Comments
mpouraryan
9/25/2016 8:31:18 PM
User Rank Platinum
Re: Netflix Learned to Read Your Mind
I don't view as being "courageous" but smart business--and part of the progress that can't be stopped despite all that our inherent fears as I noted in response to another comment on 4K TV to John--we're already here--For those using Google Inbox, For instance, it is already smart enough to anticipate your answer--so it is not just "The movies"...is it?
Fascinating times..right?
elizabethv
9/26/2016 1:03:37 AM
User Rank Platinum
Re: Netflix Learned to Read Your Mind
@mpouraryan - that's an interesting comparrison. Yet with Google "reading your mind" on what question you're going to ask, while I do agree they do tend to be right, there have been times where I've found them completely off-base too. And while I might be kind of accepting of Netflix "reading my mind" to some extent when they are right, if they get it wrong a few too many times, I'll either figure out a way to bypass the feature completely, or, like you suggested, leave Netflix as a whole. As much as that would hurt.
JohnBarnes
9/26/2016 8:04:23 AM
User Rank Platinum
Re: Interesting
afwriter,
"Yet" is a very small distinction.
Also for good or ill, to "control" anything as self-wiled, flexible, and complex as a human, Neflix's machine intelligence has to be remarkably responsive; it's like working with a bird dog (the dog finds the birds but you're the one that decides to hunt and whether to shoot) or riding a horse (things generally work better if you and the horse achieve an agreement about what should be happening).
JohnBarnes
9/26/2016 8:06:51 AM
User Rank Platinum
Re: Netflix Learned to Read Your Mind
ElizabethV,
Reading other people is even hard for other people, and we all guess wrong all the time. Over the weekend I saw a marvelous panel cartoon: couple sitting on a park bench.
Guy: "Well, I kind of feel like it's time to get -- "
Girl: "YES! I will marry you!"
Guy: ".... coffee."
JohnBarnes
9/26/2016 8:21:39 AM
User Rank Platinum
With a foot on three stools ...
Credentials: 1) I've written a lot of fiction, it was my main source of income for a long time. 2)I'm retraining in data science and have had some fairly heavy-duty coursework in machine learning. 3) Way back when I was a theatre professor and taught a lot of general humanities, and my academic area was theatre history, aka what was popular a long time ago.
So here's where Netflix is setting themselves up to get rolled someday: most of the time, people want predictability. The great majority of stories sold in all media are nothing new. Furthermore, most of the time, things creep away from diversity. For example, in 1959, Westerns and prime time game shows accounted for a majority of entertainment programming. Over time content becomes more and more routinized; the later members of a genre have a huge number of mandatory tropes.
However, there's a logical dynamic: the more routinized the popular genres get and the more rigid the routines become, and the more those genres take up audience space, the more you build up people who want to enjoy something unlike anything they've ever seen before, and who are prepared to enjoy just about anything different and oddball enough. (You could call that hipsterization.)
Add to that the tendency for people to use entertainment consumption as a way to differentiate themselves from other people. (Not a new phenomenon; look up the riots that erupted around Hernani and Ruy Blas or The Firebird).
I can see how machine learning could produce low-level novelty -- "hey, there's only one Western in the lineup but it's drawing its few fans like crazy" -- and might be able to find something like, say, Star Trek or other "sleeper" hits, eventually. But I think all that does is build up a deeper demand for greater novelty. Sooner or later you get an explosive birth of a whole new genre, and if Netflix can't find it, it will clobber them (until there's enough data for them to routinize it, and start the cycle over!)
mpouraryan
9/26/2016 1:18:48 PM
User Rank Platinum
Re: Netflix Learned to Read Your Mind
The key is to never forget in this transformative era we are witness to--that we have to be in control--and make sure we retain the creative edge always. It is in the end about us--and no machine can replace that.
JohnBarnes
9/26/2016 8:21:48 PM
User Rank Platinum
Re: Interesting
ElizabethV,
I've already seen machine-intelligence models that can include the pleasures of shopping and delayed choice -- that is, a consumer may simply prefer/like better to make a partial choice and then linger over making a final one for the fun of lingering and considering and sampling. (That's what food festivals are all about, for example!) They can and do predict what consumers will settle on but what they want to look at first (you see that in some Amazon predictions about "customers looked at this and ultimately bought that").
They're already on to you, and two jumps ahead besides.
Resistance is futile.
