My complaint yesterday about Google making their new social networking site invite-only got some flak elsewhere about being too dismissive of the usefulness of the invite-only beta-testing phase of Google+. It does make sense for a software product of any kind to be thoroughly tested for bugs before being released to the general public, and I could very well have been quite wrong about an initial closed-off phase being bad for social networking sites. The obvious counter-example is the most popular social networking site now in existence, namely Facebook. It started out being closed off to everyone except college/university students.
But that comparison points out the fundamental difference in the approach between Facebook and Google. Facebook limited its initial clientele according to social characteristics of the potential clientele: whether they were in school or not. Google's invites are being targeted, initially, at web developers, not because of any social characteristics of the web developer population but because of technical considerations: so they can debug the technical mechanics of the platform. This may mean that an initial, inter-connected and identifiable social segment of the larger population will indeed populate Google+ and thereby make it useful, so that Google+ becomes the social network of choice for tech-heads, but I still think it will be the success or failure of that initial population seeding, not any considerations of technical polish, that will set the trajectory for whether Google+ succeeds or fails.
As for the importance of debugging: it's probably not a pleasant thing for a tech developer to hear that the quality of their technical work is not actually all that important in people's decisions about what technology to use, but the entire history of technology for the masses seems to suggest that crappy-but-accessible eventually beats out polished-but-inaccessible most times: PC beat out Mac, DOS beat out OS/2, Microsoft Windows beat out everything else. This is just as true for social networking sites as for anything else. After all, nobody used Myspace because it was so well designed and so unlikely to error out. True, the much more cleanly laid out Facebook eventually superseded it, but Facebook's rise coincided with the rolling out of Facebook apps, which were still quite an attractive draw for the masses even when they (a) messed up the formerly clean layout of Facebook profiles, and (b)would initially error out as often as not. Unless there are parts of the Google+ site that currently flat out fail to function at all, then I think the potential bugginess of the platform isn't nearly as big an issue for the masses as the average tech developer might think it is.
So I could well be wrong about why I think being invite-only is a bad idea for Google+ as it currently stands, but as yet I'm not fully convinced.
Showing posts with label social tech. Show all posts
Showing posts with label social tech. Show all posts
Monday, July 04, 2011
Sunday, July 03, 2011
Software Engineering vs the Sociotechnical: a Google Plus Post
I have not been able to get my hands on an invite to Google Plus, Google's latest attempt to create a viable Facebook competitor. My only perspective to date has been from the video reviews that are now popping up on Youtube. This one, for example:.
From what I can see, Google has basically copied almost of all of Facebook's functionality, page layout and all, then added some of its own distinguishing features. These features are "circles", "hangouts" and "sparks". Looking at what they're trying to do, I'm uncertain if Google is fully understanding what they are trying to build: not just a technological platform for social communication, but a social world in itself.
The obvious first problem is making the thing invite-only. Part of Facebook's current appeal is that everybody is on there. In fact, part of the appeal of any social networking site is that other people are on there. According to Metcalfe's Law, the growth of value on a communication network grows exponentially in relation to the growth in the number of people who are on the network. Sure, Gogle Plus is in beta, and from an engineering perspective, it makes sense to only let potential debuggers on there first. But the engineering perspective is the wrong perspective to have.
In my area of study, the terminology for an approach that regards things in terms of both their social and technological factors is "sociotechnical". And I think in this case, the sociotechnical approach says: don't create an online social platform that depends on having as many people on it as possible in order for it to work most effectively, and then tell the vast majority of people that potentially want in on it that they're not invited.
The three highly-marketed new features of Google Plus (circles, hangouts and sparks) can be interpreted as new sociotechnical processes for communication. Before going into analysis of those new processes, I need to do a brief foray into some communication theory and how it relates to Facebook. Communication can be categorised in a number of ways. The divides I'm going to use are (1) "discursive" versus "presentational" forms of communication, and (2) "phatic" versus "informational" forms of communication.
Discursive communication generally refers to written and spoken language, usually thought of as privileging logic and rationality. Presentational communication generally refers to visual and aural pictures and sounds, usually thought of as privileging emotion and feeling. Presentational communication can generally be absorbed much faster and more fully than discursive communication (you can literally take it all in with a glance), but discursive communication tends to be more complete and more detailed.
