2y ago

What are your thoughts on the idea of adding an edit history feature to posts, and comments in Lemmy?

I believe that the addition of an edit history would be a massive boon to the usefulness of Lemmy on the whole. A common problem with forums is the relatively low level of trust that users can have in another's content. When one has the ability to edit their posts, and comments this invites the possibility of misleading the reader -- for example, one can create a comment, then, after gaining likes, and comments, reword the comment to either destroy the usefulness of the thread on the whole, or mislead a future reader. The addition of an edit history would solve this issue.

Lemmy already tracks that a post was edited (I point your attention to the little pencil icon that you see in a posts header in the browser version of the lemmy-ui). What I am describing is the expansion of this feature. The format that I have envisioned is something very similar to what Element does. For example:

What this image is depicting is a visual of what parts of the post were changed at the time that it was edited, and a complete history of every edit made to the post -- sort of like a "git diff".

I would love to hear the feedback of all Lemmings on this idea for a feature -- concerns, suggestions, praise, criticisms, or anything else!

This post is the result of the current (2023-10-03T07:37Z) status of this GitHub post. It was closed by a maintainer/dev of the Lemmy repo. I personally don't think that the issue got enough attention, or input, so I am posting it here in an attempt to open it up to a potentially wider audience.

59 comments

Editing a post may be to remove the password or email address you accidentally copy pasted in, or removing some potentially doxxing information, or one of many reasons you want that content gone. Github has edit history, but it also allows users to delete revisions so it seems your main concern would not be resolved by this implementation.
And as you point out, there is already a message that says the post was edited and what time.
Overall I don't see that the benefits outweigh the new issues caused.
- Sure, this is true for any public website. But at least editing it out is a form of damage control. Being able to edit and federate the change to most servers makes the problem a lot smaller.
  If edit history existed and you couldn't remove an entry, the only damage control would be to delete. This is also acceptable but I haven't seen a good argument for keeping the history yet.
- You could make it so there is a checkbox for deleting the edit history, so only the fact that it has been edited remains.
  
  To draw attention to an edit, for example to correct an erroneous statement, use a combination of strikethrough and bold (or italic if more appropriate):
  Joe Hill, who wrote songs about union organizing, was framed and ~~hung~~ executed by firing squad by the state of Utah in 1915.
  Joe Hill, who wrote songs about union organizing, was framed and ~~hung~~ **executed by firing squad** by the state of Utah in 1915.
  
  OP's argument is that people can hide that they have edited. While I'm not against the suggestion, it wouldn't solve the original problem.
- Editing a post may be to remove the password or email address you accidentally copy pasted in, or removing some potentially doxxing information, or one of many reasons you want that content gone.
  Why not just delete the post, and then make a new one with the correct information?
  Github has edit history, but it also allows users to delete revisions so it seems your main concern would not be resolved by this implementation.
  If this were to be allowed, the edit history would then be pointless.
  And as you point out, there is already a message that says the post was edited and what time.
  That is the only information that is provided. One is unable to find out what was changed.
  
