[FADCA] Re: Willie and Joe - and WiFi. (Where is the software?)
Gerry Creager N5JXS
[email protected]
Mon, 24 Mar 2003 19:37:49 -0600
Actually, while applications will be critical in making this work a lot
of them are going to work right out of the box. A lot of the apps, such
as the e-mail, web browsers, and instant messaging, won't care what sort
of network connection they get.
Any, yes, Hinternet is at the seedlig stage, although there's a fair bit
of fertilizer being applied... in one form or another.
As for Mitreks, I've been into a few of them over the years. We've been
using them for 9600 baud here (well, a few of us, anyway) for at least
10 years... Of course, the group I hang out with also puts in Mastr2 and
MastrExec repeaters for our linked system... Expanding to 56K would be
an interesting, and possibly more robust technology for links between
cities, as a backup to emergency data communications.
One consideration at microwave frequencies is that the free space
pathloss is greater than at UHF. And, if you were to try 802.11a at 5.4
GHz, you'd also see water absorption. But, I digress...
Perhaps we'll get an opportunity to discuss geodesy some day. I have
been known to spend a fair bit of time using GPS receivers for some
innovative things. Like, surveying to the cm level over extremly long
baselines. While I realize this isn't geology, it likely is something
you've met in a career like you outlined!
73, gerry
bud thompson wrote:
> Deltona, FL
> Monday March 24 1135EST
>
> Gerry:
>
> Thanks for weighing in with a response. I'd cc: you as only one of two of
> the people on the WiFi e-mail list as I'd noticed one of your notes there
> that at least I understood a little and you seem to be in the lead.
>
> Your detailed response is much appreciated, though much of it is either (1)
> out of my immediate interest or (2) very much over my head.
>
> I also interpret that (1) my concerns over end user applications/software
> are valid, and (2) the WiFi Hinternt is really at the seedling stages - not
> even grass roots yet. That makes it even more exciting to me.
>
> My main interest is applications by users - I'll leave the hardware, and
> infrastructure-building to folks like you who have much more experience in
> those matters.
>
> My audience was primarily folks on the FADCA e-mail list - a group which is
> upgrading a 1200b/9.6kb state-wide packet network. That said, many on that
> list will understand your response in more depth than I and may comment. We
> do have some rocket science-level networking guys on the list. Some even
> speak Linux!
>
> What we do have are sites - as a result of the Florida Layered Packet
> Network. Most "metro" (LAN) areas are covered, and we have vhf or UHF
> backbones between many of them.
>
> Talk about disparate levels of technology - I'm just now finishing
> converting 1980s vintage UHF Motorola Mitrek radios to 9.6kb to complete the
> upgrade for the three switches I operate between Orlando and Daytona Beach!
> Once the three are operating, I'll start a more serious study of all the
> work you WiFi guys are doing, As fast as the capabilities will likely
> emerge, I'll not likely be spending any more time or money "upgrading" my
> three switches at the AX.25 level beyond 9.6kb!
>
> If we can get 802.11(b) or (g) to the point where we can have adjacent APs
> backboned together at 30 miles, I'll not need but one more site to cover the
> populated areas of nearly 1/5 of the population in the state of Florida.
> Since you're working in Houston, you'll appreciate the flat terrain and
> vegetation problems we have here. That said, each of my sites can more than
> "see" each other. There is excess elevation above terrain- I just don't
> know how much excess for 2.4 -5.7GHz!
>
>
>>Then, I believe we can handle the metro area a bit easier than most
>>think. Recall that few of us have bought gear, pointed antennas at each
>>other and tried!
>
>
> If all goes well here, I'll get a WiFi 802.11b test set up going here by mid
> summer... perhaps over a 5-, maybe 12-mile path for some empirical
> investigations.
>
> We'll see where it goes from there.
>
>
>>. . .I'm privileged to have a lab and grad
>>students I can get working on this, and we'll all learn.
>
>
> Texas A&M? If I were 45 years younger, I'd come study and be one of your
> grad students! (Will my MS in Geology circa 1963, 34 years as a
> petrophysicist, and 51 years in ham radio get me in?)
>
> Thanks again for taking time to respond.
>
> bud Thompson N0IA
> 386 574 4124
>
> ---------------------------------------------------------------------
> "Are you still wasting your time with spam?...
> There is a solution!"
>
> Protected by GIANT Company's Spam Inspector
> The most powerful anti-spam software available.
> http://www.giantcompany.com
>
>
> ----- Original Message -----
> From: "Gerry Creager N5JXS" <[email protected]>
> To: "bud thompson" <[email protected]>
> Cc: <[email protected]>; "Pete Hoffswell" <[email protected]>;
> "Paul Miller N4WKQ" <[email protected]>; <[email protected]>
> Sent: Sunday, March 23, 2003 23:39
> Subject: Re: Willie and Joe - and WiFi. (Where is the software?)
