Don't take this the wrong way, but you spend way too much time worrying about what Hubski looks like and way too little time coming up with a way for me to skin it how I want. You have a much better framework than you have a website. Much like Wordpress, that framework needn't look the same for everyone. You've got like six themes, the best of which looks like a Toynbee Tile: And there's no possible way you'll ever come up with something that makes everybody happy. So instead of deciding what I should and should not see, why don't you invest some time in allowing me to make my own decisions? Hell - you can even mine my (and others') choices so you can see what your users actually want.
I had an idea for the site that others on the team suggested nobody would ever use, but I think you might. Currently you don't follow many people. What if you could follow a bunch of people and then click a button that "reset" who you follow and allowed you to start over, would you ever use that? I would use it. I think it would be fun to "reset" the list of people I follow and restart from scratch, including tags. What do you think? Would you or anyone else reading this ever use it?
I think I'd be more likely to use a "preset" - let me build my follows, then save it, then build another, then save it, then flip back and forth. For that matter, you could form "user groups" - a bunch of folx agree to be a curation group and if you follow that group, you get everyone in it.
Anybody can contribute to a tag. If you were to follow Knights_Of_The_Eastern_Calculus, you'd get me, myself, I, Manny, Moe, Jack, Huey, Dewey and Louie. By clicking on Knights_Of_The_Eastern_Calculus, you'd see everything all nine of us had shared across all tags. Which means Manny, Moe, Jack and myself might have to have a chat about Dewey constantly sharing those pictures of himself sitting on the copier. Keep in mind - this is not a fully-formed notion. This is a knee-jerk reaction to the idea of a "follow-delete" button.
About a year ago, we talked about something similar to this: unique tags that could be shared with specific users. Basically user-specific tags but with multiple users. Jesus, there have been a ton of weird variations that we have discussed. One that creeps back into my head from time to time is 'follower only posts' and/or 'follower only' messages. Only your followers can see them. I don't think these are good ideas. The possibilities from variations on what we have are dangerously large. But, I suspect some gems are in there.
HULK SMASH UNIQUE TAGS If I want to publish something that belongs only to me, I'll write a blog. This whole mental exercise erupted from the forehead of Zeus because thenewgreen asked me if I wanted a poison pill for my followers function. I said I'd rather have presets. Then I decided that presets would be even cooler if they were communal. It's pretty much the exact opposite direction from the way you guys are going with user-specific tags.
I like user specific tags because I can stop following "music" which often contains a sleuth of content I won't be interested in and follow #music.flagamuffin or something like that.
Again, spitballing: I'd probably run it like a modteam. Anybody could start one, and then it would be by seniority. Anybody in that group could add anyone (via invite) but the "top mod" would have the ability to purge anyone and everyone out of it. I have thought about this for exactly 30 seconds.
I think a function that is gaining usage but is still under utilized are personal tags. If you were to ask Dewey to use the #knightsoftheeasterncalculus and the #knightsoftheeasterncalculus.dewey tag you could ignore the #knightsoftheeasterncalculus.dewey tag and problem solved.
Here's something that I think all of Team Hubski is missing: The customization you guys are pushing requires more mental overhead from me. To make it work my way, I have to have a conversation. I set something up. Someone wants to look at it, they click it. Done. To make it work your way, I have to write #tag.modifier and in order for users to get my functionality out of it, they have to follow #tag and ignore #tag.modifier. It's like that list thing: in order to make it work, I have to remember where to put the colons and shit on the arbitrary six digit string that Hubski generates with the nomenclature that makes Hubski happy. Yeah, neat feature but I'm already suffering under Hubski's arbitrary (and often broken) markup. Now you're going to make me learn site-specific nomenclature? It's a very mk-up approach - "I can alter the codebase so that when I type this secret incantation things change." Except most of us don't want to carry a spell book around with us. We want to press the "up" button on the elevator and get to the 5th floor. We don't need to remember that typing 12356 in the stairwell keypad opens a secret passage to the dumbwaiter. Personal tags are not something I see any need for. Sorry. They're a voluntary way for an individual user to restrict things that works only if both the user and the viewer have perfect understanding of the functionality. The site needs to be easier to get around on, not more complex.