  Why not just delete the post, and then make a new one with the correct information?
  Sure, but then your comment chain doesn't make sense, or if it's a post them you lose all the comments.
  If this were to be allowed, the edit history would then be pointless.
  I disagree, but I do think it invalidates your reason for having an edit history.
- How much extra storage and bandwidth would it use? It's just text. It's clear that you don't like the idea, but I don't find your reasons convincing.
- Well, I was of the other opinion, but you have decisively convinced me, thank you.
  Thoughts on the cons of making all upvotes visible to all users? I've always felt that would be a net benefit, but now I'm not so sure. Any idea?
- What you think adds a feature actually takes away a feature (being able to edit posts without the edit being visible). That isn’t a bug, it’s a feature.
  Do note that a feature's mere existence doesn't necessitate that it must be a good feature.
  Increased hosting costs to operate it (storage)
  I don't believe that this is much of an issue, as text is extremely cheap to store. It would, of course, be false to state that it doesn't increase the cost at all, but I would argue that the increase in cost is most likely small enough to be of little concern. Let's make a very basic, and not overly precise example: Say, on average, there is 100 words in each Lemmy post's body. And say, on average, that a user will edit 10 words. Now, say that the algorithm that generates the changes, only stores the changes relative to the previous content, so we can then simplify this to say that it only stores the text plus, say, maybe 1 extra words worth of data for location, and linking information. So that means that each post will only add on maybe 11 words on average which would equate to a 1.1% increase in text storage requirements. Given that all of Wikipedia's Engish article text is around 20GB, a 1.1% increase in that is only about 220MB -- one should be able to see that the equivelant for Lemmy wouldn't be that terrible.
  Increased API calls and sizes (bandwidth)
  I'm not sure that I am qualified enough to make a comment on this, as I am not at all an expert in how Lemmy's (or ActivityPub's) Networking works under the hood, but how would this differ from how it already works? You can already make an edit, so the number of API requests should stay somewhat the same. The only thing I can think of is that when someone opens the edit history, they would need to make a few API calls to retrieve it all, unless all that could be retrieved in one call, then it should be the same as displaying the date of the last edit which is a feature that already exists with the only difference being the payload size in that case.
  99.999% of feature use is just typo correction
  Sure, but I don't see this as a counterargument. The whole point of it is to be able to verify that it is indeed a typo correction.
  99% of users won’t use the feature
  True, this could be seen as an investment that may not be worth it as it would really only cater to those who are, perhaps, on the upper end of paranoia, or overly persnickety.
  It invites users to review people’s edit history and nitpick/call out things that the poster edited out for a reason…
  This is a fair point. I hadn't considered this. I do think that it wouldn't be super common, it is indeed a possible issue.
  Which in turn breaks down and chills conversation as users have to be overly careful that their comment or post is 100% accurate to avoid getting nitpicked, that they fully agree with what they’re saying as they can’t take it back or edit their stance/opinion in the future, that they don’t reveal anything sensitive by mistake
  I mean, it's kind of already like this, is it not? What you say is certainly under scrutiny by the court of public opinion. Unless you mean that one cannot take something back because it would be ingrained in the edit history, but, to that, I would say that one can still delete their post.
  It invites abuse from mods by reverting edits and dictating which “version of truth” of a post is the one that everyone sees rather than the user being in control.
  Hm, I think this is a completely separate issue. A mod, or admin should not be able to do such things. This actually brings up a separate idea that I had where, ideally, a post would be signed by the user who wrote it so that one could be certain that it was the user who indeed wrote the post, and that it was not modified by an admin, or some other external entity. This censorship is an existing problem with no solution.
  Extra UI cutter is needed to handle the feature
  The button that would contain the history already exists in the form of the edit pencil that posts have. Unless you mean the diff itself, but that would only be visible if one toggles it.
  If a user posts credentials, they have to delete the entire post or comment and even then, the backend server very well could still have that log saved in a backup (legal ramifications)
  Yeah deleting would be the only option -- personally, I don't see this as a huge issue, but that's just me. As for the logs, they could already exist for a deleted post anyways. When you post something online, there really is 0 guarantee that you can ever remove it. Generally, one must accept that whatever they put online is out there, in some capacity, forever.
  Users could abuse the feature to e.g. share links to abuse material and hide it in the log requiring moderators to have to review all messages and all edit histories, greatly increasing their work load, especially if users constantly edit their posts to make moderators jobs harder to sift through all the edits to reveal what they did.
  Good point. I hadn't considered this issue. I would argue that it's the most important point of your list. I'm not sure that there is anything that could really be done about it. It would essentially have to rely on someone reporting it after having gone through the edit history, or a mod just happening to have gone through the history themself.
  will have a direct, chilling impact on all other users.
  Aha, you don't need to use such melodramatic language to try to magnify your opinion -- your counterarguments should be enough.
  if you need audit logs, you do it behind the scenes not in the UI
  Do note that this is supposed to be for the benefit of the user, and not the admins. A user cannot access logs.
  Visible changelogs on information chat / social systems make people talk less, not more.
  I would like to know your source for such a statement.
  And given how Lemmy is still in its infancy and hasn’t reached a critical mass, adding a feature like OP proposed could make Lemmy a far less inviting place to socialise.
  This is a purely subjective statement, I would argue.
- increased hosting costs
  Should be minimal since it's text. In fact, a lot of my edits reduce posts since I use it to add an edit that I would've needed to post in multiple sub-threads.
  99% of users won't use the feature
  Which further proves that it's not likely to cause many hosting costs.
  invites users to review people's edit history
  They already do this with comment history. If you don't want people digging in to your edit history, don't make controversial edits.
  People being jerks for calling out typo fixes likely will result in downvotes, thus discouraged by the community. Look at grammar police, they're frequently downvoted to the point where they're not very common (though more common than they should be).
  be overly careful that their comment or post is 100% accurate
  First, that remains to be seen. You yourself said 99% of people won't use the feature, and I think it'll turn out much like the grammar police, people calling out others for small mistakes will be shunned. I could even see mods making and enforcing harassment rules related to behavior like that.
  Second, if it improves the quality of comments and posts, I don't see that as a bad thing. Perhaps individual communities could disable it, but it should absolutely be enabled for serious communities that cover politics and news.
  abuse by mods by reverting edits
  Then don't give them that power. Just allow them to lock posts and leave a note or a flag to warn users of abuse by the commenter.
  Extra UI clutter
  Not necessarily. You can pick a client that doesn't implement the feature. Or you can have it be an optional feature, or hide it by default in an expandable menu. It doesn't cause clutter in Wikipedia, so it's not inherently a poor UX choice.
  We can bike shed the UX once we agree on the functional requirements, that's how the design process is intended to work.
  If a user posts credentials
  This is a federated platform, you should assume everything you post is there for good on some instance.
  Users could abuse the feature
  Sure, but they can do it anyway in the clear by sending DMs, changing text of links to look innocent, etc.
  I think there should be an option to show edits always, which would catch this issue. So basically you'd be looking at the equivalent of inline git diff (with strikeouts or whatever to show deleted content). That's how I'd prefer to navigate Lemmy, and I'm guessing enough others would as well to catch any attempted abuse.
  less inviting place to socialise
  Then I guess you and I see the platform very differently. I see it as a place to discuss news and politics, not a place to "socialize." It's a link aggregator, so I expect the bulk of the discussion to be about the content of links.
  That said, there are plenty of casual communities that work more like forums that want to foster casual discussion, not serious discussion. For those, edit history should probably be disabled. So make it an opt-in thing by community so those of us that want it can have it.
  
