People are social creatures and the modern Internet reflects that. Technology has made collaboration at a distance possible in new ways that present their own set of challenges. This course will introduce you to the major challenges and opportunities for creating online communities. What does the future hold? Learn how social computing can create collaboration experiences that go beyond what’s possible face to face.
What Makes Social Computing Hard?
Loading...
En provenance du cours de University of California San Diego
Social Computing
667 notes
Explore Related Videos
At Coursera, you will find the best lectures in the world. Here are some of our personalized recommendations for you
À partir de la leçon
Connecting People through Technology
When you think of social media, you might think of a popular social network, like Facebook, Twitter, or LinkedIn. These are indeed powerful, but the universe of social media is much larger. Both online and off, most everything we do has a social component that is mediated by some technology. After all, technologies have been helping people communicate since long before your favorite curly-haired CEO dropped out of college. In this module, you’ll learn about different types of social software, how to think about physical collaboration and digital collaboration using the same framework, and what the challenges to success for collaborative software are.
Rencontrer les enseignants
Scott KlemmerProfessor
Cognitive Science & Computer Science
Over the last ten years we've seen an enormous expansion of
the influence of social media.
For example, I'm about to head to the Chi 2015 Conference.
Where the musician Psy is gonna the closing keynote.
And the idea of a world famous pop-star giving a keynote talk at
a computer science conference would seem unthinkable just a few short years ago.
And what's interesting about it is in science biography
he refers to himself as a social media innovator.
Also something that it seems like,
just a few short years ago, that wasn't the sort of thing that rock stars did.
And I bet many of you know Psy but one reason that he's relevant for
our story today is that part of his fame outside of Korea.
He's been extremely well known inside of South Korea for a long time,
but part of his global fame came from his video Gangnam Style.
And this video has become so popular that it even broke the YouTube counter.
There are more than 2 billion people who viewed this video.
So many, that it broke the integer size that YouTube originally set
up to store how many views a video got in their database.
Here's a common social question.
Are you free on Thursday at 3?
And I think the answer is, well it depends on who's asking.
A few years ago, Bill Gates dropped by campus and
I had the opportunity to meet him.
And when Bill Gates calls and says hey, are you free, the answer is absolutely.
On the other hand, I think many of us have had the experience of
there's a low priority meeting where you're not that heavily involved anyway or
something that you may not have wanted to go to anyhow.
And in that case, the answer to am I free on Thursday at 3 is maybe no.
And the reason that I bring this up for
our discussion today is that as we'll see when we start to look at shared calendars.
Social Life is Contextual and the question of if it depends on who asking.
It's not a tiny trifle to be dealt with at the end.
It's actually very much at the core of our social reality.
And so, because social interaction is context specific.
You know, the way that you present yourself and handle yourself and
the conversations you have at a bar or on the beach with friends,
that's really different than the way that you might interact at the office or
a family gathering.
And that's a good thing.
It's an expression of the sophistication of human culture.
And the challenge is that technology, the stuff that's easiest to build,
often ignores that, the fact that our social lives are so
multidimensional and context-specific.
That can be a pain from the vantage point of software development.
For example, on Facebook, you know, everybody is your friend.
But there's a big difference between my sister or
a cousin or a best friend or a spouse.
Or a co-worker that you know really well, or
somebody that you met once at a conference seven years ago and barely remember.
Or somebody that you may not have wanted to respond to a friend query to but
you didn't want to be rude and say no.
So all of this things, unless we're careful can get collapsed, and
that collapsing can cause problems in the way that social media works.
And so what we're gonna look at today are what some of these problems are and
what we can do as a designer to be able to address them.
A lot of this shows up because we've moved from the individual user era
into the social computing era.
In the individual user world you've got your single computer.
And we've got our laptop here and life is good.
And there's a single person and they are at their desk doing some work.
All right.
And here the challenges of interaction design are largely about
supporting an individual's user work flow.
Is the interface clear and self-disclosing.
Is it expressive in the ways that it needs to be.
On the other hand when you think about a social computing system,
a lot of the things that your trying to support with technology,
you still have all of those, can I figure out what the user interface is doing?
Is it clear?
Do I know what's going on?
But you have all of these new challenges too.
Of supporting all of the interactions or not between users.
One of the people who describes this best is the San Diego Cognitive Science alum,
Jonathan Grudin.
