Re: [Nanog-futures] NANOG-L, Paging, and the AUP

[michael.dillon]

> But since half this thread is suggesting such a post with 
> similar guidelines I guess means that people don't read them :)

I've never read it beyond the first few lines. I just assumed
it was one of those mailing list unsubscribe reminders.

May I suggest that the email message is trying to do too many
things in one message. You would do better to collect the
FAQ type info and post it separately with a subject line
reading "NANOG Frequently Asked Questions (FAQ)".

Also, the "contact" answer that you posted does not include
some good tips which were posted in this thread. It also 
doesn't ask people to post a follow-up telling how they
eventually suceeded in making contact. And there is no
feedback loop to ensure that the netops page gets updated.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] NANOG-L, Paging, and the AUP

[michael.dillon]

> I suppose what I'm saying here is that the presence of this 
> traffic on the nanog list isn't the problem per se -- it's a 
> symptom of the problem.

I agree.

By the way, email is not the only means of contact for
operational issues. There is also the telephone, but
the problem is that most effective numbers are unlisted
and those who know them really don't want to publish
them. INOC-DBA was supposed to solve this issue, but
I don't think enough people know about it. Or perhaps
it's number plan tie-in to ASnums tends to restrict
it to only people involved in peering.

http://www.pch.net/inoc-dba/

There are a number of issues surrounding operational
contacts, that I would suggest that the topic needs
a FAQ of its own. The email filtering/ticketing issues
are just one aspect of the problem.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] NANOG-L, Paging, and the AUP

[michael.dillon]

> With actual operational content decreasing, and people using 
> the list as a paging system for email admins and NOC contacts 
> more frequently, why has the AUP not been updated to reflect 
> what seems to be a "frowned upon" use of the list?

What is non-operational about finding operational contacts?

> There is a plethora of available resources to going about 
> finding such contact information, maybe not enough 
> subscribers know about, and it needs to be addressed further 
> in a FAQ on the NANOG website?

Try the word "dearth" not "plethora". The NANOG website has
always been rather spartan in terms of contact. Outside of
archives of the list and records of the meetings there has
never been a whole lot of content, and that other content
has historically not been updated very often. 

A NANOG FAQ would be a great idea, but it takes work, and
I don't see a whole lot of NANOG members who are willing
to do that kind of work. Note that an awful lot of content
does flow through the mailing list. For a few weeks, I constructed
several pages on ARIN's IPv6 wiki http://www.getipv6.info mainly
by picking out the useful bits of messages from the list and 
collecting it into articles. A FAQ editor could probably do
the same with items that truly are frequently asked questions
on the list.

> With that being said, is that a reasonable thought towards 
> updating the AUP, or is not adequate enough?

Is there evidence that a stick works better than a carrot?
More to the point, is there evidence that NANOG subscribers
will refer to a document on the NANOG website before posting
to the list?

> and What do we need to do to weed out the junk and promote "worthy"
> content again?

The answer is simple. You need to think like a gardener, and
promote the good stuff so that the weeds have no place to flourish
or are so shriveled that they are not noticable. I used to do
this when I had more time but now, the MLC fascists bark at me
for providing a link to the subscribe info for nanog-futures.
They would do better to chastise those who post in-group messages
that only hint at information, rather than spelling it out plainly
like many of my messages.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] Bhutan discovers the "NANOG Problem"...

[michael.dillon]

> Naturally we are always wondering on how to make the 
> conference more participative, exiting, and attractive to 
> more people. Ideas would be gratefully received by the SC and the PC.

Give everyone a one-sheet writeup of meeting etiquette, along
with their badge. This would contain suggestions such as

- leave the room to make phone calls
- don't punch the keys, type quietly
- put your phone on vibrate
- move to the back of the room when you know that
  you aren't terribly interested in the preso
- move to the front of the room if you intend
  to close your laptop and pay attention to the preso
- go into the hallway when you want to have 
  a conversation with somebody
- the meeting is being broadcast out in the hallway
  so you don't have to miss anything
- yada yada yada

And then you could remind people between speakers
that it's time to change places. Quiet zone up
front, handcuffed to your laptop at the back,
and compulsive conversationalists to the hallway.

And give the speakers a similar "NANOG tips" sheet
that reminds them:

- you are being multicast on the net so don't assume
  the room is your only audience
- your preso is recorded for posterity so your audience
  could be rather larger than the room itself
- ignore the activity at the back of the room

As for the common culture vs. anonymity aspect of NANOG,
back in the day there was a communal lunch served to all.
What about bringing back a variation on this. First of
all make it optional because the power networkers don't
want to meet anyone new and will avoid it anyway. Then,
for those that sign up, randomly assign tables in the
lunchroom to break up cliques. Before you criticise this
and explain why it will not work, note that I am suggesting
an OPTIONAL thing that people sign up for, and pay extra
for. The true test is to offer it and see if there are
enough takers.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] Bhutan discovers the "NANOG Problem"...

[michael.dillon]

> If what you say is worthwhile and engaging, they'll listen. 
> The speaking style that works well at a business meeting is 
> going to leave a technical audience desiring another cup of 
> coffee, or the opportunity to catch up with a friend in the 
> row behind them. 

And it is all being transmitted live to an audience with whom
you can't make eye contact, but they are there. Not to mention
the recorded video which reaches an even larger audience.

> Let this be a vote for *no* on shutting off net access.

It's not really an either/or situation. There are ways to
have both, and, as we both pointed out, there are good reasons
why speakers should not assume that room full of laptop users
means their talk is being ignored. There is still an attentive
audience, either in the room following along on their laptop or
elsewhere watching the video today or tomorrow or next year.
For instance, look at
.
Even if you presented 8 years
ago, there is a good chance that people are still watching your
presentation and reading your slides.

