For how it distributes physical advantage, it will forever be a relevant metric of security, if only a metric here. A search makes finding the split in overall weaker individuals easier, for either good or bad, and the difference made in knowing what people select is a bit more involved.
Moving a free-form field to “Male”, “Female” and “Non-binary” really means only the latter is conveyed, and no further nebulous options.
Seeing as people have no difficulty deciding sex based on profile picture, it makes people opting for the third category searchable. How this is better for security I don’t know, but it certainly narrows it right down, or splits your candidates down the middle for convenience or malice.
Relevantly, this setting is not user-selectable, which means someone thought establishing entropy there was important for reasons I can only assume to be security-related.
In the case of females or males trying to be something else,
males carry their natural physical advantage over women.
These will largely be visually obvious, and make up the non-binary category even if not.
Allowing people to discriminate based on this metric in search makes it even easier to single out these in the greater pool, and also easier to find. I don’t see that as largely better.
If it isn’t a category at all, then there is no need to “misrepresent” ones non-binary choice for this reason. If it doesn’t have to be selected, it is a passable excuse not to make it, but then it can’t be required like it is now.
With males carrying advantageous strength into either other category, further it has to be biologically based for it to make a difference when those select “Female”.
“Man” and “Woman” doesn’t make a difference to “Male” and “Female”, and “masculine” and “feminine” is only secondary categories within that to people.
There is some personal pronoun setting, with “he/him” “she/her” and “free form”.
If the “gender” isn’t biological in the sense that personal pronoun follows, the system is rigged towards some level of stable choices, and giving away anomalies in there and in pronouns as much as possible within a user-choice system. Though right now it is right next to gender, so it tricks people that wouldn’t flag pronouns into doing so. Almost did it myself.
Whether by rationale by breakdown of crime-statistics or other underlying thought, I am not privy to it or the strategy of using it.
Will the user be prevented from changing the “gender” option when there is a pending negative review?
As the attacks to worry the most about are largely of a sexual nature, it makes a lot more sense to allow distinguishing oneself by and subsequently searching by sexual attraction. That carries the disadvantage of being protected information, so special consent is needed to collect and display it. The clever way is to have a group called “Rainbow” and then let people search by both.
I think that makes it a less convenient dating-app than its functionality would allude to, and gives more up-front info, if not a better or additional way to navigate ones security.
If profile picture is all one has to go by, people will scrutinize it more in depth, and it will be harder to pretend to be someone else than selecting a setting. That isn’t very hard, with makes it not very reliable. If you want to be anonymous for security reasons, that is largely going to select for females, and those can more-so operate without a profile picture.
It isn’t harder to fake a profile picture than it is to change a setting, but ultimately no political correctness will save anyone from using a non-available picture of someone else.
Thus, the ratings are valuable, but I think pronouns and selecting gender may create more drama, confusion and mistaken identity than benefit.
In planning to meet a woman without a profile picture, one would likely think that is fine even with no rating, I think a system of categories that can’t be changed easily lull people into thinking a “female” or “woman” has to be something other than a selection, especially since interpreting broadly from available info is the human condition.
Short of a full and actual review, the conditions for IRL damage are not met.
Having someone send illicit messages up front is better than the same ban with more interaction and reviews to be written.
Mitigating malice is easier by not only rewarding, but requiring sunken cost.
It is not only smart to go with people with extensive descriptions in their reviews because you meet a less random person that fits you better, but a potential ill-doer is looking to give away as little hard info as possible so as to get away or to be able to do it again as habitual as one allows navigating social hierarchy to be.
Writing full messages has the same vectors. At worst it gives more info to go by, and it is harder to fake.
Moderators could ask to have evidence made that you are the person in your profile picture.
Asking to have a picture of yourself or video-chat of holding up a pillow to the left of your face is harder to fake on the fly. Or something similar to it could be asked for to make an avatar.
There are plenty of ways to weed out fake profiles, but that is for another day and not something to put in recipe form.
AI can already distinguish both sex and sexual attraction by a meaningful profile picture alone.
TL;DR I am not sure pronouns and s/electable gender isn’t just bloat, when it could be free-formed into the open profile sections anyway. There it is an active choice, with which linguistic discretion can be employed if need be.