Phatic communication refers to communication that cements social bonds: the actual content of the communication is irrelevant to the purpose of publicly creating and reinforcing the social bonds between individuals. danah boyd has called this kind of communication "social grooming". Informational communication refers, obviously, to communication that has informational content. These two categories are not mutually exclusive. However, phatic communication with little informational content is very efficient compared to any kind of communication with a great deal of informational content.
In a world where less and less time is available to communicate with more and more people, the speed and efficiency of presentational and phatic communication are preferable to the slower forms of discursive and informational communication. This can be seen in the evolution of Internet technologies, from Usenet groups to blogging to social networking to microblogging: each one enables more and more communication to be done with less and less discursive and informational content. Social networking in general, and Facebook in particular, maintain their continuing appeal due to their privileging of presentational and phatic communication. For the average user, Facebook status updates, comments, and especially "likes", are a quick way to use phatic communication to efficiently maintain social bonds with their hundreds to thousands of online "friends". But probably more important is the time and effort that's been put into making the Facebook website as oriented to presentational communication as possible: everything is highly visually-oriented rather than text-oriented (with certain obvious exceptions; more on that later). Indeed, the "killer app" of Facebook is most likely the photo gallery facility. It is here that friends can share visual representations of times and events spent together, ostensibly with a great deal of privacy as well.
Which brings us to Google Plus and its new features. With their first feature, "circles", Google has taken an obvious swipe at Facebook's...shall we say, "un-nuanced" approach to privacy. In Facebook, the ability to create and maintain different groups with different sharing settings is present, but not particularly pleasant: you create the group and, one by one, type in the names of friends you want to add to it. Assigning the correct permissions to the correct groups is also a chore, I've been told. Google Plus' approach has made the process much more visually oriented: group assignment is drag-and-drop over a visual map of all available circles, not just manual assignment to one group at a time. This process will make group maintenance much easier.
Whether the circles feature will make discrete group sharing more useful is another qeustion altogether. Part of the problem with all attempts to do this sort of thing online is that the attempt to mimic contextual sharing as it exists offline (where you're willing to share some things with some people, but not others) is by its very nature very difficult to replicate online: offline, the separation of context exists by default; online, the separation is something that must be actively created and constantly maintained. This constant requirement to expend effort to keep the contexts separate online is the true obstacle to effective implementation of conextual privacy online, not just an easier ability to create and maintain the existence of different sharing groups. Does Google Plus make this effort easier somehow? I don't see how it does.
Moving on, "hangouts" isn't a particularly revolutionary breakthrough. It's Facebook chat, but video chat instead of text chat. Facebook in fact is going this way too, with Skype soon to supply such chat on Facebook. What is different, but doesn't yet seem to be implemented, is apparently the ability to have multiple people chatting over video simultaneously.
Facebook chat is probably the most glaring exception to the preference for presentational over discursive communication on the Facebook platform. Chat is text-based. It conceals a lot of the emotive content that would occur over a video-type chat. Some people actually prefer this, as it allows them to exert much greater control over their online persona and online communications: there is no requirement to be always and immediately emotionally available to the people that they're chattting with, and there is the opportunity to take time to think about and consciously craft their communication with others. Video chat would eliminate these perceived advantages. Of course, for people who find text chat distasteful precisely because it lacks the emotive content of face to face communication, they would most likely prefer the video option (assuming they have the technology to make use of it: many people still don't have cameras connected to computers). From an engineering perspective, Google has the nous and framework to easily implement video chat thanks to their work on Google Talk. From a sociotechnical perpsective, it's not a bad thing either, but only if text chat is still available alongside video chat. Is it? I'm honestly not sure.
Finally, "sparks" are a new idea to apply to social networking sites. The idea seems to be to try and use interests and the like to try and stimulate conversation among friends who want to talk about that interest (or event, or person, or object). It actually seems to be an idea from Google's business department: try to leverage Google Plus' "plus one" feature (their answer to the Facebook "like" button) to improve the search results in their core product by recording how many people "plus one" the various interests (and potential search terms) that they are exposed to through sparks. The sociotechnical question in this case is whether anyone is going to use it. The default assumption of sparks even being a good idea is that all communication is informational: it's about something. This is wrong. In fact, to date "informational" communication has actually been deprivileged on social networking sites in favour of communication that is purely phatic. As someone who has the time and inclination to talk in detail online (just look at the size of this blogpost, and I'm not even done yet :P), I would probably like this. For many people, particularly the target market of people who use social networking sites (i.e Facebook users), this feature doesn't really offer much to the user, as most of them only really use Facebook to "keep in touch" rather than to have deep and meaningful exchanges about stuff. Unfortunately, the perceived business advantage to Google, through exploitation of the "plus one" feature for search optimisation, may blind them to the reality of how communication on social networking sites actually works (i.e. it is phatic far more than it is informational).
