From the sample of people I follow, it really looks like the engagement is down from the early days: fewer posts, fewer comments and worst, fewer new accounts created.
I'm fully aware it is a completely non-scientific "study", but you would be hard-pressed to find anyone not working at Google saying "G+ is really thriving right now!"
I totally agree. I really don't see people adopting G+. Besides, the UI is extremely slow to load compared to Facebook (you can observe this on slow connection when G+ page just loads and loads, and it is not even sharp and clean to look at. Of course I am contrasting with Facebook, what else one can do? (hence the irony).
Photographers seem to be enjoying it more than other "social platforms". Alex Wild (of myrmecos, photographer and entomologist) regularly notes that G+ is significantly better than "the rest" for photographers.
I'm not one so I wouldn't know, but he seems to genuinely prefer G+, and he is clearly more active there than on e.g. facebook.
This is even more true if you think about multiple languages. I often publicly write in Italian about Italian politics, and in English about technology, and I don't want to "spam" my English followers with content they can't read (and much less care about), but I also want my posts to be public for the Italian users. This is the most important use case that G+ isn't addressing right now, imho.
I've been expecting someone to make this move sooner or later, and I got the impression that Posterous' Spaces was trying to do something like this (but I haven't used it much).
It reminds me of adding tags to blog posts and then having subdomains for different tags
Most people don't create a lot of content, and won't bother tagging their short tweet/status
This is a good point, but couldn't that also be said for the existing way Circles work? I think some form of the reverse would be better for power users without affecting the casual users.
There might be something interesting that they could do with defining a hashtag whitelist or a set of filter criteria for a circled contact.
Someone mentioned languages, too - what if autodetected language could be one of the filter criteria? I have a Russian friend who sends out a 50/50 mix of English and Russian. I would rather not see the Russian in my stream, partially because Google Translate usually doesn't do a great job, and also because I miss out on all the inside cultural references.
Yup - this is one of the things I like that you can do on Livejournal - I can specifically filter in or out certain tags that people use on their posts, so if I want to ignore someone's cat poetry then I can.
I'm not sure of the name of the feature, but I recall that G+ launched with "spark topics"(?) that I had hopes for. It seems to have devolved into saved G+ search terms now. I was hoping it would evolve into something where the platform identified common topic content of public posts together and helped strangers with those interests form a public topic circles.
Those terms can be searched, and there's additional refinement which can be added to them, in part. This seems to be under flux, and I've been spending less time in G+ over the past month or two.
That said: for content discovery, sparks are far superior to circles.
Honestly, most people that I run into don't really understand how Google+ works. Basically it is the same poor, unintuitive UI that we've come to expect from Google.
For example, how does a new user really learn what a circle is? And how does someone recognize what each input control is when sharing a post (at least one input box has no hints at all). I've personally seen people completely baffled by their first experiences with the product.
And it seems each "update" just adds more tedium and complexity. Volume for circles is a necessary evil, but it should have been done in a much less intrusive and time-consuming way.
Google seriously needs to rethink their entire UI strategy.
Eg, why should I have to set a volume for each circle? Couldn't they figure out a better way for me to provide hints as
Or the alternate solution is to show all of my posts (even those that were not marked Public) in my profile page, so that when people visit my profile, they can see what I write about and decide to circle me. There could be a mechanism to mark posts as private which should not even be shown in the profile page.
Indeed, it's backwards and it's a big problem. I find myself sharing less because I constantly have the thought of my audience in the back of my mind. I don't want to spam people.
There are several orthogonal directions in which people group. First, there are people who I have a relationship with for largely historical reasons, and who I've hung out with at some point for one reason or another. This includes my mom, old time high school friends, a friend from tennis class, etc. These people rarely ever share the same interests, and yet this grouping alone is emphasized in most social networks that currently exist.
The other significant, orthogonal and just as important portion of my connections are people who I share an interest with, irrespective of whether or not I know them personally. These are colleagues studying the same subject at a different university, hackers, bloggers, tech enthusiasts etc.
Somewhat ironically, Facebook clearly goes after the former grouping, and yet if you look through your Facebook stream you'll realize that most of it is garbage exactly because people don't have all that much to say about themselves and end up sharing mostly about their interests. It's just a mess. I was looking up to Google+ when it came out to fix this because they were coming in with the advantage of seeing the Facebook model unfold, but circles isn't it.
In short, I wish social networks emphasized and catered to the power of an interest as much as they cater to the power of a personal connection.
I agree that G+ suffers for numerous fundamental architectural failings.
I wrote about this at length in some of my early G+ engagements, apparently in comments as I can't find any specific posts on the topic. Also in some of the G+-related support forums. Oddly: Google makes searching this stuff rather difficult. And/or killed the group entirely.
Basic criticisms:
- G+ circles are asymmetric. What I call my circles (and how I classify them), and whether or not you're in them, is independent of how your classifications and/or inclusions. Sometimes you're not looking for a circle (centered on you), but a society (centered on an idea, place, thing, ...).
