Howdy @nxmc. Slant founder here, we’re building pretty much exactly what you described. Agreed it’s way too much work at the moment.
The issue I’m struggling with at the moment is how to deal with edits to pros that are synced to 2+ questions. In what conditions do edits get synced and in what conditions do they get forked? It’s ‘impossible’ for us to know if an edit was just fixing an spelling error or if it changed the context of the pro for a particular question and in the process made the pro in other questions no longer make sense.
Maybe list-dependant annotations for the pros/cons? Unless this just makes too much work. Perhaps editing pros/cons would have to be on the global page to let the user know that they’re editing globally, too.
Okay so if the option and it’s pros/cons are global allow the editor/user to make the choice of whether the edit is global or specific to that question. So before “submit” is clicked have a check box of some kind and a warning/explanation that allows the user to submit globally or specific to that one question. Basically allow the user to decide during an edits submission whether it gets forked or not.
Can’t imagine a UI that would make that intuitive to a brand new user. How would that user know if it should be global or not? Even as a moderator feature I’d worry it’s too weird of a concept to load the user with.
Maybe we have levels for users and that feature is earned with experience. So on a normal user level it just changes in the one question but then a mod (or users with the appropriate level of experience) can easily add it to all options. Thought the bigger the site gets the more mods would have to go through all changes.
Personally I do not think it would be a huge deal as long as there was a blurb that explained the option (either as a pop up when hovering over the checkbox or an explanation in red explaining it near the box) with a certain amount of concise detail. Really at some point (just like reddit subs and Wikipedia) explanations will be needed for how the site and its features work. We could have a popup for new IPs that tells users to go wherever to read the rules or at least have something in the sidebar that tells users to go read before posting. Maybe even a popup when users click on anything that allows them to contribute that says go read these rules or just lists the rules for that specific area.
Ideally we would want the site to be more intuitive but I think realistically for what we want the site to offer it may not be possible without having a few explanations here and there.
Maybe we have levels for users and that feature is earned with experience. So on a normal user level it just changes in the one question but then a mod (or users with the appropriate level of experience) can easily add it to all options. Thought the bigger the site gets the more mods would have to go through all changes.
I really like this idea, actually. Stackoverflow style, the less proven knowledgeable, the smaller scale the effect is.
So some Slant history. The initial versions of the site had ALL changes go into a pending queue. The only people who could accept the change were mods and existing contributors to that option. The idea was to prevent spamming by having a barrier in place and to educate new users about their changes before it went public. We scrapped this as the barrier was too high for new users and it’s actually a complicated of a system to resolve conflicts etc. But this means we already have the tech to have pending changes with an approval system.
One idea I’ve been toying with is changes made to a pro under some conditions get automatically synced (source changes, image changes, edits under X chars) and edits over X chars get put into a pending queue for that pro. Contributors would then be notified and could accept the edit.
The flaw in this system are changes like splitting one Pro into two. Or two Pros into one…
So what if we tweaked that even further to where the changes made over a char limit does get changed in the single questions option but does not sync to the site wide option without an admins approval? That way the specific edit no matter how large still has an effect for the user while we still retain control on what happens to the site wide option that is in numerous other questions. Maybe with that an option in the mod panel to easily revert such changes (to the one option in the one question) if they are deemed too much of a change or just not useful information over the last.
As for the splitting problem, maybe that could be resolved with the upcoming flagging system. A user can flag that particular option with a declaration of wanting to split it while also making their case as to why. Then either mods or heavy contributors to that particular subject can decide on if it is worth it, or even maybe create a voting system for users to vote on whether such a change is worthwhile. Of course the voting would not be really useful at the moment without a massive amount of users to actually vote.
Will the user be notified that the edit is only going to appear on that one questions option and not the global option? As it may not be needed information for a new user who does not know any different but for other users, they may be left wondering why the edit did not go through globally. Which I think brings us back to explanations. There is a lot of rules and behaviors that are going to be unknown to new and regular users and I think that is going to be unavoidable for what we want the site to do. I think we could use a place or places that explains the site, its rules and its actions. This could all be in one sport or appear at the appropriate time when that action is about to take place. In this instance it could be good to have something that states in the submit window that edits of a small size go global and over a certain size stay local until a mod or certain level user approves it.
Even Stack Overflow has a small blurb explaining the site on its front page and we are trying to attract a way more broad audience than them.
I’m 100% supportive of explanations in general, but in that exact use-case I don’t believe one will be needed. New users don’t need to know about or understand the global system IMO. As far as they are concerned they are unaware the Pro exists somewhere else, and they don’t need to know about it’s existence to make their edit.
I’m banging my head against this again right now. The really maddening thing is how our information is balkanized. This idea of each question being its own scope is good and I want to see it work, but the damage caused by having no cross-referencing is severe.
I feel like this needs to be very high priority, as it affects both contributors (frustrating duplication) and consumers (potentially useful info buried in corners). And while I think Stuart’s got the right idea for how to handle edit scoping, it also needs to be obvious that the option itself exists independent of any given question. So just as the question summary list allows you to browse to individual option pages, there should be a way for the user to browse other questions where that option is presented, and see pros and cons in that scope. I’d even go so far as to propose a question comparison view for options, because that seems like the only sane way to allow users to copy things between them. (Perhaps this could be worked into the option’s main Edit page, since add-an-option is a clone of that currently, and if we’re going to allow copying in pros/cons at option “creation” it’ll have to go there anyway.)
I’d just like to mention that I also think this is a really important issue to address. The situation I’ve seen is when something has a pro of “project is available on github” when just about every other competing thing in that question is also available on github. But because someone only bothered to put it into one of them, that thing gets an unjustifiably inflated pro count. You might say that this is just a bad pro, and in this case I would agree with you. But I think that the issue is much broader than that. Some types of questions seem to have many pros/cons that apply across the board to every alternative. It seems like these things would be better handled by creating a matrix of “characteristics” upon which every alternative is graded.
We have a potential solution for what you’re saying that I’d like to get your opinion on. “Available on Github” is a boolean/quantitative characteristic of a option and thus doesn’t make a great pro. Our solution for that type of information can be seen on a few test questions like this one:
So a potentially tidy solution would be to move all the pros/cons that aren’t subjective/qualitative to that matrix. There is however one catch. If there are say 6 options, and only 1 of them has github support, that’s a pretty important pro to some people and shouldn’t be buried in the matrix. The flipside is also true, if 5/6 have the support, the one that doesn’t should have a con pointing that out.
Ahh, your compare view is definitely along the lines of what I’m thinking of. I think it would be nice if you had a much more condensed view so that you could see more of them on one page at a time. Other than that, the challenge is to get rid of pros and cons that should actually be in the matrix.
I was thinking of an idea in regards to UI and adding & modifying several areas, however just within this category alone (feature suggestions) my idea is common to other categories as well. In other words and I didn’t know where the best place to post my idea / question. So I figured to post it here due to the common, in a sense, for the name for this topic of discussion - gloabal posting it.
Anyway, having said that, in reply to the last post and other posts in here, how about having some sort of accordion style, wiki-ish (expandable & collapsable) UI in regards to all the questions raised in this discission. If needed for to clarify my idea, I can into more detail on this. Like I said, this has been an idea on my mind for a few days but it applies to various topics here so I didnh know jis to best place it and how to word it in the right way without creating any confusion.