Re: High Speed Data Network

[Justin Wagner]

Jim, how about you?  Are you only in Pullman during the school year?

Tom, good idea with the Google Group!  I have added myself and included the group in the CC of this email.

Justin

On Mon, May 4, 2009 at 10:16 AM, Tom Hayward <thay...@wsu.edu> wrote:

> I agree that a meeting is needed for us to formally get some goals defined
> as well as possible solutions.

I will be returning home (Gig Harbor, WA) for summer after more tower
work this Saturday. This means it will be difficult for me to meet
until fall. If you pick a venue with Internet access, I'd love to join
via Skype or equivalent.

I'd like central place to discuss this project. Something that
archives our discussion. How about a Google mailing list? I just
created this one: http://groups.google.com/group/hswd

Tom KD7LXL

[Justin Wagner]

OK looks like it would be best to have the meeting in Spokane since you will be the only attendee from Pullman.  Do you have any conflicts with Saturday May 16th, 23rd or 30th?

Also do we have any possible locations for the meeting?

Justin

On Mon, May 4, 2009 at 12:37 PM, Jim Kusznir <j...@kusznir.net> wrote:

I am staff at WSU on 12-month appointment, and will be here throughout
the summer.

--Jim

[N7BFS]

I feel we should keep the discussion on this list for now as there are
many topics to discuss and all our schedules are too busy at this
point it seems.

I would like to invite Karl (AK2O) to the group as well as he will
have a stake in the VHF clubs part and may be able to add his
perspective.

If that is ok with the others in the list.

Doug
N7BFS

On May 4, 12:41 pm, Justin Wagner <jrw...@gmail.com> wrote:
> OK looks like it would be best to have the meeting in Spokane since you will
> be the only attendee from Pullman.  Do you have any conflicts with Saturday
> May 16th, 23rd or 30th?
>
> Also do we have any possible locations for the meeting?
>
> Justin
>
> On Mon, May 4, 2009 at 12:37 PM, Jim Kusznir <j...@kusznir.net> wrote:
> > I am staff at WSU on 12-month appointment, and will be here throughout
> > the summer.
>
> > --Jim
>
> > Justin Wagner wrote:
> > > Jim, how about you?  Are you only in Pullman during the school year?
>
> > > Tom, good idea with the Google Group!  I have added myself and
> > > included the group in the CC of this email.
>
> > > Justin
>

[Justin Wagner]

That is totally fine, I think the more people we have involved, the
better off we will be.

Justin Wagner

[Tom Hayward]

> I would like to invite Karl (AK2O) to the group as well as he will
> have a stake in the VHF clubs part and may be able to add his
> perspective.
>
> If that is ok with the others in the list.

Please invite anyone you think can contribute to the project. I
created the Google Groups reflector as a public group hoping to
attract attention and support from other hams.

Tom KD7LXL

[Jim Kusznir]

I think its great to get Carl in the group!

Its looking more and more that I won't be able to meet in person
before late May or early June (although I do have two weekends in
early June that I'm already going to Spokane).

--JIm

[Justin Wagner]

Ok, looks like June is going to be better, the only conflict I have is
on the weekend of June 27th I will be out of town, other than that I
should be free.

Justin Wagner

[Karl in Spokane]

Hi folks, I finally got a chance to find and join this group. For some
reason I keep thinking it was a Yahoo group and spent some time on
there looking around. That's what I get for spending a lot of time on
repeater maintenance and not thinking like a geek, LOL

At any rate I wish to monitor, discuss and support these types of
ideas, plans and other related discussions. Being the tech for the VHF
Club and its assets, including a nice packet network already in place
for the last 5+ years we might be able to come up with something.
What I would like to see is like what Doug mentioned that we need to
have a goal, then see how to get there technically speaking.

I don't remember who was on the cc list, however Mike Carrey has set a
meeting in July for this type of thing. Unfortunately some of the key
people may not be able to make it. I have dispatched an email Mike and
that group suggesting a re-sced.

If anyone needs to get in touch with me, there are many ways. The air
on 147.20 is best, however I have email, web site, home number, etc.
All this information can be obtained by visiting my site of http://www.srgclub.org

Regards,

Karl (with a K) in Spokane

[Jim Kusznir]

I was wondering if the 9600 baud meeting was brought about somewhat by
this, and what the cross-section was of people in this group vs the
people in the 9k6baud group.

