[Bloat] infrastructure fixes for bufferbloat.net

[Bloat] infrastructure fixes for bufferbloat.net

[Dave Taht]

It is long past time to upgrade and fix the bufferbloat.net
infrastructure. There were a ton of volunteers willing to help back
around december, but I was way too busy (and too broke) to get
anything sorted out...

It's my hope to kick off make-wifi-fast by april, and to do that, more
than a few infrastructure things need to get done... so if I could
round up a few volunteers for a few weekends prior to that to get this
sort of stuff sorted, it would be great.

But first up:

-1) Open discussion: Anything and everything we should do to be more
effective in general?

This question is not limited to discussing infrastructure problems.

...

As for known stuff on the backlog to fix that I can remember off the
top of my head

0.1) establish a mailing list for people helping hold everything
together - we have one - it's called sysmom. Been dead for a while.
i'm ressurrecting it. After this thread completes, let's move the
discussion over there.

0) Need Bufferbloat.net moved to a virtual, it's redmine instance
updated, existing spamm'ed to death userbase nuked and things like
capcha added.

That said, the world has kind of moved onto things like github, and
perhaps a more serious migration off of redmine would be better. I
really hate writing in textile in particular, and perhaps wikimedia
would be a better form for content and curation.

At the same time, I loathe php, and have been looking into more static
site generators like "hugo.io" which is markdown based. I don't think
it would be much of a problem to convert off of textile to markdown.

This has been a quandary for me for over a year now, and I think the
core problem is that I don't want to have to *care* about the darn
site, just write content once in a while, in a real editor....

Anyway....

I have obtained virtual servers from linode, and am willing to supply
root and access and a backup of bufferbloat.net for folk to fiddle
with....

And somewhere related to that was updating/taking over the lart site,
also, if that can get resurrected.

1) Email fixed -

We did a few fixes to make it better in december, but it still isn't
happy making, and I'd like to be able to easily ship patches around,
have procmail, imaps working, and so on

1.1) Mailman improved

new version of mailman has been out for a while

2) SSL fixed - I have been ignoring this primarily due to item 0.

3) Bug database sorted out.

We have a lot of valuable closed bugs. We have a lot of probably
obsolete open bugs.

4) I'd like to retire "snapon" and huchra as such. Basically the big
problem with that is merely that snapon has terabytes of storage which
costs too much to keep in the cloud. My thought is to slam a big drive
into the main dns server and call it a day. Or put a replacement for
both boxes into isc.org - which doesn't charge for disk.

Snapon, was, um, originally my desktop box.

5) ?


-- 
Dave Täht

http://www.bufferbloat.net/projects/bloat/wiki/Upcoming_Talks

Re: [Bloat] infrastructure fixes for bufferbloat.net

[Jonathan Morton]

[-- Attachment #1: Type: text/plain, Size: 304 bytes --]

Count a vote in general for static pages for web hosting where possible.
Running a server side script (in any language, but especially PHP) and
making database calls for every hit is a performance and security nightmare.

Also, spammers can't spam a comment system that doesn't exist.

- Jonathan Morton

[-- Attachment #2: Type: text/html, Size: 367 bytes --]

[Bloat] We are having it all upside down...

[Richard Scheffenegger]

Had to chuckle while reading this... :)

“There is a common misperception that latency is a dominant technical 
measure of performance for broadband,” Dankberg said.


- See more at: 
http://spacenews.com/viasats-dankberg-unfazed-by-mega-constellation-hoopla/#sthash.qbWfmpVN.dpuf

Richard 

Re: [Bloat] We are having it all upside down...

[Jonathan Morton]

[-- Attachment #1: Type: text/plain, Size: 1541 bytes --]

More of that quote:

> Dankberg and ViaSat disagree and point to Exede in the Air, the company’s
broadband service for commercial airlines, as an example of a geostationary
satellite system being given higher marks by consumers than systems using
air-to-ground terrestrial technologies, which offer reduced latency.
>
> “There is a common misperception that latency is a dominant technical
measure of performance for broadband,” Dankberg said. “Yes, latency is
important, but for the vast majority of Internet traffic, speed and
bandwidth are what’s decisive. This captures our technology strategy in a
nutshell.”