I get that there will inevitably be a learning curve but I'm all for making it more reasonable. That said, it's nice to use a site for a while and be able to navigate it on a fundamental level easily, but over time discover more functions that exist and are unique to the place where you are at. But I don't think these new functions need to be complex or unintuitive. I'm glad mk posted this, it's a good thread with some nice ideas. We will be having more conversations like this prior to making large changes etc. Also, we'll be having IRC time with the Hubski team where we can talk in real time about ideas etc... or what beer we happen to be drinking which is more the likely.The site needs to be easier to get around on, not more complex.
-This is not lost on me. At all. I have had a number of people, smart people, tell me that they tried to use the site but couldn't "figure it out."
Thanks for having this approach to developing this site further. A thing I've been thinking about lately: it might be a good idea to improve the relation between users. I find it quite difficult to go back and look up my previous conversations with people. I'd love to go back and read earlier conversations with people I interact, or read the best conversations, or the most recent. With some people around here, I know I've had good conversations with them, but I'm not grat with names so I'm not going back hundreds of pages into my own comments to find them. And I wouldn't know how to search for it. Ideally, if I went to someone's profile, I'd like to see more info about what our connection is. Like the 'how many posts have you shared by them' but going beyond that. Many things are possible, the conversations just being one.I'm glad mk posted this, it's a good thread with some nice ideas. We will be having more conversations like this prior to making large changes etc.
Hey veen, great suggestions. One thing really quickly that you might dig is that if you hover over a users name it tells you how many of their posts you've shared. After hovering over yours, I see that I've shared 10 of your posts. I like the idea of being able to quickly see more about our specific relationship to one another. mk
It's super subtle and just to the right of the name.
Yeah, I'm not a hero with names, and while I remember quite something about the people behind the usernames, I'd like to be able to go back and look at past interactions with that person. I've thought about the ability to tag people, like in RES. But that might be too intrusive to the layout. Ideally there would be a way to see what that person means to you, but I don't have a clue how to quantify that.I like the idea of being able to quickly see more about our specific relationship to one another. mk
Here's the horrible, ugly truth. If we each had a little image next to our names that was unique to us, you'd remember people a lot more easily. That said, we've discussed such things in the past and it's probably not going to happen. A very long time ago I suggested that people could have images but that these images would be automatically pencil sketched like in the WSJ. Probably a horrible idea, but my reasoning was that it would be in black and white and wouldn't make the page too distracting. I'm up for any suggestions though as to how to show relationship etc.
An image would indeed be easier, which is why I really like it that images can be added to bio pages. I for one always imagine humanodon's drawing when I read his posts, it gives a face to a name. But at the same time there's the problem I've encountered on many forums: I stop reading the usernames, and remember the image alone. People have disappeared for me because they changed their avatar. And in general they're ugly as hell. Let's get our goals clear. What I think the site will benefit from is a better way to foster relationships and connections, both existing and new. One way this would be achieved is to make them more recongnizable, e.g. avatars / images / anything visual. Have you seen this amazing idea before? MIT created a way for everyone to have their own version of their logo, using algorithms to generate 40,000 deviations. I'm not saying we should do the exact same but it is an interesting way to make people recognizable in a large operation. Another example: in WhatsApp groupchats, every person you chat with has a colored name. As I chat more with people, I started to quickly distinguish people because of that visual detail. After a while I stopped looking at the names above the messages and knew by the colors who wrote it. The first weeks here my mind tried to do the same, since I didn't have a lot of lightblue names. But I quickly made mistakes, e.g. confusing you with theadvancedapes because of the long, lightblue name that both start with the. Another way to foster connections is through the information you can know from someone else: right now I can only see how many posts I shared of you (19). It is where my idea for previous conversations comes in. There is a plethora of data that you can show to others. Which one is valuable is something to think of. Preferably, viewing someones profile, you would be able to assess who they are in relation to yourself. Maybe a number that says how many posts you shared that they shared as well. I'd love to hear more thought & discussion on this.
I like veen's thoughts. To add some more: how about being able to set a profile image, and if you hover over someone's name for > 1 second, a little 180x180 profile image pops up. (greater than a second so it doesn't pop up while you're mousing through a page.) Furthermore, how about a popup box when you click someone's username that has all of the info that normally appears in the left column of the profile page, so you can view someone's info without leaving the current page. You could then click their username from the popup or something to view their full profile with posts, shares, comments, etc. This would be similar to how the Twitter webpage (not mobile app) works when you click on someone's name.
mk, b_b, forwardslash, insomniasexx, Some interesting ideas from veen and doesntgolf. -By the way, have you ever tried golfing? It really can be quite fun.