After he graduated, he started working on social computing, and
the challenges of doing that motivated him to
write this really great article about the difficulties of group work.
And one of his favorite examples is that of a calendar.
If you think about a shared calendar from the vantage point of somebody who's
a manager, well these things are great because a manager's got a lot of meetings.
And, so you got some time block off here, and some time block off here,
and some time block off here and on Friday you go surfing.
It's really nice to be able to see other peoples' calendars as well.
So we might see some more calendar information
from other people that overlays and
if I want to be able to make a meeting we can find a block of free time that works.
But what about from a workers perspective?
Well here there are a couple of challenges.
The first one is that a lot of the information about what you need to do
might be just in your head and not in the system itself.
Another one is that the time that's not scheduled on your calendar,
that's when you have actual work.
That's when work gets done.
So free time isn't free, that's the real work time.
Also, different people might use different infrastructures.
So one person might use a Microsoft system, somebody
else is using Google Calendar, somebody else is using a paper day planner.
Each of which has it's own merits and reasons why you use it and all that stuff.
But they may or may not interact.
And the fourth one, as we talked about at the beginning of this video,
is that the availability depends on who's asking.
So here we see a central challenge of making cooperative software work.
That there can be a disparity between the person who does the work and
the person who gets the benefit.
And you see that in the case of calendars,
because managers who have to coordinate all these different schedules, they get
a big benefit by being able to have a consistent view of everybody's work.
On the other hand, the individual users, if they keep a shared calendar for
the purposes of coordination, well for them, the cost exceeds the benefit.
And so we have for workers the cost is higher and
for managers well, the benefit might be higher.
But the shared calendar only works if everybody uses it.
There's a network effect.
And so to get the benefit of being able to line all these up,
you need to have everybody use it.
Now, there's lots of ways that you can address the challenges, and
I think different systems have done that in different ways.
One way of addressing the challenge is, is what we see with Doodle.
So Doodle is a web application
that enables you to schedule an individual meeting.
And so if we want to be able to find a couple of free hours to meet this week,
maybe for a dissertation defense it's important to be able to find free time.
All of the people involved may use different calendaring systems or
have preferences about when they prefer to meet or not, and again,
our availability depends on who's asking.
But by putting this in something like Doodle you're asking an answer not
in general are you free, but are you free at this particular time?
Here's another example with Google Calendar where as the Google
infrastructure integrates more and more services,
the cost of being able to keep a shared calendar gets smaller.
The user interface gets easier, the integration gets smoother, and
the benefits start to get larger.
And so even for an individual worker the relationship between those two can flip.
One way that you can see this is by integrating e-mail and calendar.
And so one way of both lowering the friction of using a piece of software,
decreasing the cost and adding value is that we can, by showing my calendar,
in the context of an email that comes in asking about a calendar relevant query.
It makes it a lot easier to answer that.
And so I'm incented to participate and keep things in there.
And then I can just hit a single button to be able to add that to the calendar.
Another challenge of building social software
is being able to handle exceptions.
One thing that is extraordinary about human social interaction
is that people are extremely good at extemporizing and exception handling.
So much so that we often don't realize how much of it we're doing.
We've seen this especially in recent years with regard to medical records.
You don't realize how much extemporizing is going on until you
try to replace it with a system that doesn't allow that kind of extemporizing.
So Christian Heath and Paul Luff and
many other researchers have looked at how doctors actually use medical records.
In part because when electronic health records started to appear,
there was a surprisingly large amount of push-back and frustration for
those that were using it.
One of the things that you see with paper medical records is that they encode
a lot more than just what officially goes into the form fields.
So, for example, when a paper medical record is on somebody's desk you know
that they are likely to be the person who has the most recent information.
If a paper medical record is worn and dog-eared.
Well that means that there's been a lot of activity, whereas one that's brand new and
stiff means it's seen less interest.
When something is provisional it might get written in, in pencil, and
then only later transferred to pen.
You can even tell, sometimes, who it was that entered that information.
And be able to infer, based on that, some additional information.
Or know who to ask if you have a question about that.
Now, all of these benefits of paper health records
doesn't mean that there aren't benefits to the digital world.
So if we look at a digital interface, we see that it's extremely easy to search.
It's extremely easy to aggregate across patients.
