What do you guys think of these proposed features?
Hi folks, want to get feedback on these set of features:
1) Aggregated followability index, based on his twits and network
2) Summarized labels stating his characteristics
3) Conversations with your friends
4) Filter - see people who are at your "influence" level, not famous people
Screenshot:
Screenshot (with labels):

1) Aggregated followability index, based on his twits and network
2) Summarized labels stating his characteristics
3) Conversations with your friends
4) Filter - see people who are at your "influence" level, not famous people
Screenshot:
Screenshot (with labels):

54
people like this idea
I like this idea!
Tell me when this idea gets some attention.
The more people who like this idea, the more it gets noticed.
The more people who like this idea, the more it gets noticed.
The best point from everyone
-
This whole concept is long overdue. I like very much that the past six months, in particular, have given rise to more metrics beyond just the same cumulative profile data that has been there since day 1, but most people don't live in the range of the ones at the very top of those heaps. Rankings of that nature are very contrary to where the value lies - in the high relevance and entropy one gets by following people who fit how they are using the service at that moment.
My concern with even your approach is that it will continue to repeat a common mistake: to focus too much on accumulation, rather than recency. I Twitter differently now than I did a year ago, largely because my network has grown slowly to fit my needs. It would be great if some/all of these metrics could have separate ratings for long-term and short-term behavior. I wouldn't want this system assuming things about me because of how I Twittered on Election Day, for example.
It will be very interesting to me to see how this might change my own use of Twitter. I mainly follow only local members or members within my academic/professional community. A small percentage extend beyond this group and include people of interest, usually with some notoriety. When Twitter Grader added their recommended members feature (which is secondary to their grading system), I noticed a number of follows from regular people (not spammers, not locals) and attribute that to the Grader recs. Your approach seems like it could be even more effective and thus enticing to expand my network beyond my current rules.
Looking forward to this launching.
I’m delighted
3 people think
this is one of the best points
Create a customer community for your own organization
Plans starting at $19/month
-
Inappropriate?This whole concept is long overdue. I like very much that the past six months, in particular, have given rise to more metrics beyond just the same cumulative profile data that has been there since day 1, but most people don't live in the range of the ones at the very top of those heaps. Rankings of that nature are very contrary to where the value lies - in the high relevance and entropy one gets by following people who fit how they are using the service at that moment.
My concern with even your approach is that it will continue to repeat a common mistake: to focus too much on accumulation, rather than recency. I Twitter differently now than I did a year ago, largely because my network has grown slowly to fit my needs. It would be great if some/all of these metrics could have separate ratings for long-term and short-term behavior. I wouldn't want this system assuming things about me because of how I Twittered on Election Day, for example.
It will be very interesting to me to see how this might change my own use of Twitter. I mainly follow only local members or members within my academic/professional community. A small percentage extend beyond this group and include people of interest, usually with some notoriety. When Twitter Grader added their recommended members feature (which is secondary to their grading system), I noticed a number of follows from regular people (not spammers, not locals) and attribute that to the Grader recs. Your approach seems like it could be even more effective and thus enticing to expand my network beyond my current rules.
Looking forward to this launching.
I’m delighted
3 people think
this is one of the best points
-
Hi Kevin, thanks a lot for your insightful note.
Would you mind elaborating on what your "rules" are, and how they have been changing?
We are definitely looking into some form of "state / rules" as you call it, but we need to understand user needs better.
Thanks for the great articulation of what i think a lot of people are thinking of, but were not able to articulate yet! -
The rules have evolved over time.
When I first started using Twitter, the main purpose was just to communicate with my wife and kids, specifically for a spring break when they headed to Florida and I stayed north to work on academic stuff. That expanded quickly to a small set of 20 or so people, which included both friends and fellow grad students and a couple of celebrities (Steven Wright and Jim Gaffigan, neither of whom make any use of the medium).
As more people in our grad program started using Twitter, I became interested in using it for some research. My focus then became identifying Informatics students, faculty and alums, and following anyone who lived in my region. This got me up to about 100. My geographic rules included nearby towns, because few people were using Twitter at the time.
As it became more successful, I created a second account for my academic tracking and started pruning some of the locals and regional people I followed to increase relevance. I went to a few events where more people wound up following me than I returned, mainly because I wanted to focus just on people in town. I started following just a few people outside of the academic-geographic range if they were of interest, but many of those were not reciprocated (Lee LeFever et al). There was also an increase in Twitter application accounts and a few blogs (Comix Mix, The Onion) that I followed. My network was approaching 300 by this time, far more than I ever anticipated following.
