[Cerowrt-devel] eero gains competition in plumewifi

[Cerowrt-devel] eero gains competition in plumewifi

[Dave Taht]

I sure wish I knew how they are implementing diversity routing and if
they are bothering to pay attention to make-wifi-fast

https://www.plumewifi.com/

-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Cerowrt-devel] eero gains competition in plumewifi

[Erkki Lintunen]

On 06/20/2016 09:16 AM, Dave Taht wrote:
> I sure wish I knew how they are implementing diversity routing and if
> they are bothering to pay attention to make-wifi-fast
>
> https://www.plumewifi.com/


Quite a staffing for a startup, the web site lists 46 names and 
positions from which 26 are named as engineers. Wondering if they 
already are in the business of WiFi chip and RF engineering, which 
enable them to do some "magic" to make a network of their own from the 
plumewifi plugs. On staffing resources this isn't on par with Eero, I think.

- Erkki

Re: [Cerowrt-devel] eero gains competition in plumewifi

[Erkki Lintunen]

(resent, wrong sender address at first send)

On 06/20/2016 09:16 AM, Dave Taht wrote:
> I sure wish I knew how they are implementing diversity routing and if
> they are bothering to pay attention to make-wifi-fast
>
> https://www.plumewifi.com/


Quite a staffing for a startup, the web site lists 46 names and 
positions from which 26 are named as engineers. Wondering if they 
already are in the business of WiFi chip and RF engineering, which 
enable them to do some "magic" to make a network of their own from the 
plumewifi plugs. On staffing resources this isn't on par with Eero, I think.

- Erkki

Re: [Cerowrt-devel] eero gains competition in plumewifi

[Maciej Soltysiak]

[-- Attachment #1: Type: text/plain, Size: 2023 bytes --]

On Mon, Jun 20, 2016 at 8:16 AM, Dave Taht <dave.taht@gmail.com> wrote:

> I sure wish I knew how they are implementing diversity routing and if
> they are bothering to pay attention to make-wifi-fast
>
> https://www.plumewifi.com/


"Plume is a cloud coordinated WiFi system that exponentially increases the
signal strength and quality of your WiFi. It monitors your WiFi activity
and balances your network load without sacrificing the performance of one
device over another. Plume utilizes multiple WiFi channels in the same home
to communicate between Plume Pods, eliminating congestion as your WiFi
demands change and increase. Our cloud algorithms figure out which band to
use for each device to ensure your devices have access to the speed they
deserve, avoiding interference along the way. Plume WiFi adapts to the
physical space you live in, your online devices, your household, and even
to your neighbor’s WiFi usage patterns (yes, their networks also impact
your environment!). It adapts real-time to your personal needs."

I'm reading this as: we setup the APs on all 3 (1,6,11) channels, have the
plumes record which channels have best radio and network parameters at a
given time; store that on their disks over the web (sorry, cloud); then
dynamically set power and perhaps other radio chip and QoS parameters to
something a la heuristics: e.g. send X type of traffic on ch11 after 6pm
Mon-Fri.

Doesn't sound like they're making wifi fast in Dave Taht sense. More like
working around the issues by multiplying channels and squeezing what they
can from the radio. That may work, who knows.

Anyway, this is what a consumer like me reads it.

Best regards,
Maciej




>
>
> --
> Dave Täht
> Let's go make home routers and wifi faster! With better software!
> http://blog.cerowrt.org
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel
>

[-- Attachment #2: Type: text/html, Size: 2993 bytes --]

Re: [Cerowrt-devel] eero gains competition in plumewifi

[Dave Taht]

On Tue, Jun 21, 2016 at 1:43 AM, Maciej Soltysiak <maciej@soltysiak.com> wrote:
> On Mon, Jun 20, 2016 at 8:16 AM, Dave Taht <dave.taht@gmail.com> wrote:
>>
>> I sure wish I knew how they are implementing diversity routing and if
>> they are bothering to pay attention to make-wifi-fast
>>
>> https://www.plumewifi.com/
>
>
> "Plume is a cloud coordinated WiFi system that exponentially increases the
> signal strength and quality of your WiFi. It monitors your WiFi activity and
> balances your network load without sacrificing the performance of one device
> over another. Plume utilizes multiple WiFi channels in the same home to
> communicate between Plume Pods, eliminating congestion as your WiFi demands
> change and increase.

I love how marketing folk can eliminate congestion with a wave of a
word processor.

> Our cloud algorithms figure out which band to use for
> each device to ensure your devices have access to the speed they deserve,
> avoiding interference along the way. Plume WiFi adapts to the physical space
> you live in, your online devices, your household, and even to your
> neighbor’s WiFi usage patterns (yes, their networks also impact your
> environment!). It adapts real-time to your personal needs."

I am glad they sort of recognise that oh, my, merely because you have
a WPA password does not stop you from radiating outside your home/apt.
The fact that this needs an ! to point out is a sad commentary on the
state of humanity.

> I'm reading this as: we setup the APs on all 3 (1,6,11) channels, have the
> plumes record which channels have best radio and network parameters at a
> given time; store that on their disks over the web (sorry, cloud); then
> dynamically set power and perhaps other radio chip and QoS parameters to
> something a la heuristics: e.g. send X type of traffic on ch11 after 6pm
> Mon-Fri.

