AT&T will have a premium streaming service this fall

How many streaming services will AT&T have?

Will any be successful? So far their track record isn’t too good in the streaming field.


Sent from my iPhone using the SatelliteGuys app!

This would be AT&T's first SVOD service and will be built around content it owns/controls. The other two (DirecTV Now, AT&T Watch) are vMVPD services that require licensing contracts with various linear TV channel providers (aka, competitors). So the WarnerMedia service is more of a cash-in, costing AT&T relatively little beyond app R&D, customer service, and marketing. They can also leverage it to market other services (cellular, internet, etc.) as an add-on perk with relatively little cost to them.
 
This would be AT&T's first SVOD service and will be built around content it owns/controls.
AT&T is building their entire house of cards out of HBO. They'll be other stuff but HBO is becoming a two or three-trick pony just like the rest of the networks and that's not a sustainable anchor for a suite.

To suggest that HBO Kids is a direct competitor to the Disney suite is asinine but that's what you do when you have only one horse.

I'm also compelled to remind what a marginal job of OTT that AT&T has done thus far and they're clearly trying to come from way behind again with precious little detail and a whole lot of chest pounding about HBO being in their back pocket.

The answer won't be found in on demand (unicast). The answer is figuring out how to make multicast work and the MVPD services have been doing it for a loooonggg time.
 
The answer won't be found in on demand (unicast). The answer is figuring out how to make multicast work and the MVPD services have been doing it for a loooonggg time.

Nah, an ever-increasing amount of the video we consume is non-live/on-demand, and therefore unicast. Eventually, just about everything except live sports and live newscasts (and maybe local broadcast channels) will be watched on-demand. Multicast will have a place for the live stuff but not so much outside of that. I expect we're going to see MVPD channel bundles dwindle in the coming years as it becomes the norm to watch the bulk of our TV through SVOD services (unicast) and supplement that with "cable TV" for sports, news and locals. (Although some of us just use an antenna to get the locals for free.) I think this is the main reason why AT&T is far more excited about their upcoming HBO-centered SVOD than they are about DirecTV Now. SVODs are the future and the cable channel bundle is the past.

The next couple years will be really interesting to watch as Netflix, Hulu and Prime Video are joined by Disney+, HBO+ (or whatever the Warner SVOD is called) and Apple TV+ this fall, and the upcoming NBCU and Discovery/HGTV/BBC Nature SVODs in 2020. I think the effect will be to accelerate "cord-cutting," with an increasing number of consumers wanting several of the streaming SVODs but not willing to pay for all that stuff PLUS a cable channel package. Something will have to give and for many of them (mainly those who aren't big sports fans), it'll be the clunky old cable or satellite box and its program guide with lots of channels they never watch.

Ultimately, as more and more viewers ditch the cable bundle, I think you're going to see some of the key cable network owners (who will all also own at least one SVOD) allow MVPDs to distribute skinny bundles consisting of just locals, sports and news channels. (They'll monetize the rest of their content through their SVODs.) Hulu has talked about offering such a skinny live channel bundle and they may be better positioned than anyone to do it, given that Hulu is 2/3 owned by Disney (owner of ESPN and ABC), and 1/3 owned by Comcast (owner of NBCSN, MSNBC and NBC). But if Comcast participates in such a bundle for Hulu (i.e. Disney), except ESPN and ABC (i.e. Disney) to show up in such a bundle to be distributed by Comcast cable. Getting Fox (which is now just the main broadcast net plus their sports and news cable channels) on board shouldn't be hard. RSNs will largely be owned by Sinclair, it looks like, and they only own local network affiliates otherwise, so they should be game to participate.
 
You've got some sure-fire solution to the bandwidth issues created by this approach?
Verizon and Att do...its called a cache server...so that multiple streams of the same movie don't clog bandwidth...the cache server is usually in the closest serving office to the sub

Sent from my SM-G950U using the SatelliteGuys app!
 
Anyone get the impression AT&T is just slinging stuff up on the wall to see what sticks? Counting U-verse they'll now have 5 avenues all competing against themselves for market share. Not a very promising business model.
 
  • Like
Reactions: comfortably_numb
Verizon and Att do...its called a cache server...so that multiple streams of the same movie don't clog bandwidth...the cache server is usually in the closest serving office to the sub
And this will scale to everyone going unicast?

How do they pull this off without some sort of MITM situation?

Verizon and AT&T aren't the way that the bulk of TV viewing currently travels and probably won't be in the future.
 
You've got some sure-fire solution to the bandwidth issues created by this approach?

I'll leave that to the network engineers employed by companies -- ISPs, CDNs, etc. -- that make nearly all of their money now serving up more and more bandwidth to customers who are increasingly demanding more of it. CableLabs, for instance, already has symmetrical 10 Gbps speeds on their roadmap for HFC DOCSIS networks. Comcast appears to be making cloud DVR -- rather than an actual physical DVR in your home -- the default option for new TV subscribers. And of course using a cloud DVR means a unicast stream each time you play a recording. And Comcast isn't the only one moving to such a system. Other cable and telco TV providers are too. They don't seem to be worried about their networks' ability to handle the additional traffic.

