On Thu, 20 Mar 2014 21:56:28 -0400, Danny Messano drmessano@gmail.com wrote:
(Please trim inclusions from previous messages) _______________________________________________ I've been little more than an observer for some time, but I must say ALL the points in this e-mail are right on. I began thinking about 44 almost 5 years ago, and to reiterate a few points made:
- Coordination at first seemed almost impossible. I was told by a few
other hams, "Good luck getting an address. 44-net is dead". I sent a number of emails over a span of 2 years before I got the attention of who I believe is the most recent coordinator for my area. I was greeted with a "You DO know that a /24 is 253 routable IPs?" After 3+ years of trying to get anything, I asked for a sizeable segment that I could break up and route as needed. I didn't want to wait 3 years for every /29 I needed for a new project. I'm sure the portal has made things easier, but with all the space available, why the grief over a /24? I didn't ask for a networking lecture, I know what a /24 is..
As a coordinator I've never hesitated when assigning a block of any size, including /24's to any ham who asked for one. Perhaps I am exceptional but I didn't see a reason to be stingy with addresses. In the beginning I used to assign individual addresses in 44.18.0.x in continuation of my predecessor's policy, since everyone was on 1200 baud on 2 meters and routing was manual but when it came time to assign for entities I assigned whole subnets and eventually designed a frequency/geography scheme for assigning subnets and if someone wanted a single IP, they got one from within that subnet. My biggest problem now is determining whether an address is even being used anymore and whether or not I need to care.
Brian originally reserved the middle two bits of the dotted quad to future expansion, resulting in a somewhat conservative approach in the minds of the sub-coordinators toward their net blocks. I believe that reservation no longer exists.
- Some of my experience with even getting addresses coordinated, along
with what Steve pointed out in the Terms and Conditions, as well as the overall complexity, I haven't found 44 to be very welcoming. If a club posts their repeater rules, and they consist of "Feel free to use the repeater. The tone is unpublished... you'll need to figure it out, here's a list of 20 things NOT to do, oh, and we're watching you!", I am probably not going to use the repeater.
This is unfortunate. I've never considered it my job to be a conservator of a scarce resource or to "regulate" what hams were doing with their address blocks. IP addresses are not repeaters and as a former repeater owner I don't have time to be a nanny and never did like being a nanny. But then again, I've heard some pretty foolish things done on repeaters that the "offending" ham should have known was against the regs and unlike IP address coordinators, the guy who's callsign is on the repeater can be held responsible for what people do with it. I also don't feel I need to teach TCP/IP protocol to someone just because he's new. I used to have a list of resources I'd pass out or just point them at the UCSD FTP site for all the code and docs.
YOU have to understand that if the environment seems to be unwelcoming it might just be because experimenters are experimenters and not necessarily good teachers, coders can be good coders and more often than not, they are lousy documentation writers. I've found some brilliant individuals who are doing some interesting things in their net blocks but they don't publish articles or how-to's, leading to a lot of hidden effort.
I challenge you to take a leadership role and publish your work and create some documentation and develop scripts and methods that people can use as templates so we can start making some forward progress instead of reinventing the wheel.
I have a strong networking background, and I am far from being an appliance operator, but it's hard to make a pizza if you've never seen one, the recipes are so vastly different that the concepts aren't clear, and it almost appears intentional.
Danny
What have you done to publish your recipes? How many new hams have you brought online in your community?
On Thu, Mar 20, 2014 at 7:45 PM, Steve Wright stevewrightnz@gmail.comwrote:
(Please trim inclusions from previous messages) _______________________________________________ Ok here's my opinion.
Technically, it's difficult for prospective members to connect a 44 subnet of any type, using any method. It is not clear at all how this is ACTUALLY done or what options are available.
The wiki should be the authoritative document, but ;
1.) The main page is all about how to edit the wiki and a logo competition, and ONE LINE on how to set up a gateway - which the whole reason people went to the wiki.
2.) The "Setting_up_a_gateway_on_Linux" wiki page has a broken link leading to "common instructions for setting up a gateway", inviting newcomers to consider that there ARE NO such instructions, at which point they'll probably completely give up.
3.) The three main options, munge script, rip44d.pl and rip44.c are not stated clearly, nor are there links to any such subsection, nor are these options grouped from the users' perspective - namely their chosen platform, be it JNOS, x86 Linux, OpenWRT, or METARouter.
4.) There's no real index to what people are actually DOING over the 44net, and people ARE DOING some cool stuff. If there were some page in the wiki where people shared what they were making, then others might duplicate their efforts.
Sysadmins on the portal are reluctant to issue /24s, when there's lots and lots available.
The portals' "Law and Jurisdiction" section in the terms and conditions insults the user. Most of the rest of that section is pretty unsavoury too.
WISPs and others who want to peer don't have access to any toolkit or support.
Some stuff in the portal doesn't (or didn't) work, and it's not clear which.
There's not really an apparent reason WHY newcomers might even WANT to number a network with 44. It's simpler to just throw a DHCP server at an interface and add some routing - easy peasy, why number the network with 44, and if they did - HOW to do that?
It's not really clear to network builders, that they can actually number up with 44 right now, and worry about connecting to other 44/xx Networks later when they're ready. If they want to expose several 44/24's to the wild internet, then that doesn't really affect anyone else but themselves.
All this tunnelling really is an unstable mess. Apart from allowing the wild internet to connect inbound, why not just route the whole thing?
HTH, Steve
Steve,
Wiki's are a collective effort, what have you done to fix the flaws you see there? Have you signed on as a Wiki editor? Have you written articles for inclusion on that Wiki? Links go stale, there has to be an "ugly bag of mostly water" behind the keyboard to keep a document tree fresh and healthy, otherwise off-site links go bad when someone drops dead or an organization folds. What are you doing in YOUR spare time.
I see a whole lot of room for improvement and a whole lot of networking experts who can "advance the state of the art" but who don't seem to be inclined to actually publish what they know.
What I see above from both of you is "this is a mess, someone needs to clean it up, but that someone isn't going to be me". I must boldly state that if you have the time to discern a problem and criticize a state of affairs, you have the time to take ownership of that problem and fix it.