other views will not be available to you)
Take a vacation, relax for a week! Believe me, without you, our planet won't collapse into a black hole in just one week! Prepare a piece of code that deletes all discussions/issues, etc after a certain date. Warn everyone that this is happening so they don't bother you! I think everyone will understand! You've done a lot for all of us! :-)
I believe that at this point in my statement this issue can be deleted!
But I also want to speak out: I'd tell you then: I understand perfectly well that it's impossible to 100% deleting a message from another node, since the other node could recompile the source code on their end, which wouldn't accept the message deletion command... BUT! The command to delete a specific message (from one node to another), I believe, must be transmitted. And such deletion will occur with a 99% probability, since most people use pre-compiled final programs and don't make changes to the source code before using these programs.
And finally, here's a universal truth: efficiency (coefficient of performance) can't be 100%! This means that the word "guarantee," which you use (and see my sentence as such in the code), can't even "guarantee" itself.
Also, in this context, it is worth saying that deleting even 1% of the number of messages will leave only 99% of the trace.
At first, can only delete text messages:
The simplest way is a command (without any IDs) (a simple command):
delete the fifth message from the end convHystory, containing "Hello world!"...
Yes, this isn't a precise command, but it's valid, since this conversation was created by two fully-fledged, equal nodes.
(Both creators of the conversation are creators of the conversation and have equal rights to this conversation.)
One of which wanted to delete something (or maybe the entire conversation at once)
= 0.0001% of the trace is erased!
Yes, I understand that, for example, can't delete messages from websites, from sensor commands, from voice streams, etc....
But tomorrow, a smart developer will come along and offer a solution for this too!
For example, in the help on creating websites, can suggest that developers have a good form
to have a deletion interface.
The simplest thing (even now) is to introduce one universal command "Delete()"...
And let the developer of each program, using the protocol, decide for themselves - what it means.
(This way, little by little, can move toward complete trace removal.)
I really hope that you will change your mind and engage in dialogue (with others) rather than making definitive decisions.
Here can consider that I found and described a bug in Mark itself. ))))))
Liam and team! Please close this thread if you feel the same way as Mark. (I won't be offended for this :-))