- In emission, circles are akin to tags in that I can indicate my focus for a conversation. But I can't tell you what that tag is. In receipt, circles indicate my interest areas, but again, I can't communicate that with those whom I'm receiving content from. This is particularly annoying in following someone who's notable in a particular area, or in several areas, but posts on a wide range of topics. It's been interesting to discover that Linus has a skin-diving hobby ... but I'm not particularly interested in those posts.
- Confounding "Public" with circles is an error.
- Confounding sharing with scope of rebroadcast is an error.
- Past a certain size, "streams" become too fucking big. For a company whose first key innovation was identifying relevant online content, Google+ presents a shit-ton of irrelevant content.
My suggested fixes:
- Create true semantic tags, and allow them to be collaboratively assigned. Yes, tags. Author can tag stuff, readers can tag stuff (optionally enabled/disabled) and/or vote on tags.
- Split scope of distribution from topic of conversation. I may want to have widely inclusive or narrowly scoped discussions of politics, religion, rape as a mode of social control in the Byzantine empire, WoW, or deadlift technique. Grant me this. A distribution might be strictly limited, limited-but-expandable, or wholly public.
- Sane foreign language management. If you're going to translate, then translate. If not, then unless I specifically say so, don't bother me with content not in my native language.
- Allow better stream filters. I'm probably going to want to weight stuff persons of interest present, but have an awareness of more widely discussed topics (breaking news, emerging trends). Allow multiple conversations to be aggregated (G+ divides, rather than unites). Don't strictly time-order stuff.
- Give me a quiet space to generate posts. Edit-box-over-streams doesn't work well.
- +1 implies -1. Make it so.
- Better crap management: trolls, spams, etc. Fer-fucin' reals, Googlez!
- As far as the social management aspect goes: less tedium, better tools. Without bugging the crap out of me, manage stuff. Clues as to when you're about to make a really bad disclosure would be good.
- Full anonymity / pseudonymity support. End of story.
I'm not sold on G+ by a long shot. It's been an interesting experiment. I'm not sold on "social networking" by a long shot either. I'm more interested in having useful networks to share thoughts/ideas/information. No, I really don't trust anyone's network/systems to my personal information, and I very much share the view expressed by some (RMS particularly articulately) that SN is a huge, voluntary, surveillance network.
Yes, this problem became obvious very quickly with G+. As an alternative to post categories, tags or hashtags would go a long way to solving the problem.
Tag your content, and let your subscribers filter by tag. To create the filter, tag your circles to give them context, and when someone in a circle makes a post with a matching tag, show the post, otherwise suppress it.
So for example, I tag my ‘Functional Programming’ circle with #Haskell #Ocaml #Scala #MapReduce #Lisp #Clojure #Python, and only posts with at least one of those tags are shown in my feed. Posts in that circle tagged with #VacationPictures #DinnerPictures and the like are suppressed.
Do tagging the way StackOverflow does – when you start typing a tag for a post, show all the related ones, let you click to select pre-existing tags. That should keep the number of slightly-differently-spelled tags to a minimum.
This and similar schemes have been repeatedly suggested since G+ went live. They added hashtags, hashtag search, and hashtag autocomplete, but not the killer app – hashtag filter – yet. The tagging infrastructure is all there, I don't know what the holdup is. Even if this is just a 90% solution, it's still infinitely better than the current state of G+.
Makes you wonder, do they use their own product in the same way their audience does? Are they not feeling this pain point too?
25 comments
[ 0.24 ms ] story [ 62.9 ms ] threadAnd yes, I still believe it would really help salvage G+ from the situation it is in right now...
I'm fully aware it is a completely non-scientific "study", but you would be hard-pressed to find anyone not working at Google saying "G+ is really thriving right now!"
I'm not one so I wouldn't know, but he seems to genuinely prefer G+, and he is clearly more active there than on e.g. facebook.
It reminds me of adding tags to blog posts and then having subdomains for different tags
This feature is only useful for a very minor audience. Most people don't create a lot of content, and won't bother tagging their short tweet/status.
Power users who do write a lot should already have blogs like the one the author is using.
This is a good point, but couldn't that also be said for the existing way Circles work? I think some form of the reverse would be better for power users without affecting the casual users.
Someone mentioned languages, too - what if autodetected language could be one of the filter criteria? I have a Russian friend who sends out a 50/50 mix of English and Russian. I would rather not see the Russian in my stream, partially because Google Translate usually doesn't do a great job, and also because I miss out on all the inside cultural references.
That said: for content discovery, sparks are far superior to circles.
For example, how does a new user really learn what a circle is? And how does someone recognize what each input control is when sharing a post (at least one input box has no hints at all). I've personally seen people completely baffled by their first experiences with the product.
And it seems each "update" just adds more tedium and complexity. Volume for circles is a necessary evil, but it should have been done in a much less intrusive and time-consuming way.
Google seriously needs to rethink their entire UI strategy.
Eg, why should I have to set a volume for each circle? Couldn't they figure out a better way for me to provide hints as
Indeed, it's backwards and it's a big problem. I find myself sharing less because I constantly have the thought of my audience in the back of my mind. I don't want to spam people.