I was also wondering if people from this group wanted to get together
before the 9k6 meeting, and if any of our stuff should be discussed
there or not.

I'm leaning a little bit towards "not", as at this point, I consider
our work as "experimental", and I get the impression that the meeting
is about "production" packet systems.

I do see an opportunity to provide "worm hole" through our network
if/when its built, to provide a backbone to the existing system.

It also appears that a critical mass of people wanting to improve from
1200 baud may have been reached. This could help with acceptance of
our system, if/when deployed.


I do want to meet with everyone and start hammering out some goals and
the like. I've got some ideas for goals in my mind; if interested, we
can discuss here or deffer to an in-person meeting.

--Jim, K7LL

[Tom Hayward]

> I do want to meet with everyone and start hammering out some goals and
> the like.  I've got some ideas for goals in my mind; if interested, we
> can discuss here or deffer to an in-person meeting.

I won't be able to make it back to your side of the state for the
meeting, but I'd love to join via Skype, etc.

Tom KD7LXL

[Justin Wagner]

I didn't even know about the 9k6baud group meeting, so I guess that makes at least one who is just on this group and not that one.  I am pretty flexible with meeting times and I would love to get together to discuss plans etc.  I have a feeling that there is a lot for me to learn about existing packet systems to be able to be very helpful in the process.

Justin KC7JRW

[N7BFS]

Hi Jim, I noticed you are doing some work with the LinuxRMS project
and so that will more than likely dovetail into the '9600' meeting as
that is the direction we are wanting to go towards (at least for the
short term). There is a need for faster access to the Winlink system
and having a 9600 RMS would help in that direction, I have the
hardware from the old BBS and will be using that to setup a LinuxRMS,
so if I can get your input/help that would be very much appreciated.
That can be taken off this list and kept on the yahoo group if you
like.

With that said I would think that there will be overlapping on this
list as there are not that many players in our line of interest, so we
should just make the best of it but also try and keep our projects
clear so we don't get fractionated.

Lets talk about goals on here, its easier on a list like this as we
can do it in the lulls in our busy lives (well I mostly speak for
myself on that last one)...

Have fun this weekend on Field Day!

Doug
N7BFS

> > All this information can be obtained by visiting my site ofhttp://www.srgclub.org

[Jim Kusznir]

Ok, so here goes my initial statement attempting to answer the "why":

First and foremost, Amateur radio pushes the envelope. Technology is
always evolving, and Hams tend to deploy this. When there are no
developments in a field for a time, there is a decent chance that
interest will wane. That has happened in non-HF data communication
(One can even argue in HF: New modes have come out, and much focus
has been switched to them...). Packet communication, or other local
area data methods, haven't really evolved in the last 2 or so decades.
As a result, we get a lot of statements like what I heard about
packet for Bloomsday: Each year, many of the packet volunteers
haven't done it since last year's bloomsday, and so they've forgotten
how to make it work and have to spend a while to fix it back up". We
also get reduced numbers of users, and consequently reduced interest
in the infrastructure.

So, Goal #1 is to revitalize interest in local data over ham radio
communications by providing a newer, more relevant mode.

Building on that is the relevance / usefulness. Packet has generally
always been about moving data. HF it often is more about getting
contacts...Sure, you're moving data, but the focus is being able to
talk to people. However, since I've gotten involved in packet (mid
1990's), the focus wasn't so much about making contacts, but moving
specific blocks of data (often e-mail) from one place to another. In
its early days, it was a new and novel thing: We could actually get
an electronic message across the US in a couple days to a week
(usually beating US Mail), and across the world in a little bit more
time (but definitely beating the mail). We were one of few "games in
town" that could do this, and the only one accessible to many hams.
As such, packet flourished. Then came the Internet. Hams were
involved from the beginning in the Internet, and we even got our own
class A of IP addresses assigned: 44.x.x.x. There were grandeurs
plans of connecting the packet radio to the Internet and of hams
having IP addresses, etc. However, the Internet continued to evolve
at an unheard of rate, pushing consumer data moving technologies, and
soon most Hams who were interested in moving data across the country
(or next door) were finding ways to do it on the Internet much faster
and easier than on packet radio. The Internet also began to offer a
much richer application suite that far surpassed that which Packet had
to offer. Pretty soon, development stopped on Packet, as it was just
too far behind the Internet and there wasn't enough demand anymore to
push it.

