[Bloat] [Cerowrt-devel] fcc initial comments due sept 10

[Livingood, Jason]

Hi David - See some comments inline below. Hope you are well!
- Jason

On 8/10/18, 11:35 PM, "Bloat on behalf of dpreed at deepplum.com" <bloat-bounces at lists.bufferbloat.net on behalf of dpreed at deepplum.com> wrote:

    Of course, the folks who wanted to see Internet connectivity spread and restructure the economic structure of communications decided not to look this "gift horse" in the mouth. Congress was going to fund rollouts of Cable TV (new and upgrades), and as a side effect, maybe some better Internet.

[JL] How and when did Congress fund DOCSIS network investments (either in 2007-2008 or before)? Do you have any citations for this? 
    
The architecture of the Cable TV service distributes *every* channel simultaneously to every endpoint, consuming outrageous bandwidth compared to the 25 Mb/s diddly squat usage on the "broad band" cable or fiber.

[JL] This is changing more rapidly than you may realize, through a combination of factors. One is that most cable companies have been deploying IP-based set tops for several years, which enables more unicast delivery using IP, which frees up channels for DOCSIS Internet services (otherwise where would cable companies find the channel space to 24 or 32 downstream channels plus D3.1 OFDM compared to 4 or 8 downstreams a few years ago?). In addition, on the demand side, customers increasingly watch programs via their DVR or On Demand library rather than watching something live (typical exceptions being sports, news, weather, etc.). Finally, they are also watching increasingly on non-set-top devices like tablets and laptops.

    So, in responding to this notice by the FCC of a "standard" for Broadband, just realize that you might want to question the entire proposition that Broadband is a thing that needs a standard.

[JL] I thought it was just the usual annual NOI on the definition of "advanced telecommunications capabilities". The use of the word "broadband" just appears to be shorthand for that. See https://docs.fcc.gov/public/attachments/FCC-18-119A1.pdf which says:

"Section 706 of the Telecommunications Act of 1996, as amended (1996 Act), requires us to determine and report annually on “whether advanced telecommunications capability is being deployed to all Americans in a reasonable and timely fashion.”FN1
 
FN1 47 U.S.C. § 1302(b). For simplicity in past inquiries, the Commission has sometimes used the term “broadband” to refer to “advanced telecommunications capability.” However, “advanced telecommunications capability” is a statutory term with a definition that is narrower than the term “broadband.” See 47 U.S.C. § 1302(d)(1) (“The term ‘advanced telecommunications capability’ is defined, without regard to any transmission media or technology, as high-speed, switched, broadband telecommunications capability that enables users to originate and receive high quality voice, data, graphics, and video telecommunications using any technology.”). As this definition makes clear, while all services providing advanced telecommunications capability are “broadband,” not all broadband services provide advanced telecommunications capability."
    
    What "high speed" means in the Internet is a much more meaningful question, but that's a truth in advertising question. And it really has to do with "response time" more than any guarantee of an "up to" speed. How long does it take Netflix to buffer enough content to play the rest of the show without interruption? (that's why you need burst rate) How fast do your trigger pulls on your multiperson VR shooter get reflected on all the other players' displays?
    
    People do pay more for that. But the standard of what "that" is - high speed, very high speed, ... isn't just 25 Mb/sec. It's also bufferbloat, for example.
    
    THink about that, and don't get sucked into comments on what Cable TV should do for minimal Internet quality.
    
[JL] Putting aside all the stuff about cable and broadband, it seems your bottom line is that buffer bloat, latency, and latency under load are also meaningful and should be considered in addition to speed/throughput measures in this definition. That seems a worthwhile technical question to raise.