  increased hosting costs
  Should be minimal since it’s text. In fact, a lot of my edits reduce posts since I use it to add an edit that I would’ve needed to post in multiple sub-threads.
  If you make a post which is 1000 chars in length, then you edit it to be only 800 chars, the 1000 chars still need to be stored. And federated and everything. That is the actual idea being presented here. It might not be a total of 1000+800=1800 chars because there are clever ways of compressing stuff, but it is still 1000 and certainly 800. And as @fartsparkles also pointed out you need to track meta data for each edit in addition to the text.
  It doesn’t cause clutter in Wikipedia, so it’s not inherently a poor UX choice.
  Interesting comparison. Wikipedia has a very robust system for tracking changes, because it is a core feature of the project. It is a collection of collaboratively edited documents. Since that's the whole idea of the project, they have rules, software, code, humans, robots, meetings, arguments, computers, etc to manage it because it is really complicated.
  Sometimes, it is too much and they just wipe it away https://en.wikipedia.org/wiki/Wikipedia:Selectivedeletion
  Threadiverse is not a collaboratively edited collection of documents so why introduce that? There is no compelling argument presented.
  Also mentioned is git, which like wikipedia is primarily a tool for collaborative editing. It also has the ability to permanently remove: https://git-scm.com/docs/git-filter-branch Not to mention using git is a very specialized skill primarily attained through formal education and employment.
  Both wikimedia and git are known as very complicated to use pieces of software which take years of practice to be good at. Both have their own subcultures. They have to be like this because they are trying to accomplish a complicated task, which is to allow large number of people to collaborate together. I think compare/contrasting these to threadiverse does a great deal to show what actually happens when you need to have changetracking like this and how difficult it is to design properly in such a way that it can be easily used by a common person without significant study.
  
