⭐ Merited by dkbit98 (1)
Now, I don't see any advantage in clicking this icon and only seeing my posts.
Yup, it's a little confusing, but I think it'll make sense to you when the time comes...The small feature you're referring to is basically just a throw-in (it's something that I can add, effectively "for free", once the rest of the code is in place). The broader feature is mostly about being able to "teleport" around large topics by focusing and then unfocusing on the posts made by a specific user (but, that's very difficult to appreciate until you've seen it in action).
For example, I often find myself in a position where I'm on one of theymos' posts in an old topic, and I'd like to jump to the next post (in that topic and if one exists) by theymos. Once you're used to that idea (being able from within a topic to one-click see just the posts made by a specific author), it becomes convenient to sometimes see just your own posts within a given topic (to refresh your memory of that topic from your own perspective, by seeing all the things you said, and who you said them to). For that particular use-case, the icon I was talking about ends up being a very natural place to hang a link (it stands to reason that if that's the icon that informs you that you've already posted in that topic, it should also be the icon that lets you discover what you've posted in that topic).
Speaking about ignore, I would like to see a patch that would improve and expand forum Ignore feature.
Yup, ignore-related features have been on my mind for a long time...I dipped my toes into this back when I agreed to look into a topic-ignore feature:
Hehe, yeah I'll look into this. 
I'm busy helping theymos with something that might keep me occupied for a while, and then after that I'd like to send him the 2FA patch that I've been talking about for months. Once those two things are done, I'll see about this proposal; haven't thought too deeply about it, but it seems like a good idea to me, and I think it's something that I would personally use.
I think maybe a logical place to put it would be here (bottom right):
And then, obviously something in "Modify Profile" to maintain the list of topic IDs.
I'm busy helping theymos with something that might keep me occupied for a while, and then after that I'd like to send him the 2FA patch that I've been talking about for months. Once those two things are done, I'll see about this proposal; haven't thought too deeply about it, but it seems like a good idea to me, and I think it's something that I would personally use.
I think maybe a logical place to put it would be here (bottom right):
And then, obviously something in "Modify Profile" to maintain the list of topic IDs.
The thing is, my to-do list has since picked up enough ignore-related items that the approach I'm most drawn to at this point is to just scrap the existing features and replace them with a new unified ignore system that could handle everything [1]: users, boards, topics, etc. I'd even like to add ignore expressions, so that you could (effectively) say something like: "In my topic in the Lending section, only show me posts made by Jr. Members and above."
That previous thought-arc is actually pretty common: With a lot of the suggestions I've internalized, I think about them on their own for a time, and then I think about them on-and-off in the context of all the other ideas from that same universe, and then I sometimes come up with something cool that would tie the whole collection of ideas up into a neat bundle with a nice pretty bow, except for the fact that it cuts across code that I don't have access to, or would involve too much code, either for me to safely/sensibly write or for theymos to comfortably review... so I put the whole thing on ice.
SMF has just about the worst possible architecture for feature development: It's much harder to make bug-free modifications to it than I think most people can really appreciate. By far, most of the time I've spent thinking about Bitcointalk has actually been dedicated to finding a way to lay the technical foundation for me to work a bit more independently and much more safely, but, for now I'm basically stuck working on things beneath a certain complexity threshold (which is why I titled this topic the way I did).
[1] xkcd 927 doesn't apply: I'm talking about replacement not addition.