I have no idea why "the cloud" is needed, as the amount of storage and
computation required to do this fits into a few dozen k.

Well, what I can imagine is the radios sending a constant stream of
statistics to the cloud (thus leaking who is on, when), giant
databases written in python to manage it, plume reselling your usage
data and patterns to third parties, a whole team dedicated to
developing the json api to manage all this, and the gui for your smart
phone relying on hole punching to get the conf data out to the cloud,
so the json api controlling the parameters could make it back into the
device over unsecured telnet over port 22.

... and one or two guys cross compiling openwrt and crossing their
fingers if it will work in the shipped part of the product, and
utterly ignoring rfc7084 and homenet in developing their sort of meshy
technology.

Of all the things in the preso that bugged me most, it was their
proposal for you to throw out all your existing APs for their stuff,
and secondly not documenting how to interop with the real open source
software they are probably leveraging.

Welcome to our silo! Trust us, we fixed everything you've been bugged
about! Thanks for supporting us on kickstarter!



> Doesn't sound like they're making wifi fast in Dave Taht sense. More like
> working around the issues by multiplying channels and squeezing what they
> can from the radio. That may work, who knows.

One of these days I should stress more that the inverse square law for
radio propagation *vastly* trumps anything we can do in
make-wifi-fast, that we 'canna violate the laws of physics, captain!'
No matter what I try for example, the make-wifi-fast AP I have in the
office cannot reach the kitchen.

So I have definately longed to see the end of your typical wifi
"repeater" that rebroadcast on the same channel, and

I would also have liked it if some "meshy" 2 radio diversity routing
product that plugged into a wall like these things do... also had
ethernet over powerline support. *way* better answer that wifi
diversity routing is in many cases.

> Anyway, this is what a consumer like me reads it.

This non-consumer woke up on the wrong side of the bed this morning.

> Best regards,
> Maciej
>
>
>
>>
>>
>>
>> --
>> Dave Täht
>> Let's go make home routers and wifi faster! With better software!
>> http://blog.cerowrt.org
>> _______________________________________________
>> Cerowrt-devel mailing list
>> Cerowrt-devel@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/cerowrt-devel
>
>



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Cerowrt-devel] eero gains competition in plumewifi

[Dave Taht]

On Tue, Jun 21, 2016 at 1:28 AM, Erkki Lintunen <ebirdie@iki.fi> wrote:
> (resent, wrong sender address at first send)
>
> On 06/20/2016 09:16 AM, Dave Taht wrote:
>>
>> I sure wish I knew how they are implementing diversity routing and if
>> they are bothering to pay attention to make-wifi-fast
>>
>> https://www.plumewifi.com/
>
>
>
> Quite a staffing for a startup, the web site lists 46 names and positions
> from which 26 are named as engineers. Wondering if they already are in the
> business of WiFi chip and RF engineering, which enable them to do some
> "magic" to make a network of their own from the plumewifi plugs. On staffing
> resources this isn't on par with Eero, I think.

Well, it seems that post-nest there was a wave of VC investment into
making "quality" things that could be made hi-margin as market demand
(and consumer foolishness) was demonstrated to exist, by that.
Certainly wifi APs were ripe for improvements... and my own bias
towards fixing the standards and making the main APs better (e.g.
ISP-supplied or open source capable) a slower path.

The actual overheads for a company of building a shippable product
(e.g. marketing resources, manufacturing, structured testing, software
and hardware development, etc) I've long recognized, but I would have
loved to have actually had this sort of backing to go and develop a
product that worked right in the first place, long ago, and ship it.
And be making a living while doing so. I guess I'm jealous.

I really hadn't hit on the idea of the bundled, multiple AP approach
as restoring margins enough for new ideas to make it to
productization.
(I still think the all or nothing silo approach is a hard sell, even
for the classic enterprise AP vendors).

eero has published their GPL'd sources, at least, and this is the
latest of a string of products that promised way too much for what
they can possibly deliver on first ship. I keep hoping the onhub will
improve, but so far, it hasn't. And then there are so many failed
attempts with under-staffed over-promised products that have been
through kickstarter.

Ah, well, maybe this time these boxes will be be better in a
demonstrable way. Hope springs eternal.

> - Erkki
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel



-- 
Dave Täht
Let's go make home routers and wifi faster! With better software!
http://blog.cerowrt.org

Re: [Cerowrt-devel] eero gains competition in plumewifi

[Dave Taht]

wow this was an old message thread. plume had last I looked 3
different firmware versions and had been unable to make 'em work with
prior versions of aql....

On Mon, Mar 30, 2020 at 4:21 AM Erkki Lintunen <erkki.lintunen@iki.fi> wrote:
>
> On 06/20/2016 09:16 AM, Dave Taht wrote:
> > I sure wish I knew how they are implementing diversity routing and if
> > they are bothering to pay attention to make-wifi-fast
> >
> > https://www.plumewifi.com/
>
>
> Quite a staffing for a startup, the web site lists 46 names and
> positions from which 26 are named as engineers. Wondering if they
> already are in the business of WiFi chip and RF engineering, which
> enable them to do some "magic" to make a network of their own from the
> plumewifi plugs. On staffing resources this isn't on par with Eero, I think.
>
> - Erkki
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel



-- 
Make Music, Not War

Dave Täht
CTO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-831-435-0729