So, Google Plus offers some new things. But does it offer enough to encourage people to switch from Facebook, where all their friends and contacts are still set up, and have been for some time? Well, the obvious answer is that people can't switch anyway while the whole thing is still invite-only. Once Google Plus goes public, the answer may change. My tentative answer right now is that it will appeal to some people, namely those who have friends there already (which means friends of web-developers and other people currently entitled to Google Plus invites will move across), those who have the desire and available technology to use video chat (which means people in the upper half of the Western world's income bracket), and those who have both the time and inclination to make talk online. This means Google Plus would be the preferred location for people who are comfortably well-off, employed in white-collar jobs that are not too time-consuming, and who are comfortable with communicating online about things that they're interested in, in detail.
Hmm. If Google Plus takes off, and that becomes its expected demographic, what then becomes the expected demographic of Facebook?
From what I can see, Google has basically copied almost of all of Facebook's functionality, page layout and all, then added some of its own distinguishing features. These features are "circles", "hangouts" and "sparks". Looking at what they're trying to do, I'm uncertain if Google is fully understanding what they are trying to build: not just a technological platform for social communication, but a social world in itself.
The obvious first problem is making the thing invite-only. Part of Facebook's current appeal is that everybody is on there. In fact, part of the appeal of any social networking site is that other people are on there. According to Metcalfe's Law, the growth of value on a communication network grows exponentially in relation to the growth in the number of people who are on the network. Sure, Gogle Plus is in beta, and from an engineering perspective, it makes sense to only let potential debuggers on there first. But the engineering perspective is the wrong perspective to have.
In my area of study, the terminology for an approach that regards things in terms of both their social and technological factors is "sociotechnical". And I think in this case, the sociotechnical approach says: don't create an online social platform that depends on having as many people on it as possible in order for it to work most effectively, and then tell the vast majority of people that potentially want in on it that they're not invited.
The three highly-marketed new features of Google Plus (circles, hangouts and sparks) can be interpreted as new sociotechnical processes for communication. Before going into analysis of those new processes, I need to do a brief foray into some communication theory and how it relates to Facebook. Communication can be categorised in a number of ways. The divides I'm going to use are (1) "discursive" versus "presentational" forms of communication, and (2) "phatic" versus "informational" forms of communication.
Discursive communication generally refers to written and spoken language, usually thought of as privileging logic and rationality. Presentational communication generally refers to visual and aural pictures and sounds, usually thought of as privileging emotion and feeling. Presentational communication can generally be absorbed much faster and more fully than discursive communication (you can literally take it all in with a glance), but discursive communication tends to be more complete and more detailed.
Phatic communication refers to communication that cements social bonds: the actual content of the communication is irrelevant to the purpose of publicly creating and reinforcing the social bonds between individuals. danah boyd has called this kind of communication "social grooming". Informational communication refers, obviously, to communication that has informational content. These two categories are not mutually exclusive. However, phatic communication with little informational content is very efficient compared to any kind of communication with a great deal of informational content.
In a world where less and less time is available to communicate with more and more people, the speed and efficiency of presentational and phatic communication are preferable to the slower forms of discursive and informational communication. This can be seen in the evolution of Internet technologies, from Usenet groups to blogging to social networking to microblogging: each one enables more and more communication to be done with less and less discursive and informational content. Social networking in general, and Facebook in particular, maintain their continuing appeal due to their privileging of presentational and phatic communication. For the average user, Facebook status updates, comments, and especially "likes", are a quick way to use phatic communication to efficiently maintain social bonds with their hundreds to thousands of online "friends". But probably more important is the time and effort that's been put into making the Facebook website as oriented to presentational communication as possible: everything is highly visually-oriented rather than text-oriented (with certain obvious exceptions; more on that later). Indeed, the "killer app" of Facebook is most likely the photo gallery facility. It is here that friends can share visual representations of times and events spent together, ostensibly with a great deal of privacy as well.
