[Bloat] [Cake] dslreports is no longer free

Sergey Fedorov sfedorov at netflix.com
Fri May 1 19:35:34 EDT 2020


Hi Michael,

This blog post <https://netflixtechblog.com/building-fast-com-4857fe0f8adb>
describes how
the test steers to the server(s).
Noted on the other thread, I hope to add the url param option reasonably
soon.

SERGEY FEDOROV

Director of Engineering

sfedorov at netflix.com

121 Albright Way | Los Gatos, CA 95032



On Fri, May 1, 2020 at 3:07 PM Michael Richardson <mcr at sandelman.ca> wrote:

>
> {Do I need all the lists?}
>
> Sergey Fedorov via Bloat <bloat at lists.bufferbloat.net> wrote:
>     > Just a note that I have a plan to separate the loaded latency into
>     > upload/download. It's not great UX now they way it's implemented.
>     > The timeline view is a bit more nuanced, in the spirit of the
> simplistic
>     > UX, but I've been thinking on a good way to show that for super
> users as
>     > well.
>     > Two latency numbers - that's more user friendly, we want the general
> user
>     > to understand the meaning. And latency under load is much easier than
>     > bufferbloat.
>
>     > As a side note, if our backend is decent, I'm curious what are the
> backends
>     > for the speed tests that exist that are great :)
>
> Does it find/use my nearest Netflix cache?
>
> As others asked, it would be great if we could put the settings into a URL,
> and having the "latency under upload" is probably the most important number
> that people trying to videoconference need to know.
>
> (it's also the thing that they can mostly directly/cheaply fix)
>
>     > SERGEY FEDOROV
>     > Director of Engineering
>     > sfedorov at netflix.com
>     > 121 Albright Way | Los Gatos, CA 95032
>
> Very happy that you are looped in here.
>
> --
> ]               Never tell me the odds!                 | ipv6 mesh
> networks [
> ]   Michael Richardson, Sandelman Software Works        |    IoT
> architect   [
> ]     mcr at sandelman.ca  http://www.sandelman.ca/        |   ruby on
> rails    [
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.bufferbloat.net/pipermail/bloat/attachments/20200501/dd42805f/attachment.html>


More information about the Bloat mailing list