There are several orthogonal directions in which people group. First, there are people who I have a relationship with for largely historical reasons, and who I've hung out with at some point for one reason or another. This includes my mom, old time high school friends, a friend from tennis class, etc. These people rarely ever share the same interests, and yet this grouping alone is emphasized in most social networks that currently exist.
The other significant, orthogonal and just as important portion of my connections are people who I share an interest with, irrespective of whether or not I know them personally. These are colleagues studying the same subject at a different university, hackers, bloggers, tech enthusiasts etc.
Somewhat ironically, Facebook clearly goes after the former grouping, and yet if you look through your Facebook stream you'll realize that most of it is garbage exactly because people don't have all that much to say about themselves and end up sharing mostly about their interests. It's just a mess. I was looking up to Google+ when it came out to fix this because they were coming in with the advantage of seeing the Facebook model unfold, but circles isn't it.
In short, I wish social networks emphasized and catered to the power of an interest as much as they cater to the power of a personal connection.
I wrote about this at length in some of my early G+ engagements, apparently in comments as I can't find any specific posts on the topic. Also in some of the G+-related support forums. Oddly: Google makes searching this stuff rather difficult. And/or killed the group entirely.
Basic criticisms:
- G+ circles are asymmetric. What I call my circles (and how I classify them), and whether or not you're in them, is independent of how your classifications and/or inclusions. Sometimes you're not looking for a circle (centered on you), but a society (centered on an idea, place, thing, ...).
- In emission, circles are akin to tags in that I can indicate my focus for a conversation. But I can't tell you what that tag is. In receipt, circles indicate my interest areas, but again, I can't communicate that with those whom I'm receiving content from. This is particularly annoying in following someone who's notable in a particular area, or in several areas, but posts on a wide range of topics. It's been interesting to discover that Linus has a skin-diving hobby ... but I'm not particularly interested in those posts.
- Confounding "Public" with circles is an error.
- Confounding sharing with scope of rebroadcast is an error.
- Past a certain size, "streams" become too fucking big. For a company whose first key innovation was identifying relevant online content, Google+ presents a shit-ton of irrelevant content.
My suggested fixes:
- Create true semantic tags, and allow them to be collaboratively assigned. Yes, tags. Author can tag stuff, readers can tag stuff (optionally enabled/disabled) and/or vote on tags.
- Split scope of distribution from topic of conversation. I may want to have widely inclusive or narrowly scoped discussions of politics, religion, rape as a mode of social control in the Byzantine empire, WoW, or deadlift technique. Grant me this. A distribution might be strictly limited, limited-but-expandable, or wholly public.
- Sane foreign language management. If you're going to translate, then translate. If not, then unless I specifically say so, don't bother me with content not in my native language.
- Allow better stream filters. I'm probably going to want to weight stuff persons of interest present, but have an awareness of more widely discussed topics (breaking news, emerging trends). Allow multiple conversations to be aggregated (G+ divides, rather than unites). Don't strictly time-order stuff.
- Give me a quiet space to generate posts. Edit-box-over-streams doesn't work well.
- +1 implies -1. Make it so.
- Better crap management: trolls, spams, etc. Fer-fucin' reals, Googlez!
- As far as the social management aspect goes: less tedium, better tools. Without bugging the crap out of me, manage stuff. Clues as to when you're about to make a really bad disclosure would be good.
- Full anonymity / pseudonymity support. End of story.
I'm not sold on G+ by a long shot. It's been an interesting experiment. I'm not sold on "social networking" by a long shot either. I'm more interested in having useful networks to share thoughts/ideas/information. No, I really don't trust anyone's network/systems to my personal information, and I very much share the view expressed by some (RMS particularly articulately) that SN is a huge, voluntary, surveillance network.
"be careful not to conflate gossip with real news"
first web hits: http://www.merriam-webster.com/dictionary/conflate http://www.thefreedictionary.com/confound
I think the poster is fine.
Conflating is a combining, which may be positive or negative.
Confounding is a combining with the result that the parts combined cannot be distinguished, and such that the result confuses the combined parts.
Tag your content, and let your subscribers filter by tag. To create the filter, tag your circles to give them context, and when someone in a circle makes a post with a matching tag, show the post, otherwise suppress it.
So for example, I tag my ‘Functional Programming’ circle with #Haskell #Ocaml #Scala #MapReduce #Lisp #Clojure #Python, and only posts with at least one of those tags are shown in my feed. Posts in that circle tagged with #VacationPictures #DinnerPictures and the like are suppressed.
Do tagging the way StackOverflow does – when you start typing a tag for a post, show all the related ones, let you click to select pre-existing tags. That should keep the number of slightly-differently-spelled tags to a minimum.
This and similar schemes have been repeatedly suggested since G+ went live. They added hashtags, hashtag search, and hashtag autocomplete, but not the killer app – hashtag filter – yet. The tagging infrastructure is all there, I don't know what the holdup is. Even if this is just a 90% solution, it's still infinitely better than the current state of G+.
Makes you wonder, do they use their own product in the same way their audience does? Are they not feeling this pain point too?