In my opinion, for a data network to be relevant today, it needs to
support IP protocols so that the applications and services that
society has become accustomed to can run over it, and it needs to run
at speeds at least within that which end users can get commodity
Internet access at. After all, if you're going to run a web server on
1200 baud packet, most web pages will take a long time to load...far
beyond what the Internet has conditioned us to accept. That's just
from the protocol overhead and slow singling rate. Even at 9600 baud,
things would be pretty bad. I'd say you'd need to go up to at least
56k for anything to begin to be taken seriously. It also appears that
these days, neary any network technology outside of ham radio supports
IP. People have even gone to great lengths to support IP on older
technologies. New protocols that failed to support IP natively (eg,
ATM) more or less died even though they may have been superior just on
account of not being built around IP. For a network to be relevent in
today's age, it must support IP.

So, Goal #2 is to make radio-based data communication relevant by
providing IP protocol support, and speed and response time that is
within the rage society has become accustomed to. Otherwise, people
won't use it on a regular basis (which is what we have with Packet
today).


So, those are my two primary "meta-goals" of this project. While
these are more idealogical than practical, they lead to a plethera of
practical outcomes. However, I think they poise a more important goal
than all the individual practical goals: They represent revitalizing
ham's interest in wireless data communications, and hopefully, start
making data communications a commonplace thing again.

But without some practical goals, we're still left with little to
actually get our hands dirty with and little to measure progress with.
So, here are some shorts on that subject:

1) start by providing a high speed backbone to the existing packet
network. Currently, it takes paiinfully long times to cover any
reasonable distace because of node hops. At 1200 baud, it takes
several seconds minimum for each node hop. With a high speed network,
that drops to milliseconds. If we had a statewide high speed network,
you could connect (at 1200 baud) into your local packet node, then
connect to a node or other station in Olympia or vancover, and still
get return times similar to talking to another node/station in
Spokane. Also, by increasing the backbone size, it becomes much
harder to saturate it and move large amounts of data over it.

2) Provide some end-user access, espicially access into the above
(AXIP probably). This would at least allow individuals in the field
to connect with high speed equipment and do things like winlink at
speeds that can only be comprabal to direct internet connections. It
can also allow for high speed ham-to-ham connections. For example, in
an emergency communications situation, a served agency needing to get
a large file or perhaps some high resolution pictures to another site
in the region.

3) provide BASIC internet connectivity. This can provide hams in
remote locations the ability to browse the web page, or better yet,
when those of us who work on mountain top gear find they need
information about a particular piece of equipment, and they forgot to
bring that manual, they can just go download it or do the research
they need while still at the site! This can also be useful in
emergency communications situations: Being at a field command center
for a fire and being able to download information directly from NOAA
and other useful sources, etc.

4) With the above done right, it becomes possible to locate IRLP nodes
directly on the repeater controllers at the mountain tops! It also
allows for moving of some analog links to digital, but keeping them
completely over radio (for example, repacing the Evergreen Interntie
with a digital, but still all over radio, method of linking.

5) we could even go so far as to develop our own VoIP network,
complete with IP phones. This would result in EOCs having "hamphones"
that can call other hamphones (say, a county emergency manager
communicating directly to a neighboring county's emergency manager.
Of course, we have radio to do this too, but it may be simpler, and if
the material is somewhat sensitive, it doesn't have to be broadcast
wide-area. It also allows us to cary more concurrent conversations
rather than tying up wide-area repeaters with a single conversation.

6) How about doing fast-scan Amateur Television, but the
easy/inexpensive way? Just do a video-over-IP solution on the high
speed network. Now you can send video anywhere over the network and
not need all the expensive/somewhat rare gear...Anyone with an
ethernet-enabled webcam or a laptop and a USB cam is now equipped!

The list can keep going on and on, of course.....

There are also the technical goals:

Construct a reliable, meshed, decentralized network running at high
speeds covering large regions, and growable....

Anyway, this should be enough to get things started.

--Jim, K7LL