[Starlink] revisiting the old "starwrt" plan

[Starlink] revisiting the old "starwrt" plan

[Dave Taht]

[-- Attachment #1: Type: text/plain, Size: 2452 bytes --]

A genesis point for the creation of this email list was this failed, too
ambitious, and too multi-faceted, project proposal from over 2 years back.
See and comment on it here:

https://docs.google.com/document/d/1rVGC-iNq2NZ0jk4f3IAiVUHz2S9O6P-F3vVZU2yBYtw/edit

In reviewing it today,  I am so pleased that everyone here has leaned in on
the measurement side described therein, and also there seems to be a ton of
information about how starlink´s spot beams, etc seemed to work that may or
may not be obsolete within it that I would like to update.

In retrospect, I gave up on the idea of making all of openwrt, and cool
subsystems like samba or matter,  interoperate with the mesh code on the
starlink routers, or of producing a build that "just worked" on their
routers. It was too hard, and too locked down, to even try.

I think broader mesh interoperability would be great however - living in an
age where an eero cannot interoperate with a tp-link, and so on, evades the
true beauty of wifi, and I still hold dear the ideas of starlink powering
village connectivity and local services like email, etc, as well as being a
better backup to other links, gaining BGP support, in particular.

On the embedded front, their first router was IPQ8014 based, for which I
had labored for a couple years to finally produce a very good wifi
subsystem for, the second (and presently most popular I think? 1+ million
in the field) is mt76 based which is getting better all the time, and their
latest wifi6 product I know nothing about, and being deeply cynical about
all the wifi6 devices I had tested before exiting the embedded the market,
would like someone else to test.

I personally exited the embedded market in favor of working on libreqos
over a year ago.

... and it was mostly the dishy that in the end I felt just needed a few
tweaks, and their headends.

Anyway: given that openwrt is nearing a final release in this development
cycle and I have also been scoping out improvements to CAKE based on what
we have learned from the mikrotik and LibreQos deployments, thinking about
what is still wrong with wifi was on my mind today. In particular, I
enjoyed reminiscing about the fire I had had in my veins back then about
getting in there and making it the best ISP router on the planet.




-- 
Oct 30: https://netdevconf.info/0x17/news/the-maestro-and-the-music-bof.html
Dave Täht CSO, LibreQos

[-- Attachment #2: Type: text/html, Size: 3111 bytes --]

Re: [Starlink] revisiting the old "starwrt" plan

[Dave Taht]

Sometimes I look at this old document and dream about what might have been.

On Thu, Sep 21, 2023 at 7:37 PM Dave Taht <dave.taht@gmail.com> wrote:
>
>
> A genesis point for the creation of this email list was this failed, too ambitious, and too multi-faceted, project proposal from over 2 years back. See and comment on it here:
>
> https://docs.google.com/document/d/1rVGC-iNq2NZ0jk4f3IAiVUHz2S9O6P-F3vVZU2yBYtw/edit
>
> In reviewing it today,  I am so pleased that everyone here has leaned in on the measurement side described therein, and also there seems to be a ton of information about how starlink´s spot beams, etc seemed to work that may or may not be obsolete within it that I would like to update.
>
> In retrospect, I gave up on the idea of making all of openwrt, and cool subsystems like samba or matter,  interoperate with the mesh code on the starlink routers, or of producing a build that "just worked" on their routers. It was too hard, and too locked down, to even try.
>
> I think broader mesh interoperability would be great however - living in an age where an eero cannot interoperate with a tp-link, and so on, evades the true beauty of wifi, and I still hold dear the ideas of starlink powering village connectivity and local services like email, etc, as well as being a better backup to other links, gaining BGP support, in particular.
>
> On the embedded front, their first router was IPQ8014 based, for which I had labored for a couple years to finally produce a very good wifi subsystem for, the second (and presently most popular I think? 1+ million in the field) is mt76 based which is getting better all the time, and their latest wifi6 product I know nothing about, and being deeply cynical about all the wifi6 devices I had tested before exiting the embedded the market, would like someone else to test.
>
> I personally exited the embedded market in favor of working on libreqos over a year ago.
>
> ... and it was mostly the dishy that in the end I felt just needed a few tweaks, and their headends.
>
> Anyway: given that openwrt is nearing a final release in this development cycle and I have also been scoping out improvements to CAKE based on what we have learned from the mikrotik and LibreQos deployments, thinking about what is still wrong with wifi was on my mind today. In particular, I enjoyed reminiscing about the fire I had had in my veins back then about getting in there and making it the best ISP router on the planet.
>
>
>
>
> --
> Oct 30: https://netdevconf.info/0x17/news/the-maestro-and-the-music-bof.html
> Dave Täht CSO, LibreQos



-- 
https://www.youtube.com/watch?v=BVFWSyMp3xg&t=1098s Waves Podcast
Dave Täht CSO, LibreQos