Thanks! Parent twice for my comment by the way. And maybe it is a good time now to start a new post on this? I'm not gonna. I've golfed a bit! I had the chance to get 6 golf courses on stances and movement for free, after which we went to a pitch and putt track to put it to the test. I ended up coming back four more times there, quite like it as a relaxed sport. The satisfaction of landing a perfect shot is really great.
accepts as compliment I hear what you are saying, but this is one case where a user suggested to me that the design was significantly affecting usage. I think the conversation was worth having if only for the insight it provided. At this point, it feels like a push to me. There are good points both for and against a change. We did discuss moving the follow count up top to profiles last week as the badges did, and if anything, this might just give a bit more validation to that idea, as it's a pretty nominal thing to begin with.You've got like six themes, the best of which looks like a Toynbee Tile:
Don't take this the wrong way, but you spend way too much time worrying about what Hubski looks like and way too little time coming up with a way for me to skin it how I want.
Never, ever think I'm against conversation. Quite to the contrary - I'm against changes implemented without.I hear what you are saying, but this is one case where a user suggested to me that the design was significantly affecting usage. I think the conversation was worth having if only for the insight it provided.
Customization = good, but I think you picked a bad example. The themes are functional. hubski is essentially a forum, and I've never been on a forum that benefited from too much thought about its skins. (See: livejournal.) I've spent time on a thousand forums, and basically every time I tried some skins and defaulted to basic out of sheer necessity. What other kind of customization do you want? Usage changes? Specific UI stuff? I think I'm stuck in the reddit mindset where I can't see a reason for everyone's hubski to be slightly different.
I'm not saying I want to change the way Hubski looks to you. I'm saying I want to change the way Hubski looks to me. UI is important, and UI is personal. With Livejournal, you're impacting what everyone else sees (see also: Myspace). The reason Facebook won that one is that by forcing conformity on everyone, it set certain standards for legibility. (well obviously there was more to it than that, but still). However, if you take a look at Protopage, you can tweak it any way you want - it's just a UI that allows you to arrange things to your liking. I've been playing The Last Story pretty heavily and damned if I don't spend nearly as much time fucking with peoples' wardrobes as I do fighting monsters. And you know what? They built that in, and they incentivized it. And what it equals is more gameplay for less effort and - as previously discussed, The Ikea Effect. Customization counts.
Right, but how? hubski is a basic forum. The rule of function over form applies here, I think. Like you say, we have certain standards for legibility, and this is good. Anything else seems a bit superfluous. If our admins here spent all their time on this and were paid and so on then yeah, maybe I'd look for some behind the scenes customizability, but hell, a forum's a forum. We're here to read text and follow links.I'm saying I want to change the way Hubski looks to me.
Colors, fonts, etc. Then why are there six presets? The discussion, already in progress, is "messing with the look." The "admins" are, as previously stated, spending time on it already. My argument, if you'll read back, is that I'd much rather have the customization in my hands than theirs because then I'll get what I want without having to worry about anyone else's notions. Notions such as how "dark" is actually "asphalt."Right, but how?
hubski is a basic forum. The rule of function over form applies here, I think.
If our admins here spent all their time on this and were paid and so on then yeah, maybe I'd look for some behind the scenes customizability, but hell, a forum's a forum.
Okay. Now that I think about I guess I'd like some more markup. Hex color palettes maybe. Some cleaner menus (mk, the domains followed popup is messy as hell -- sorry to keep bugging you about insignificant shit but hey). I've only ever used one but I sort of assumed all that changed was the background color ... as in they all have the same function over form aspect. From here it just sounded like you were advocating for extremely personalizable CSS stuff, impacting where things were on the site, menus etc. -- what I would consider overkill for a forum. Maybe because I'm unfamiliar with wordpress I jumped to the wrong conclusion. Either way. I'm happy with what we have and would probably continue to be happy with whatever we ended up with.Colors, fonts, etc.
Then why are there six presets?