  99% of users won't use the feature
  Which further proves that it's not likely to cause many hosting costs.
  This is a good point -- I missed that.
  invites users to review people's edit history
  They already do this with comment history.
  What do you mean by this? You can't see comment history currently.
  If you don't want people digging in to your edit history, don't make controversial edits.
  Hm, well, an edit is only controversial if you know that it was edited in a controversial manner. You wouldn't look in the edit history because you knew that it was controversial, you would look in the edit history and find that it was controversial. Unless, you meant to say "controversial posts" to which I would say that I disagree with that opinion.
  People being jerks for calling out typo fixes likely will result in downvotes, thus discouraged by the community. Look at grammar police, they're frequently downvoted to the point where they're not very common (though more common than they should be).
  This is a fair point.
  I see it as a place to discuss news and politics, not a place to "socialize."
  This is a rather one-sided/dubious statement. For one talking about news and politics could be deemed as socializing, plus a forum is just a medium of discourse in the general sense -- it doesn't really have any explicitly defined topic unless stated by an individual communtiy.
- better done on the client side with a browser extension
  or fuck it, install a keylogger on yourself
Nah, never liked the feature, wouldn't appreciate it here.
Side note, external images can be embedded in markdown like this:
![alt description](https://example.com/cool-image.png)
- Nah, never liked the feature, wouldn’t appreciate it here.
  Would you mind elaborating on why you feel that way?
  Side note, external images can be embedded in markdown like this:
  ![alt description](https://example.com/cool-image.png)
  Thank you for that info! I'll update my post.
  
  It adds nothing to the discussion. Use cases where it would have been useful I can count with my fingers. I made many more edits due to typos and brain-farts (that made the sentence look like I just learned English yesterday) than that.
  Edit: Also, I'm hosting my own instance (for others as well) and the (unoptimized) storage use is already huge. No need to pay for something I don't really care about.
Personally I like the idea of that history simply because I have seen people go back and edit their posts, as a form of trolling by getting into an argument with someone, and then changing their posts to completely obfuscate what the argument was about
It's not something I would care about or ever use. It comes with significant unresolved problems already pointed out, and it mostly just seems like you want it for reasons of idle curiosity or paranoia.
Most importantly, if a lemmy dev already said no, and you aren't willing to do the work, then it's dead, and opening a thread about it isn't a helpful way of fixing that.
- It’s not something I would care about or ever use.
  I think it's better to look at this not from the perspective of one's own personal gain, but the benefit that it provides to the site on the whole.
  It comes with significant unresolved problems already pointed out
  Would you mind stating the exact "unresolved problems" that you are referring to?
  it mostly just seems like you want it for reasons of idle curiosity or paranoia.
  I believe that the feature's existence provides the passive benefit of increasing the average quality of posted content.
  Most importantly, if a lemmy dev already said no, and you aren’t willing to do the work, then it’s dead
  What's bothersome about that is that the dev didn't just say that they didn't want to work on it, they closed it. I completely understand if the dev doesn't want to work on it personally, but closing it gives one the feeling that future discussion on the topic is not wanted -- not to mention that it also greatly reduces its visibility.
  opening a thread about it isn’t a helpful way of fixing that.
  No, but I wanted to have more discussion that what was had on GitHub. I figured that posting about it here would yield a much larger audience, and, perhaps, less biased opinions.
  