>
>
>
>>Inline (sparse) comments below... You make some good observations...
>>
>>bud thompson wrote:
>>
>>>Hi, Florida Amateur Digital Communications Association e-mailers - and a
>>
> few
>
>>>others who may be helpful.
>>>
>>>Deltona FL 3/23/2003 10:13:43.8766
>>>
>>>(The topic is Wireless Fidelity - WiFi - I'll get there in a couple of
>>>paragraphs.)
>>>
>>
>>...
>>
>>>Over the past month I've subscribed to the ARRL-sponsored HSMM-related
>>>WiFi e-mail reflector. I've read most, and scanned the rest, of the
>>
> notes.
>
>>>(Geeezee, is it a HAMTHING to flame, or a general male trait? After only
>>
> two
>
>>>weeks the flames were gaining in numbers. After working my way through
>>
> the
>
>>>10 to 15 e-mails a day I have come to the following conclusions:
>>>
>>>1. Most of the 802.11(b) folks are interested in the RLAN (i.e. local
>>
> area
>
>>>net) aspects of a nationwide system that would allow ANYONE with a
>>
> laptop
>
>>>and wireless network card and a short stub of an antenna INTERNET access
>>>anywhere in CONUS. - A nationwide wireless dream.
>>
>>I tend to think linearly when faced with a problem of this sort. From
>>my perspective, we have to first handle the local area problems of (1)
>>authentication, (2) identification (to a security geek, they're NOT the
>>same) and, (3) capabilities management.
>>
>>Then, I believe we can handle the metro area a bit easier than most
>>think. Recall that few of us have bought gear, pointed antennas at each
>>other and tried!
>>
>>The rather strident call for full-mesh networking is something I'll be
>>testing over the next month... I'm privileged to have a lab and grad
>>students I can get working on this, and we'll all learn. However, mesh
>>networking isn't easy, and yes, Houston, it does approach rocket
>>science. Given a choice of Layer 2 protocols to try to build a mesh
>>with, it'd be Asynchronous Telecommunications Mode, not some form of
>>Ethernet.
>>
>>Finally, I'm hard over that, if we WANT to make this a connect-anywhere
>>system, we're going to have to leverage all the potential benefits and
>>gegaws available in the IPv6 protocol, and not ever look back.
>>
>>
>>>2. Many of the hams involved are SHF RF geeks who want to prove they
>>
> can
>
>>>extend RLAN to RLAN backbones with either (a) less or (b) more power
>>
> output
>
>>>than anyone else.
>>
>>Been around AMSAT lately? And that's not really a slam at AMSAT, but
>>said as one who's played with them as a member for years.
>>
>>
>>>3. Some (as usual) are ham barroom lawyer wannabees who challenge
>>>everything, reciting part and parcel from Part 97, offering little else.
>>
> > >
>
>>>4. Not many are promoting specific applications (i.e. uses by end
>>
> users)
>
>>> - just Internet access sans telephone lines or CATV cable.
>>>
>>>Here I come with ideas for applications OTHER THAN just wireless
>>
> Internet
>
>>>access, but without sufficient COMPUTER applications experience. (I've
>>
> been
>
>>>'in' personal computers since 1981 but only got my first sound card in
>>>October last year. Echo Link is the only sound card application I've
>>
> used.
>
>>>The only computer game I've played since PONG is Hearts. There MUST be
>>
> a
>
>>>breadth of computer applications out there that can help with
>>
> integrating
>
>>>WiFi RF networking to ham applications (i.e. EMCOMMs) that do not use
>>
> the
>
>>>wired Internet.
>>>
>>>ARRL has coined this Ham Internet the HINTERNET.
>>>
>>>REFERENCE Page 28 April 2003 QST - . Figure 2. The Access Points (AP)
>>>represent LAN/Switches just as with FPAC/ROSE on our layered network.
>>
> The
>
>>>HAM1, HAM2, HAM3 represent EOCs, Served Agency offices, tactical
>>
> stations,
>
>>>(a.k.a. users). (See http:www.fadca.org for the existing packet
>>
> network.)
>
>>>Now - get out of that box and stop thinking radios. No vhf/uhf/HF radios
>>
> or
>
>>>antennas anymore. Only WiFi cards and sufficient antenna to reach an
>>
> AP. No
>
>>>TNCs any more. No telegraph keys, semaphore, or smoke signals any more.
>>>
>>>What USER computer software/applications are presently available to do
>>
> the
>
>>>following (all of which is described as possible applications on page
>>
> 29)?
