[Make-wifi-fast] On building your own routers and the mass market

[Dave Taht]

Changing the topic. I'd like the arstechnica discussion to remain political....

On Fri, Mar 11, 2016 at 11:31 AM, Paul Vixie <paul at redbarn.org> wrote:
> Jonathan Morton wrote:
>>> On 11 Mar, 2016, at 20:22, Luis E. Garcia<luis at bitamins.net>  wrote:
>>> Time to start building our own.
>> A big project in itself - but perhaps a worthwhile one. We wouldn’t
>> be able to compete on price against the Taiwanese horde, but price is
>> not the only market force on the table. Firmware quality is a bit
>> abstract and nebulous to sell to ordinary consumers, but there is one
>> thing that might just get their attention.

I  totally approve of people getting together and building things they
think the market needs, or merely what they think they need.
Ironically, in 1999, the designs I was shopping around for a "home
wireless AP" were designed to plug into where the thermostat was in a
house, had a touchscreen (and ran x-windows remotely), and sank
without a trace past a VC's glazed eyes over and over again.  Then the
wrt happened. in 2002... and 15 years later, Nest happened... sigh.

The segment of "makers", the essential R&D, ultimately feeds into more
polished products, and people should try making those too. There are
billions of devices left to be built in the next decade, and plenty of
room for innovation, and the best, not necessarily the cheapest, like
tp-link - will win - and it's my hope that certain values we have here
will propagate into the successful products. We're not particularly
good at final products, however.

... I had a nice visit over at eero yesterday. Their CTO greeted me
with a flash stick outstretched, containing their (debian-based) code.
;) I was very impressed that they went from kickstarter to a US-wide
launch of their product(s), with enough people, and with a business
model that makes sense (selling three 2-radio diversity meshed routers
at a time to get better coverage (and improve margins)), that they
seem to understand the need for remote updates deeply, AND the need to
make things simple for ordinary users. They also seem to grok things
we are really bad at, like getting finished products on shelves.

... and I'm going to order a couple, 'cause their wifi is not as good
as it could be (nobody's is), and he said I could visit periodically
with make-wifi-fast's upcoming fixes. https://eero.com/

> the cznic team has already done this.
>
> https://omnia.turris.cz/en/

I am rooting for them, also, hard! They've made all the right choices
technically and it's (aside from the wifi, currently) the best chipset
possible from an open source perspective. They - unlike nearly all the
home router makers and CPE vendors, have a perspective driven by the
problems they've seen by being a registrar. They have money in the
bank, and low costs, and an idealism about open design that seems
rooted in the maker movement. I'd love to see this market
arduinotized.

It would be my hope, that with another year's worth of software
development that their design would be suitable for more mass market
CPE, particularly to ISPs in the fiber space. (And that they will be
essentially first to have better wifi as I plan to get stuff working
on that board as soon as I get one in my hot little hands).

A huge percentage of the former "home router" market is going away as
more and more ISPs bundle the router/wifi and other options into a
single device. That concerns me a lot, and the only thing I can think
of worth doing there is to join this org: http://rdkcentral.com/ and
encourage others to also do so.

Still, there has to be a lower cost option (for which I'm mostly
chasing mediatek, and sad to have had to have dropped both netgear and
tp-link from further consideration).

- but who knows - with serious volumes, on a standardized platform
that gets 100ks of deployments by ISPs, manufactured for the long
term, the omnia's costs could drop significantly enough to be
competitive with cheap crap, or, as in the raspberry pi market, a
whole new segment using more standardized firmware will emerge.

I'd really like to see more devices interoperate in the IoT, as well,
things like making mdns scale better, ipv6 work with naming more
right, etc - perhaps this new wave of folk entering the market  - and
others exiting - might actually be able to pull more of the pieces
together.

>
> --
> P Vixie
> _______________________________________________
> bufferbloat-fcc-discuss mailing list
> bufferbloat-fcc-discuss at lists.redbarn.org
> http://lists.redbarn.org/mailman/listinfo/bufferbloat-fcc-discuss