JohnBarnes
9/26/2016 8:25:32 PM
User Rank Platinum
Re: Netflix Learned to Read Your Mind
freehe,
Au contraire, people who like mixtures tend to like predictable mixtures, often with a lag effect (if they've watched a lot of vampire musicals lately, they are more ready for superhero romantic comedies, unless their baseline includes a lot of anime noir, in which case they'll want alt history with zombies ...), and that's well-studied and well-understood. If you like to watch a wide variety of things, simple linear regression and hierarchical clustering models can do a very good job of predicting how much your mixture will vary and what you will do to maintain it.
Resistance is futile.
JohnBarnes
9/26/2016 8:46:52 PM
User Rank Platinum
The contradiction at the heart of the issue
A friendly reader who says he's too shy to get into the argument, but who knows me, just emailed me and pointed out that some of my posts seem to say "Oh, no, don't think you can escape, machine intelligence and big data are much more powerful than that," whereas others seem to be saying that popular entertainment will always escape the algorithms. Here's why I'm saying that, Mr. Anonymous-Boy McChickenPants, and darn it, come out and argue your own points -- that's why we have a comment section!
Anyway, there's a confusion that goes back to Enlightenment philosophy in the West where we tend to think that "having freedom" means "being free to choose." So we think that to have our freedom about Grey's Anatomy has to somehow mean we choose to watch or not watch based on our own priorities. We do choose, of course, based on those priorities -- but those priorities are not only not altogether ours, they are knowable and manipulable. But that doesn't really mean resistance is futile (despite what I keep saying); it just means the shopper-and-shelf model isn't how things really work (or ever have).
Acting on the choice feels like freedom. But you don't wholly control your choice. But -- and here's the kicker -- neither does anyone else. Your choice just happens, based on your personal history, temperament, who you hang out with, whether someone you like said something mean to you this morning, your digestion .... in the aggregate those influences become stochastic and they can predict pretty safely that by pitching it in a particular way to 135,000 people like you, they'll get 100,000-110,000 viewers. (That's much better than 30 years ago when they had to pitch to two million and didn't know if it would be 5000 or 500,000). But whether you will ultimately land in the conformist majority or the dissident minority -- that's probablistic, not deterministic.
There's another sense of freedom, after all -- the sense of "free fall", aka weightlessness, the thing that means that although there certainly is gravity in space (that's what keeps your space station in orbit!) that as long as you and everything around you moves in accord with gravity, you don't feel it. And that's the freedom that big data and machine learning can't take away; all they can do is help you to stay in the orbit.
"Resistance is futile" -- the Borg
"Resistance is part of the game" -- Carse
"Resistance is voltage divided by amperage" -- Ohm
Michelle
9/27/2016 12:56:27 PM
User Rank Platinum
Queue queue
Netflix does a fair job of suggesting new titles for my household. I don't like the new queue set up. They removed the "continue watching" queue and replaced it with "watch next". The "watch next" queue is just another suggestion list like all the others below it. Suggestions are based on past viewing and ratings just like everything else. I want to resume some shows of my own choosing, not be fed constant suggestions.
|
|
Italy's 5G auction could exceed a government target of raising €2.5 billion ($2.9 billion) after attracting interest from companies outside the mobile market.
The emerging-markets operator is focusing on the humdrum business of connectivity and keeping quiet about some of its ill-fated 'digitalization' efforts.
Three UK has picked Huawei over existing radio access network suppliers Nokia and Samsung to build its 5G network.
Vendor says that it's its biggest 5G deal to date.
Verizon skates where the puck is going by waiting for standards-based 5G devices to launch its mobile service in 2019.
Orange has been one of the leading proponents of SDN and NFV. In this Telco Transformation radio show, Orange's John Isch provides some perspective on his company's NFV/SDN journey.
10/16/2017
Huawei Network Transformation Seminar
The adoption of virtualization technology and cloud architectures by telecom network operators is now well underway but there is still a long way to go before the transition to an era of Network Functions Cloudification (NFC) is complete.
The Small Cell Forum's CEO Sue Monahan says that small cells will be crucial for indoor 5G coverage, but challenges around business models, siting ...
People, strategy, a strong technology roadmap and new business processes are the key underpinnings of Telstra's digital transformation, COO Robyn ...
Eric Bozich, vice president of products and marketing at CenturyLink, talks about the challenges and opportunities of integrating Level 3 into ...
Epsilon's Mark Daley, director of digital strategy and business development, talks about digital transformation from a wholesale service provider ...
Bill Walker, CenturyLink's director of network architecture, shares his insights on why training isn't enough for IT employees and traditional ...
All Videos
|
On-the-Air Thursdays Digital Audio
Special Huawei Video
Tweet This