App Design Philosophy: Informing and empowering users rather than attempting to control them

[Oshyan]

This has come up a number of times in various interesting and potentially valuable idea posts, so I want to make a stand-alone topic to see if there is general agreement on this important principle.

As an overall app and system design philosophy, I want to strongly encourage good behavior and positive, active engagement in the platform. But there are very few things I think we should truly mandate. I am really in favor of leaning toward informing users, both those who are doing something that may limit their visibility or engagement, and those who are looking for people such that they can make an informed decision whether to A: include those people in their searches/feed, and B: decide whether and how to interact with them and how to set their expectations if they do see/match.

There are a few things we might reasonably make a requirement, for safety or if that choice would clearly make a major difference to matching potential and quality, positive engagement, etc. But there are a couple major problems with mandating things in general so we should be cautious and limit the number. Two significant considerations that come to mind are:

1: The most obvious is that we will already have big challenges getting enough critical mass of users to make an effective dating platform. Anything we mandate is likely to turn off some proportion of potential users. So the more we mandate, the lower our overall user numbers are likely to be. Now there are some people we just don't want on the platform, of course, and I think any actual user/profile requirement we implement should be aimed at those specific, problematic people. But again we should be very careful and considered in doing this.

It's also worth noting that often the very fact of something being "required" will turn people off (i.e. some people just don't like being controlled, even in some cases if they might agree with the basic "reason" for that thing not being optional). Yet if that same thing (e.g. turning off notifications) were simply handled as information to the person, but they could still do what they want, many who might otherwise not have signed-up on principle may still end up using the platform. In the case of disabling notifications #133 some would probably proceed with notifications off after accepting the consequences, but I bet many would ultimately leave them on as the entire point of the "soft" (user-informing) approach is to encourage good behavior. We want people who want to turn off notifications (or any other behavior/choice that we feel is against this or other user's interest) to second-guess that choice frequently, but still to be able to make that choice so that it doesn't feel like we're controlling or infantilizing them. Users who feel informed generally feel empowered, and this is a good thing.

2: We don't know what all users want or what their tolerance level is for various behaviors, etc. For example, while having to be persistent is arguably an unwanted feature of many current dating app models, it is nonetheless a very real aspect of modern dating, and some people are persistent and end up finding good relationships by doing so. By "persistent" I mean, in part, messaging 1 or more times after the other person has stopped responding, for example. So if we were to consider banning people for ghosting, for example, that might limit these persistent people's ability to engage in the way that has worked for them. Now of course there is the danger here, too, that this "persistence" might be seen as creepy, abusive, harassment, etc. but I think there are other tools (such as unmatch, report, etc.) to handle that. Point is someone should be able to look at a profile and say to themselves "I see this person doesn't respond much, but they sound really great, and they're just my type. I'm going to risk it", match with them, and then continue to message for a while even if that person stops responding. They should be able to make that choice, in my view, as long as we're clearly informing them in the process.

I've also spoken to this previously, e.g. here:
#118 (comment)