  What's bothersome about that is that the dev didn't just say that they didn't want to work on it, they closed it. I completely understand if the dev doesn't want to work on it personally, but closing it gives one the feeling that future discussion on the topic is not wanted -- not to mention that it also greatly reduces its visibility.
  This is the part where you should recognize that its not a feature they want on lemmy. It doesn't need "more visability", it's their project, and they get to choose what they want to do with it. We just use it as a byproduct of it being free and open.
I actually don't think it is required to trust people on a forum in the way you suggest.
If I was in what I perceived to be a really high stakes discussion (read: flamewar) where I was worried about this, I would take my own measures to ensure I could "trust" the other parties. I would save my own copies locally. Reddit RES had a button you could add client side for just this kind of petty bullshit. If you really want the feature, implement it in your browser/device.
Really though friend, try to have a bit of a sense of humor and distance from your online posting and interactions with unknown people. If someone is going to such lengths as to edit their post so it looks like you are responding to something else to make you look bad, it is either: a) a boring joke, or b) they are really pathetic and sad trying to sabotage you. Either way, it's not the end of the world. If it sticks in your craw, you can just go edit your comment to say "edit: the comment to which I am replied was substantially edited after I posted so what I said no longer applies". You can either delete what you said, or correct it, or leave it as-is with a caveat.
- Your post made me realize that I haven't heard the word "flamewar" in a long while.
- I actually don’t think it is required to trust people on a forum in the way you suggest.
  Why not try to improve it though?
  If I was in what I perceived to be a really high stakes discussion (read: flamewar) where I was worried about this, I would take my own measures to ensure I could “trust” the other parties. I would save my own copies locally. Reddit RES had a button you could add client side for just this kind of petty bullshit. If you really want the feature, implement it in your browser/device.
  I don't really understand the argument hat you are trying to make. You are admitting that this concern is justified, and that there are scenarious where one could be expected to want to take such measures, but you don't want a feature for this built in. Instead, you'd want a 3rd party plug-in...? I must ask: Why? Also, TIL about Reddit RES. Neat.
  If someone is going to such lengths as to edit their post so it looks like you are responding to something else to make you look bad, it is either: a) a boring joke, or b) they are really pathetic and sad trying to sabotage you. Either way, it’s not the end of the world. If it sticks in your craw, you can just go edit your comment to say “edit: the comment to which I am replied was substantially edited after I posted so what I said no longer applies”. You can either delete what you said, or correct it, or leave it as-is with a caveat.
  The point that I am trying to make isn't that this is for my own benefit, it is that this sort of behaviour detracts from the quality, and usefulness of the information on this site on the whole. Information shouldn't be purely ephemeral. The reliable exchange of information on forums is invaluable in the modern age. I couldn't even hope to count the number of times that I have gone through old forum posts reading people's opinions, and conversations when conducting research on a topic, or troubleshooting an issue.
  
  I have been on lots of old forums too. That is irrelevant to this thread. This thread is about the ability to investigate the typos on the old forum posts. How often are you on some phpBB site thinking "this would be so much better if I could see what incorrect information was edited out in 2009."? Nobody fucking cares.
I really appreciated reddit's ninja-edit window, where you had about three minutes to fix typos and grammatical errors without getting the this-was-edited indicator.
The root shortcoming is that changing one letter gets the same flag as replacing the whole comment or adding a wall of text.
- The root shortcoming is that changing one letter gets the same flag as replacing the whole comment or adding a wall of text.
  Fair point.
- Your are absolutely right. I guess I had accidentally copied the wrong link. Thank you for letting me know! I have now updated my post.
Edit: Comment deleted
Considering that, especially with federation, editing or deleting stuff doesn't really make the old version go away, I personally think the better move would be to pull a Slashdot and just not have the ability to edit or delete at all to begin with. That way, it makes it clearer what is inherently true: that once you post something, it's out there for good.
- I disagree, I like editing comments to fix typos, add links, or lift up responses I've made to popular posts so I don't have to repeat myself in a half dozen sub-threads.
  We should just make edits create a new post links to the previous one and clients can offer a diff feature.
- disagree because for example
  reading sequentially posted items when the original author wishes to correct themselves is really annoying
  there are reasons other than personal embarrassment someone would want to correct something they'd said. like if you give advice but later realize the advice could be dangerous for a group of people you hadn't considered

59 comments