When somebody moves, shipping digital files is a lot easier.
They can also be in multiple places at the same time.
There's lots of benefits of the digital world.
Computers generally write better than doctors, so
there's a handwriting benefit there as well.
And it's easier to automatically integrate sensors, enrich multimedia, and
all that stuff.
Electronic health records clearly have huge benefits.
And so, when we look about the benefits of paper records, I think the question is
not so much are we gonna keep paper around forever, for every single task.
But rather how can we marry the benefits of the paper world
with those that we see in the digital world?
And I think that health records have become particularly challenging,
because if you think about the relationship between a doctor and
a patient in the paper world,
well you've got a doctor and they've got the paper in front of them.
And you've got the patient just right here, and they can see each other.
But when you change over to the computer world then you can have a situation
where the doctor now is typing into a computer.
And then the patient is over here, and
they're wondering why on earth nobody's paying attention to them.
And this loss in bedside rapport is something that can have
real health consequences.
Where, if I tell you something looking in the eyes,
you may be much more willing to listen to it.
And if I'm, on the other hand,
turned around, that's not gonna work nearly as well.
You may be even less disclosive if we're not having eye contact.
And so, I think here you can see there's lots of ways that you can remedy this and
people are working on it.
I think the best electronic health systems address many of these issues.
What I think is notable,
is that with all of the examples that we've seen in group work here.
The extemporizing, the coordination, the change in the locus of power, the change
in who gets the benefit, what things get easier and what things get harder.
If you don't pay attention to all of these subtle variables, the ship of your social
computing system can mysteriously sink without you knowing why.
Here's another wonderful example.
For many years, possibly still to this day,
airline traffic controllers have used paper flight strips.
This is research from Wendy McCaya and colleagues.
What I love is you can see these yellow paper flight strips.
They encode all of the relevant detail about a particular flight.
And what you see here in this picture is that there's an airline traffic controller
who is responsible for the flight strips that are directly in front of them.
To anybody in that room can know who's responsible for what.
Furthermore, if I want to be able to hand a flight to you, I do that physically,
and the passing of the physical token makes it extremely clear who's in charge.
There's little room for ambiguity.
Often digital systems lose that really strong token passing.
And that open loop communication where you've lost the confirmation and
the common ground awareness that I know that you know that I know.
That's when systems have problems.
And again, it's not like we couldn't replace this with a digital system.
But that many of the subtle coordination benefits that paper and
physical system support.
We want to think about how we might make those more possible and
also get the benefits of the digital era.
And something that's almost taught a logical, but I think,
nonetheless profound is that if a technology is gonna provide an advantage,
well at some point the correspondence with the real world has to break down.
The reason that we're looking to a digital health record is because it
can do all of these things that a paper health record might not be able to do.
If we jump too far, too fast,
without checking what happens we can run into real trouble.
So one good heuristic for being able to design effective social computing
systems is to be really attentive to current practice.
Think about what people do now.
And then all things equal, work to minimize that distance.
The benefit there is not only might you keep some of the interactional benefits of
the paper world, but also people have to relearn less.
And so onboarding of new users can be a lot easier.
And so in conclusion, what we see here is that if we want to build social
computing apps, that a world of extemporizers actually adopts.
In general, the more flexible things are the better.
For example, Twitter, which I think of as an extremely successful social computing
platform in terms of its enormous adoption.
This short messaging platform is extremely free-form.
There's not a lot of structure.
And the structure that has emerged, things like hashtags,
many of them first emerged as user community conventions.
That over time became more and more strongly adopted into the platform.
The flexibility enables that end user innovation.
And I think Twitter is a great example of how these flexible,
light-weight tools are much more used than the more highly prescribed ones.
And that paper is often one of our best examples of something that is lightweight,
that's flexible, and that supports extemporizing.
And so now the design challenge for
all of you, as we build the digital world, is how we can enable one that has
all of this wonderful micro-coordination that we're used to in the analog one.
Explorer notre catalogue
Rejoignez-nous gratuitement et obtenez des recommendations, des mises à jour et des offres personnalisées.
Coursera propose un accès universel à la meilleure formation au monde,
en partenariat avec des universités et des organisations du plus haut niveau, pour proposer des cours en ligne.
© 2019 Coursera Inc. Tous droits réservés.