I continue to follow all new Informatics students indiscriminately and have added some locals who have joined Twitter. For a brief time, I followed sets of fictional characters on Twitter (Battlestar Gallactica, Buffy, Laura Ingalls Wilder, etc) but wound up pruning them back because the fiction wasn't as intriguing as real people.
There is another Startup Weekend event coming up next month, this time in Indy. I am certain that the people I meet there will be followed, expanding my network again. I've also got a book due out in the Spring, so I anticipate another spurt of growth around that.
In a nutshell, every time I say there is no way I am going to follow more people, my network grows to a new size. Also true: I only do so when I am ready to grow. So some of the people on my list never would have been followed last year. It also may have been intimidating to follow everyone in Informatics, say, if I wasn't there when that Twitter community was small and manageable. -
It is very impressive how you have such an excellent grasp of how your networks evolve, your needs at each point, and how you are able to articulate it so well. Do you feel that way?
I have a few questions for you (hopefully we will get to use you as a case study one day!!)
1) There are key usage differences between information gathering, relationship building/networking, collaboration, and staying in touch. Do you feel that these use cases are relatively distinct for you? How would you like to handle them?
2) How do you feel about overlaps in your networks, and do you feel conscious that you are broadcasting it to all of them when it might be meant for a single group?
3) Would groups help in your case? There is of course Tweetworks, which is attempting to provide the groups functionality on twitter. -
1) I wouldn't separate "relationship building" from "staying in touch" To me, that is all part of the cycle of disconnection and reconnection. The other two usage cases you mention are specific tasks that benefit from strong relationships. Collaboration on Twitter is not a deep activity. Things are initiated there and reality checks done to poll progress throughout, but the work itself is done elsewhere.
2 & 3) I know there are people who crave the ability to separate subgroups of followers, even going to far as to create and manage multiple Twitter identities (one for local, one for professional). This really needs to be separated into to different aspects: publishing and organizing.
Identity management is not a part of Twitter for me, however. The ability to unfollow (or, with Twalala, even mute people for a while) is always there for people to use. I try to be true to myself, which means sometimes I am noisy and sometimes what I type interests mainly one group. I'm not interested in tweeting different things to different groups. In doing so, I'd think I might impose what I think certain people should know rather than just letting everyone else make up their own minds about any value a tweet might hold.
The second aspect is organization, and for that I am still waiting for some kind of directory management that allows me to tag, flag and search for people in my network in a meaningful way. The best way to search right now is actually to guess someone's Twitter handle (which I rarely see anymore, with use of Full Names in Twitterrific and the ability to click reply buttons). There is nothing that would allow me to track former people I follow, maybe with a note reminding me when/why I stopped. These kinds of things have nothing to do with the outward relationship or publication of content, but it is a way larger and larger networks can make sense. Maybe even something that connects them to other social network pages, like LinkedIn. -
Inappropriate?1, 2 & 4 Likin it, 3 not so much...
-
Hi Des, thanks! Why do you tink (3) is not useful? -
Inappropriate?1 & 2 - maybe. Tough to tell how useful it would be.
3 - I'd be looking to desktop clients to handle that sort of thing. Rarely use the web for posting.
4 - sounds useful -
Hi Adrian, thanks for your note. For (4), what criterion do you tink would be useful - location? influence level? -
Inappropriate?I like the ideas.
It is a little hard to say how things will be received given the changing direction most people have with Twitter.
I would use the suggested items to better understand my community, who is on the edges and how to grow my interests or network. Right now a way to aggregate or segregate is most important. Once I have that I might then want to move to the next level where the needs are unclear.
I find that many of Kevin's comments strike a cord with me. I am doing some similar things without the academic angle. More informal might be the way to put it.
I’m excited
-
Inappropriate?I like the suggested options although for 3, it would only be useful as an additional data point to determine if this is someone I want to follow.
I've been pushing for groups support in twitter for a while but it does not look like it will happen soon. So for the time being, using something like tweetdeck would be useful in breaking the conversation down to manageable bits.
Considering Dunbar's number, I think it's important to make the distinction that for different context, the set of people may overlap or vary greatly so the total number of people you could reasonably interact with could be quite a bit bigger than 150.
It would be interesting then if mrtweet could then recommend different people filtered on different keywords corresponding to those groups of people. For example, my car buddies do not overlap with my interest in data center technology.
I’m happy
-
Inappropriate?while its nice 2 have help on whos not following who, i dont think i need too much help on deciding who 2 follow. i can c their tweets & their favorites. i can also look at other things that help me decide, like how many theyre followin and how many follow them. the last part isnt terribly important really. but it might be to some. theres no accounting 4 taste. some of twitters top guns look like spam royalty 2 me. so i wouldnt mind seeing the stats but when it comes 2 matchmaking... well i dont think i'll have high confidence in the system.
Loading Profile...


EMPLOYEE