I think he's managing to confuse coverage and reliability with speed.
That's a really serious blind spot. A system based only on ground stations
is going to have a lot of dead spots on it, even at 30000ft where line of
sight is much further than at ground level; most notably all those pesky
oceans where passengers have nothing to look out of the window at.

So of course they're going to be more satisfied with a system that works
anywhere than one that drops out seemingly at random - even if it does add
a full second of round trip latency. A system that manages to work
anywhere, reliably, AND with lower latency is going to be preferred over
that.

I did note Dave's ping trace from GoGo, with its interplanetary-scale
latencies. I can't believe that anyone on board would have a functioning
service under such conditions - I hope they weren't paying for it!

- Jonathan Morton

[-- Attachment #2: Type: text/html, Size: 1664 bytes --]

Re: [Bloat] We are having it all upside down...

[Dave Taht]

On Thu, Feb 12, 2015 at 11:04 PM, Richard Scheffenegger <rscheff@gmx.at> wrote:
> Had to chuckle while reading this... :)
>
> “There is a common misperception that latency is a dominant technical
> measure of performance for broadband,” Dankberg said.
>
>
> - See more at:
> http://spacenews.com/viasats-dankberg-unfazed-by-mega-constellation-hoopla/#sthash.qbWfmpVN.dpuf

Well, they claim to be at least routing stuff down from geosync.

I don't know what gogo-in-flight is using, but presently, GoGo seems
to be routing packets through one of the Mars orbiters as the GoGo
ping times jg measured were in the 12 minute range. So whatever viasat
is using is hopefully better than that...

... until the stuff in LEO arrives.

> Richard



-- 
Dave Täht

http://www.bufferbloat.net/projects/bloat/wiki/Upcoming_Talks

Re: [Bloat] We are having it all upside down...

[Joey Padden]

Good salesmanship at least. Can't blame a guy for trying. 

I recently attempted the in flight internet deal and found ~800 ms average ping times to Google DNS. I had 7 Mbps DL and .3 Mbps UL. The impact to browsing was annoying, enough that I got my email loaded and stopped using the connection.  

Hard to believe he is using the in flight services as an example supporting his claim. I'd think it is the perfect counter example. 

-Joey


> On Feb 13, 2015, at 9:49 AM, Dave Taht <dave.taht@gmail.com> wrote:
> 
> On Thu, Feb 12, 2015 at 11:04 PM, Richard Scheffenegger <rscheff@gmx.at> wrote:
>> Had to chuckle while reading this... :)
>> 
>> “There is a common misperception that latency is a dominant technical
>> measure of performance for broadband,” Dankberg said.
>> 
>> 
>> - See more at:
>> http://spacenews.com/viasats-dankberg-unfazed-by-mega-constellation-hoopla/#sthash.qbWfmpVN.dpuf
> 
> Well, they claim to be at least routing stuff down from geosync.
> 
> I don't know what gogo-in-flight is using, but presently, GoGo seems
> to be routing packets through one of the Mars orbiters as the GoGo
> ping times jg measured were in the 12 minute range. So whatever viasat
> is using is hopefully better than that...
> 
> ... until the stuff in LEO arrives.
> 
>> Richard
> 
> 
> 
> -- 
> Dave Täht
> 
> http://www.bufferbloat.net/projects/bloat/wiki/Upcoming_Talks
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] We are having it all upside down...

[Kathleen Nichols]

"If you just look at the efficiency measures, the bigger satellites" are
going to make it easier to for an "unfriendly" to disrupt global
communications.

I think, on the whole, given a scrappy communications protocol, I'd
prefer the lemonade.

	Kathie