Point is that as the market demands more and more of something, service providers will aim to deliver it -- for a decent enough profit, of course. There are folks who today say that the internet will never grow/evolve/advance enough to handle all of our video consumption. But I'm sure those are the same people who a decade ago would never have thought we'd be where we are today, with video accounting for about 3/4 of all internet traffic, and rising. Figures from Nielsen last August seemed to indicate that the average US adult consumes about 20% of their video via streaming vs. regular cable/sat/broadcast sources. But since that doesn't include kids, who skew heavily toward streaming, I'd say that figure undercuts reality. (Plus I'm not so sure about the accuracy of those Nielsen diaries.)

The amount of our video diet served by unicast streams will only increase, particularly as older Americans die and there are more and more young ones raised on YouTube and Netflix.
 
I'm also compelled to remind what a marginal job of OTT that AT&T has done thus far and they're clearly trying to come from way behind again with precious little detail and a whole lot of chest pounding about HBO being in their back pocket.

The answer won't be found in on demand (unicast). The answer is figuring out how to make multicast work and the MVPD services have been doing it for a loooonggg time.
Unicast isn't entirely terrible if you can get reasonable statistical multiplexing. This is the hidden secret to Netflix' success -- sure they have 80 million subs, but at any given instant they only have a couple million active streams.

Where ATT went with DIRECTV NOW and WatchTV is the worst case scenario from an infrastructure standpoint. Not only did they build it on appointment TV, but started without any DVR functionality so people had to watch at exactly the same time if they wanted to see a certain show.

I still think the grand equalizer in all of this is going to be cost. As streaming services scale up the costs will exceed the distribution costs of cable/sat, and then we'll get a more honest look at what value people place on convenience.
 
And this will scale to everyone going unicast?

How do they pull this off without some sort of MITM situation?

Verizon and AT&T aren't the way that the bulk of TV viewing currently travels and probably won't be in the future.
verizon and att carry the bulk of internet and gasp cellular traffic..this may come a surprise but all cell towers connect to a landline network..the cast majority controlled by att and verizon..I won't get into cache servers because its beyond the scope of this forum,,but all it is a locally cached copy of a popular movie or show..its not a new idea..its kinda like cacheing a webpage in a browser..you can't have on demand and unicast..that would be like a broadcast tv
 
There is much more to the Internet backbone than what AT&T and Verizon provide. The problem probably isn't with the backbone but out in the far reaches where stuff travels through cable and fiber that isn't so readily scaled.

Broadband providers will have to upgrade their plant and/or make the nodes much smaller. There's some light coloring and stuff that can be done to increase bandwidth incrementally, but we're talking perhaps orders of magnitude increase in demand.
 
There is much more to the Internet backbone than what AT&T and Verizon provide. The problem probably isn't with the backbone but out in the far reaches where stuff travels through cable and fiber that isn't so readily scaled.

Broadband providers will have to upgrade their plant and/or make the nodes much smaller. There's some light coloring and stuff that can be done to increase bandwidth incrementally, but we're talking perhaps orders of magnitude increase in demand.
You don't understand the structure of the internet..you have a good grasp on local loop but you have no concept of the Internet as a giant computer network

Sent from my SM-G950U using the SatelliteGuys app!
 
  • Like
Reactions: comfortably_numb
You don't understand the structure of the internet..you have a good grasp on local loop but you have no concept of the Internet as a giant computer network
That's an odd statement, but I challenge you to explain how bulking up at one level can help if all of the other levels don't expand in lockstep. Pretty much everyone's broadband connection is carried by multiple companies and each has to do their own scaling to serve the new requirements. The last mile is every bit as important as every other mile. We see this in cable broadband and we see it in wireless. You can't readily build expressways into a star network; you have to create whole new constellations closer to the ends.

Unlike water and electricity, there will be few opportunities for conservation of bandwidth.
 
70CA9CE9-002B-4DD1-97F6-04ABE4292AB2.png
 
  • Like
Reactions: MikeD-C05
That's an odd statement, but I challenge you to explain how bulking up at one level can help if all of the other levels don't expand in lockstep. Pretty much everyone's broadband connection is carried by multiple companies and each has to do their own scaling to serve the new requirements. The last mile is every bit as important as every other mile. We see this in cable broadband and we see it in wireless. You can't readily build expressways into a star network; you have to create whole new constellations closer to the ends.

Unlike water and electricity, there will be few opportunities for conservation of bandwidth.
Explaining the internet is beyond the scope of this forum...look up cache servers and start there

Sent from my SM-G950U using the SatelliteGuys app!
 
Explaining the internet is beyond the scope of this forum...look up cache servers and start there
It's real easy to fire off a hit-and-run post like this and use buzzwords that most have heard but you can't just say that it works like it always has without making some connections.

How do caching servers cache bits of video streams?

Web caching caches files that come with URIs but how do you cache and index video segments for later recall? It is easy to presume that time code would have to be part of the equation but there's still the question of how you identify what you're caching. If the client is having problems, you would presumably also need to cache multiple qualities of the stream.

If the choke point is upstream from the caching server you can't cache what you can't get.