>
>>>1. End-to-end digital messaging (e-mail) between "users".
>>
>>Netscape/Mozilla mail. Heaven help me, but, IE/Micro$oft Lookout.
>>Evolution. GAIM. Jabber.
>>
>>
>>>2. Voice over IP (actually full duplex telephone) between/among users.
>>
>>NetMeeting. OpenH323/Gnomemeeting. A slew of Polycomm and other
>>commercial h.323 apps.
>>
>>
>>
>>>3. Video and video conference between/among users' computers.
>>
>>Same. Voice and video properly fall, today, under H323. Maybe
>>tomorrow, it'll be SIP.
>>
>>
>>>4. Web Cam applications.
>>
>>Plethora. Too many to enumerate. Recall, however, that a lot of
>>streaming webcams want 12 Mb/sec (USB 1.1) and will have to have size
>>and framerate dropped to not be bandwidth hogs on a fixed 11 Mb link...
>>
>>
>>
>>>5. Remote Control by HAM1 of radios, etc at HAM3 location.
>>
>>Does it currently work over a wired network? Next question?
>>
>>
>>
>>>Microsoft Netmeeting is referenced in the article. How much of this will
>>
> it
>
>>>handle?
>>
>>All you want, but not necessarily well. NetMeeting is not exactly H.323
>>compliant, but, in general, it's close enough.
>>
>>
>>
>>>I know there are 'chat' and 'instant' messaging schemes that probably
>>
> look
>
>>>like ham digital stuff (upper screen receive text, lower screen transmit
>>>text), but the controlling interface between two users is on the
>>
> internet.
>
>>>E-mail, per se, requires servers, presently though the internet. How is
>>>such handled on our emerging Hinternet w/o an internet connection? (Rick
>>>KN6KB: For e-mail messaging can Airmail-to-Airmail be the answer if both
>>
> use
>
>>>Telnet terminal? I'll bet there needs to be a Telnet server someplace
>>
> (e.g.
>
>>>WL2K PMBO).)
>>
>>Er, let's use something like AIM/GAIM/Jabber for instant messaging, and
>>something that speaks SMTP for mail.
>>
>>
>>
>>>I know there is software for all of these applications, but since I've
>>
> not
>
>>>used it, I don't know how much of it supports computer-to-computer w/o
>>
> some
>
>>>interface. I suppose that "connections" could be made based on IP
>>
> addresses
>
>>>if the software doesn't need an intermediate controlling interface. We
>>>certainly need more than a ping test from the Orange County EOC in
>>
> Orlando
>
>>>to the Daytona Beach Red Cross office!
>>
>>If it runs on a wired connection, it'll run on a wireless connection, as
>>long as you don't overwhelm the bandwidth/headroom. We're all talking
>>about 802.11b here; by the time this goes into the Ham Mainstream, we're
>>really gonna be talking about 802.11g, which, although not yet ratified,
>>is showing up on the market. Overlaying the 802.11b stuff.
>>
>>
>>
>>>Are we going to have to supply local or regional "servers" on our
>>
> Hinternet?
>
>>>Presently, my ISP has a server that gets me to the internet, another
>>
> that
>
>>>gets my incoming e-mail, and another that takes my outgoing e-mail.
>>
>>Some ham or club should step up to provide the server(s) as needed. for
>>the ham population in my hometown, we're looking at a surplus AMD K2/450
>>with 256 MB RAM and an interface to HSMM and one to an ISP. It can do
>>mail, DHCP (which I do NOT support for this; earlier comments about
>>IPv6); DNS and routing.
>>
>>
>>
>>>My concept is that at user end points (EOC, Served Agency, ham QTH,
>>
> mobile,
>
>>>or fixed tactical station), a ham would sit in front of a computer with
>>>headphones/microphone along with keyboard/mouse. Each user must have a
>>
> way
>
>>>to connect to a single or to multiple user(s). A connect would also
>>
> include
>
>>>media/mode type (e-mail, telephone, video, etc) I imagine this could
>>
> look
>
>>>like the Echo Link menu (point, click, connect) - but without the
>>
> Internet,
>
>>>where is
>>>this software to support end user to end user applications?
>>
>>It's not nice to read my mind... But if it looks like Echolink, I'm not
>>too sure I'll like it. Personal bias. There are apps that do this
>>without Echolink, which ALMOST but not quite employs H.323.
>>
>>I'm a standards sort of guy. If you're gonna write the code using a
>>piece of a standard, find the minimum spec and get to that point, at
>>least. Microsoft has proven we can do otherwise, or can attempt to
>>redefine the spec regardless of international (or national) groups
>>working together. But, tell me: Is that the right way to do it?