Which brings us to Google Plus and its new features. With their first feature, "circles", Google has taken an obvious swipe at Facebook's...shall we say, "un-nuanced" approach to privacy. In Facebook, the ability to create and maintain different groups with different sharing settings is present, but not particularly pleasant: you create the group and, one by one, type in the names of friends you want to add to it. Assigning the correct permissions to the correct groups is also a chore, I've been told. Google Plus' approach has made the process much more visually oriented: group assignment is drag-and-drop over a visual map of all available circles, not just manual assignment to one group at a time. This process will make group maintenance much easier.
Whether the circles feature will make discrete group sharing more useful is another qeustion altogether. Part of the problem with all attempts to do this sort of thing online is that the attempt to mimic contextual sharing as it exists offline (where you're willing to share some things with some people, but not others) is by its very nature very difficult to replicate online: offline, the separation of context exists by default; online, the separation is something that must be actively created and constantly maintained. This constant requirement to expend effort to keep the contexts separate online is the true obstacle to effective implementation of conextual privacy online, not just an easier ability to create and maintain the existence of different sharing groups. Does Google Plus make this effort easier somehow? I don't see how it does.
Moving on, "hangouts" isn't a particularly revolutionary breakthrough. It's Facebook chat, but video chat instead of text chat. Facebook in fact is going this way too, with Skype soon to supply such chat on Facebook. What is different, but doesn't yet seem to be implemented, is apparently the ability to have multiple people chatting over video simultaneously.
Facebook chat is probably the most glaring exception to the preference for presentational over discursive communication on the Facebook platform. Chat is text-based. It conceals a lot of the emotive content that would occur over a video-type chat. Some people actually prefer this, as it allows them to exert much greater control over their online persona and online communications: there is no requirement to be always and immediately emotionally available to the people that they're chattting with, and there is the opportunity to take time to think about and consciously craft their communication with others. Video chat would eliminate these perceived advantages. Of course, for people who find text chat distasteful precisely because it lacks the emotive content of face to face communication, they would most likely prefer the video option (assuming they have the technology to make use of it: many people still don't have cameras connected to computers). From an engineering perspective, Google has the nous and framework to easily implement video chat thanks to their work on Google Talk. From a sociotechnical perpsective, it's not a bad thing either, but only if text chat is still available alongside video chat. Is it? I'm honestly not sure.
Finally, "sparks" are a new idea to apply to social networking sites. The idea seems to be to try and use interests and the like to try and stimulate conversation among friends who want to talk about that interest (or event, or person, or object). It actually seems to be an idea from Google's business department: try to leverage Google Plus' "plus one" feature (their answer to the Facebook "like" button) to improve the search results in their core product by recording how many people "plus one" the various interests (and potential search terms) that they are exposed to through sparks. The sociotechnical question in this case is whether anyone is going to use it. The default assumption of sparks even being a good idea is that all communication is informational: it's about something. This is wrong. In fact, to date "informational" communication has actually been deprivileged on social networking sites in favour of communication that is purely phatic. As someone who has the time and inclination to talk in detail online (just look at the size of this blogpost, and I'm not even done yet :P), I would probably like this. For many people, particularly the target market of people who use social networking sites (i.e Facebook users), this feature doesn't really offer much to the user, as most of them only really use Facebook to "keep in touch" rather than to have deep and meaningful exchanges about stuff. Unfortunately, the perceived business advantage to Google, through exploitation of the "plus one" feature for search optimisation, may blind them to the reality of how communication on social networking sites actually works (i.e. it is phatic far more than it is informational).
So, Google Plus offers some new things. But does it offer enough to encourage people to switch from Facebook, where all their friends and contacts are still set up, and have been for some time? Well, the obvious answer is that people can't switch anyway while the whole thing is still invite-only. Once Google Plus goes public, the answer may change. My tentative answer right now is that it will appeal to some people, namely those who have friends there already (which means friends of web-developers and other people currently entitled to Google Plus invites will move across), those who have the desire and available technology to use video chat (which means people in the upper half of the Western world's income bracket), and those who have both the time and inclination to make talk online. This means Google Plus would be the preferred location for people who are comfortably well-off, employed in white-collar jobs that are not too time-consuming, and who are comfortable with communicating online about things that they're interested in, in detail.
Hmm. If Google Plus takes off, and that becomes its expected demographic, what then becomes the expected demographic of Facebook?
