[NNagain] nn back in the news

Frantisek Borsik frantisek.borsik at gmail.com
Sun Aug 18 15:32:19 EDT 2024


I think that top level items you have mentioned, Dave, for kicking-off a
follow-up NOI filling: "harping on our latency points, pointing to
progress, and the need for more IXPs" are MORE than enough for starters.
There are some improvements on the latency front (notably Starlink's
implemenation of FQ-CoDel) so it's a good set of points to start with, IMO.
______________

I have returned from Africa yesterday, where I was attending MwNOG-1 and
SAFNOG-9 <https://safnog.org> in Lilongwe, Malawi. Imagine a country of
almost 22 million people, having opened up ONLY its 2nd IXP, a few days
ago: https://lionex.org. Literally anyone and their mother, with any means
at our hands, should offer them some kind of help. There was a nice
development during the event - a pledge by MACRA: Malawi Communications
Regulator Authority, to open up the 3rd IXP in Malawi, this time in the
city of Mzuzu, their third biggest city, with an important university:
https://www.linkedin.com/feed/update/urn:li:activity:7229821715822899200/

I don't want to hijack this thread, but if there is someone here from IXP
world, willing to explore ways how to help Malawi, please - contact me
privately here or on LinkedIn and I will connect you with Brian Munyao
Longwe <https://www.linkedin.com/in/brian-munyao-longwe-25432/> - he and
quite a lot of amazing people around him, are getting things done down
there in Malawi and I'm so thankful for a chance to see that. By any means,
feel free to contact him directly if you want to: brian at ctnmw.net

_______________

Getting back on track: I recall Paul Vixie mentioning PAIX somewhere on
LinkedIn this week - couldn't find it today, so I was just searching the
Internet and found this:
https://www.enog.org/presentations/enog-2/12-2011-11-28-paix-history.pdf

Isn't  there a time to try it again? Or something similar, for that matter?
Lots of things has changed - a global level of understanding of IXP
importance, for example...
I mean, even with a current uptake in installation we can see at
https://libreqos.io, with what Preseem, Cambium Networks QoE, Paraqum and
Bequant are doing with QoE middle-boxes as  force multipliers to fix
latency, IXP component is badly needed.


All the best,

Frank

Frantisek (Frank) Borsik



https://www.linkedin.com/in/frantisekborsik

Signal, Telegram, WhatsApp: +421919416714

iMessage, mobile: +420775230885

Skype: casioa5302ca

frantisek.borsik at gmail.com


On Thu, Aug 15, 2024 at 8:06 PM Jack Haverty via Nnagain <
nnagain at lists.bufferbloat.net> wrote:

> IMHO, educating politicians and regulators is difficult but necessary.
> Latency is an important and overlooked component of "broadband service",
> not included in the definition of such "service".  But there are others
> that might also catch their attention.
>
> I live in an area in California where wildfires are an issue.  My location
> officially has broadband service, according to the FCC databases.  I get IP
> through cable, provided by Comcast.   The IP{ service is fast and latency
> hasn't been an issue for me.
>
> The government (state, local, and federal) have all spent a lot of money
> and effort to put technology in place to  mitigate the risks of wildfire
> and other such natural risks, to help keep the public safe.  That's their
> job.  We have sensors and cameras that spot smoke columns or seismic
> events.  We have apps and communications channels that spread the word
> fast, so that people can get out of harm's way.  The Internet has made
> quite a difference by providing the infrastructure to make such safety
> mechanisms possible.  (See app.watchduty.org for an example)
>
> The local electric utility has also implemented mechanisms to avoid
> causing wildfires.  When it's dry and windy, and therefore the risks of a
> falling tree limb starting a fire are high, they simply shut off the power
> to the area.  When the risk is over, they turn power back on.  It might be
> a day or two, but while the risks are high, a fire won't be started by the
> electric grid.  Many people have installed generators to keep essential
> things going in their homes.   The electric utility is even offering
> subsidies for people to buy such equipment.
>
> However...
>
> Comcast relies on the electric grid for power.  Some areas have equipment
> with backup battery power, but that only works for an hour or two.  After
> that, communications shuts down - in my neighborhood that means telephone,
> TV, and Internet all cease to work.   All of the information from those
> sensors, cameras, and emergency service agencies doesn't get to us, the
> public it's supposed to help.
>
> Despite that behavior, officially the database says we have broadband
> service available.  It of course is available, most of the time.  It turns
> off when you need it the most.
>
> Thinking techologically for a bit, the problem is a lot like Latency.
> Latency is not specified in the definition of "broadband service".
> Neither is "Reliability".
>
> I've filed a "challenge" to my own database entry, pointing out that we
> don't actually have reliable broadband service.  Even if, over a month or
> so, the service is 99.xxx% "up", that's not much help if it's down when you
> actually need it the most.   The definition of "reliable" needs to be
> revised.
>
> I don't expect any action from my challenge as just a lone voice in the
> crowd.
>
> Perhaps educating the politicians and regulators about reliability would
> grab some attention.  It might be a bit easier to understand than just
> "latency".
>
> Jack Haverty
> Nevada City, CA
>
> _______________________________________________
> Nnagain mailing list
> Nnagain at lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/nnagain
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.bufferbloat.net/pipermail/nnagain/attachments/20240818/4078e582/attachment.html>


More information about the Nnagain mailing list