>>
>>
>>
>>>I'm guessing we would need UMBRELLA "menu software" of some sort to tie
>>
> the
>
>>>user addressing to media/modes. It would be a hassle to have to run a
>>>separate program for each mode (e-mail, telephone, video, etc.) each of
>>>which had its own point and click connect scheme. The menu should be in
>>>user-supplied English, not IP addresses.
>>
>>No, I disagree. We need a suite of apps laid out logically on the
>>desktop. They need to look like, if not be, the apps folks use
>>regularly. Some of the apps, specifically the H.323 stuff comes to
>>mind, will require a _little_ training, but not too much. The
>>overwhelming umbrella menu restricts what people can do. Often to bad
>>ends, because it doesn't get fully debugged until you're in the midst of
>>an operation.
>>
>>
>>
>>>How much of this can Microsoft Netmeeting handle?
>>
>>Voice and video over IP. Probably satisfactorily. Instant messaging,
>>to a MUCH lesser degree, and not well.
>>
>>
>>
>>>There are other applications questions - Say at an EOC - We presently
>>
> have
>
>>>three hams each with a radio. One is vhf, one uhf, and one HF. There
>>
> might
>
>>>even be another radio for packet with a TNC and computer. How do all of
>>>these guys use the Hinternet? Does each have to have a separate
>>>computer/WiFi card, or can we use something like WinXP or Windows 2000
>>
> and
>
>>>have multiple users on a single computer via wired LAN? Each ham would
>>
> only
>
>>>have a keyboard, mouse, headset/mic and monitor.
>>
>>*LINUX* said the Linux bigot. XP and 2k approach the stability we need,
>>but neither does networking well, and neither do it right. Microsoft
>>has screwed up almost as many networks as Cisco support.
>>
>>That said, XP is better than any previous MS offering. Active directory
>>should be banned from all these networks, and MS network filesharing
>>should similarly be banned. Novell is right out. These are for
>>technical reasons. I can expand if I need to.
>>
>>For those of you who haven't figured it out, I'm not a Microsoft fan.
>>However, pragmatically, we need to plan around its limited use. It's
>>possible to incorporate a local area wired LAN that allows MS
>>"networking" and the wireless side that doesn't. That's a good approach
>>if we are stuck with someone who can't think outside the MS box.
>>
>>Being able to identify the necessary apps is a start. Assembling them
>>on a platform is another one. However, most of what I'm doing in HSMM
>>hardware and software deals with a bit of the networking implementation,
>>not the apps side. I'm trying to concentrate at layers 1-4 of the ISO
>>model; you're generally talking 5-7. We both have things to do that
>>will impact each other. For instance, if you want a webcam app and
>>won't or can't throttle your camera, I'll do it for you with Quality of
>>Service/traffic shaping. I'll ramp its priority down to a low enough
>>level that we can ignore it 'til the bandwidth's available. Same thing
>>with using Microsoft network file sharing. I will throttle it down on
>>my networks to prevent gridlock. Microsoft's backwards method of
>>handling contention ("Gee, if I take over and get the file moved, it'll
>>be better for everyone, and I'll be happier, too") instead of being a
>>good citizen and sharing the link, is just the far side of intolerable.
>>
>>So. Brand me as uncooperative, if you must, because I don't like (or
>>use, willingly) M$ products. We don't have to use 'em, and if we can
>>identify a stable implementation of Linux, with applications, and lock
>>it down, we're good to go.
>>
>>73, gerry n5jxs
>> > How much of this can Microsoft Netmeeting handle?
>>
>>>If not MS Netmeeting, where's the software to do the job?
>>>
>>>Thanks for any help you can supply in my continuing education.
>>>
>>>bud N0IA
>>>
>>>---------------------------------------------------------------------
>>>"Are you still wasting your time with spam?...
>>>There is a solution!"
>>>
>>>Protected by GIANT Company's Spam Inspector
>>>The most powerful anti-spam software available.
>>>http://www.giantcompany.com
>>>
>>
>>
>>--
>>Gerry Creager -- [email protected]
>>Network Engineering -- AATLT, Texas A&M University
>>Cell: 979.229.5301 Office: 979.458.4020 FAX: 979.847.8578
>>Page: 979.228.0173
>>Office: 903A Eller Bldg, TAMU, College Station, TX 77843
>>
>
>
--
Gerry Creager -- [email protected]
Network Engineering -- AATLT, Texas A&M University
Cell: 979.229.5301 Office: 979.458.4020 FAX: 979.847.8578
Page: 979.228.0173
Office: 903A Eller Bldg, TAMU, College Station, TX 77843