Saturday, May 07, 2011
A bad prediction from the dot-com bust
I'm reading some historical research on the growth of new media. In the wake of the dot-com bust in 2000, critics of the more grandiose claims made about the Internet got a lot more prominent. A financial writer for the New Yorker, one John Cassidy, had this to say not long after the crash:
"[the Internet} was not a disruptive technology that would destroy any company locked into the old way of doing things, such as selling books in stores, printing news on paper, or using people to sell stocks. The bookstores, newspaper companies, and brokerage houses are still in business, and most of them are doing fine"
Given what's happened to all three of those industries in recent times, I find this amusing.
"[the Internet} was not a disruptive technology that would destroy any company locked into the old way of doing things, such as selling books in stores, printing news on paper, or using people to sell stocks. The bookstores, newspaper companies, and brokerage houses are still in business, and most of them are doing fine"
Given what's happened to all three of those industries in recent times, I find this amusing.
Monday, February 04, 2008
The promise and problems of OpenID
Playing around with an OpenID provider and something bothers me.
The idea behind OpenID of course is to try and get past the problem of having lots of different accounts on lots of different websites. The existence of a site like Useless Account is testimony to the problem.
OpenID, as I understand it, gets you to sign on to a single OpenID provider site, like My Open ID, which you can then use to sign into other sites rather than having to explicitly create a new user/password combination for each and every site you want to use.
The main obstacle at the moment seems to be a lack of major sites that will authenticate using OpenID. There does appear to be some recent momentum, with web2.0 site aggregator Plaxo and the Blogger.com comment system now being accessible through OpenID authentication. But that's not what's bothering me.
What's bothering me is that I already have a lot of accounts on a lot of sites which aren't tied to my OpenID account, and I don't see any way to tie those accounts to my OpenID account. Worse, plenty of existing sites like LiveJournal also act as OpenID providers: you can sign onto an OpenID-compatible site using your LiveJournal details. As a result I now not only have multiple accounts around the place, but two of those multiple accounts are both OpenID accounts. This doesn't bode well for a system aimed at reducing the number of superfluous user/password combinations I have to keep in mind.
Maybe there's something I missed in the protocol, but to my knowledge there's no way to associate my previously existing accounts with my OpenID accounts. Nor do I know any way to make my existing accounts on disparate OpenID providers aware of each other so that I could easily alternate between them, or even subordinate one to the other.
Perhaps someone could point out whether this is currently possible? If it isn't, then I fear OpenID will go down the road of things like the DVORAK keyboard: a technological improvement that is superior, useful and fails because it desn't take entrenched social realities into account. To succeed, OpenID needs to be able to assimilate existing accounts somehow.
The idea behind OpenID of course is to try and get past the problem of having lots of different accounts on lots of different websites. The existence of a site like Useless Account is testimony to the problem.
OpenID, as I understand it, gets you to sign on to a single OpenID provider site, like My Open ID, which you can then use to sign into other sites rather than having to explicitly create a new user/password combination for each and every site you want to use.
The main obstacle at the moment seems to be a lack of major sites that will authenticate using OpenID. There does appear to be some recent momentum, with web2.0 site aggregator Plaxo and the Blogger.com comment system now being accessible through OpenID authentication. But that's not what's bothering me.
What's bothering me is that I already have a lot of accounts on a lot of sites which aren't tied to my OpenID account, and I don't see any way to tie those accounts to my OpenID account. Worse, plenty of existing sites like LiveJournal also act as OpenID providers: you can sign onto an OpenID-compatible site using your LiveJournal details. As a result I now not only have multiple accounts around the place, but two of those multiple accounts are both OpenID accounts. This doesn't bode well for a system aimed at reducing the number of superfluous user/password combinations I have to keep in mind.
Maybe there's something I missed in the protocol, but to my knowledge there's no way to associate my previously existing accounts with my OpenID accounts. Nor do I know any way to make my existing accounts on disparate OpenID providers aware of each other so that I could easily alternate between them, or even subordinate one to the other.
Perhaps someone could point out whether this is currently possible? If it isn't, then I fear OpenID will go down the road of things like the DVORAK keyboard: a technological improvement that is superior, useful and fails because it desn't take entrenched social realities into account. To succeed, OpenID needs to be able to assimilate existing accounts somehow.
Subscribe to:
Posts (Atom)