NANOG has always been more than merely a conference.

--Michael Dillon
 



___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] Bhutan discovers the "NANOG Problem"...

[michael.dillon]

> Having said that, providing its just the main plenary then 
> sure give it a go - why not try a 1/2 day in the next nanog 
> and then collect the feedback after to see how it went.

Essentially, this suggestion is just one form of a blended
solution to the issue, i.e. a blend of connectivity and 
non-connectivity. There are other ways to do this which
should be considered.

One possible approach is to shut off connectivity during
panel discussions in the plenary but have it on at other
times. The goal here is to achieve more participation.

Another approach is to arrange wireless coverage so that there
is a dead-zone at the front of the room. People who want to listen
will sit further forward, people who want connectivity will sit
at the back of the room. The stage area can be supplied with
wired connectivity for those who need it.

And a third approach is to have no wireless or wired connectivity
at all in the plenary, except for wired access to the stage. But
set up a separate room with full audio-visuals, i.e. big screen
video feed with good speakers to serve the whole room. Put a
microphone in this room as well, and give it Internet access
bothe wired and wireless. People can still participate fully
but the noise of keyboards and chatting is contained. Some of
you may recognize this as the "terminal room" of the mid 1990s.

Before making a decision on this it would be a good idea to 
propose several variations and then solicit input from people
who have experience at other conferences, as well as talking
to the people who actually setup and operate the audiovisual
and wireless services. For instance is it really feasible to
have a deadzone at the front of the plenary? Does an audio-visual
feed with microphones in a separate room complicate the AV
operations too much?

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] [Outages] Outages have an Outage? (fwd)

[michael.dillon]

> > That's the reason we need list moderators, to CULTIVATE the 
> list and 
> > encourage more signal in the postings.
> 
> I don't know about "moderators", but I do think list admins 
> are appropriate.  You want someone to watch the list and 
> curtail inappropriate behavior (and postings), and not really 
> to moderate the posts.  Moderation suggests some control over 
> what gets posted to the mailing list, and unless you make the 
> list fully moderated, that is impossible.
> 
> Unfortunately, one of the large disadvantages of an 
> unmoderated mailing has versus a moderated mailing list or 
> forum is the inability to retroactively remove posts that 
> have been made that were not on-topic, or otherwise not 
> appropriate.  That said, I'd rather have an unmoderated 
> mailing list over either of the other options.  The 
> information exchange is so much more important than the 
> ridged control or inappropriate behavior.

Yes, I agree that we want an unmoderated list. But we do need
to have some "people" called something or other, who will 
cultivat the list and encourage more signal in the postings.

For example, if the first offender was privately reminded that
their posting about outages was too cryptic and privately asked
to follow up with a URL etc., then they likely would do that,
with an "oops, I forgot" kind of apology. Then the "somthing
or other" person could follow up to the whole thread thanking
the "oops" person for following up, and all would see that
cryptic posts are not the NANOG way.

--Michael Dillon

P.S. the outages example was just that, a handy example. If only Gadi
had posted his helpful reference to the main list...

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] [Outages] Outages have an Outage? (fwd)

[michael.dillon]

> I've been seeing two kinds of discussion on the NANOG mailing 
> list about the Outages mailing list. 

So have I.

Strangely enough, none of those people seem to be able to document where
the outages list is located and how to get on the list, except in the
most oblique way. Imagine someone doing a web search for an outages list
who gets dozens of hits from the NANOG list archives but no real info. 

That's the reason we need list moderators, to CULTIVATE the list and
encourage more signal in the postings.

--Michael Dillon


___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] thought regarding meeting-locale chatter on theNANOG list

[michael.dillon]

> Just an idea from the peanut gallery. [Along similar lines, 
> it seems a shame that the meeting wiki effort we kicked off 
> for NANOG 39 seems not to have continued as far as this 
> meeting. At least, I couldn't find a NANOG 43 topic.]

That's because all the real action is on Multiply
http://nanog.multiply.com/

-- Michael Dillon I

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] MLC post-mortem]

[michael.dillon]

> I don't think that is a very good idea.  Blogs tend to be 
> good for the blogging crowd, which is probably similar to the 
> forum crowd, but is different from the mailing list crowd.  
> They may be overlap between the three groups, but you will 
> definitely loose a large number of participants by moving 
> from the mailing list.

If you go back to the message which I replied to, he was suggesting
that on a list as big as NANOG, there are too many participants.
Reducing the number of participants might be a good thing. In any
case the number of spectators should not change because the list 
will still be distributed as it always has. 

>   Honestly, I do not think that there really is a need to fix 
> it, as it is not really broken.

On the other hand, this is the nanog-futures list and it doesn't hurt
to kick some ideas around. Sometimes the end result is that you realize
the brokenness isn't as bad as it seemed.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] MLC post-mortem]

[michael.dillon]

> I'll say it You, M Dillon, are out of touch with what NANOG is.
> NANOG is not a mailinglist software development 
> company/group, nor should it be developing it's own custom 
> one-off of off-the-shelf products.

Which is why we use Mailman to run our mailing lists and which is why
I suggested that we should pick some existing blog software to run
our blog/list combo. However, the fact is that if we want to solve the
scaling problems of 10,000 plus list members by moving to blogging 
software, we also have to find a way to maintain the email delivery
of the lists. That is the place where somebody needs to do some
minor development work because, unfortunately, there isn't a nice
off-the-shelf package with a GUI control panel and a checkbox for
this function.

Perhaps you are not that familiar with the world of Open Source, but
the software that I referred to with a URL is a selection of
off-the-shelf
blog servers. No custom development needed, unless you want to do
something
a bit different. And since many of these tools either have an explicit
plugin architecture (PyBlosxom) or are frameworks designed to be
modified
(Django-based blogs) the scope of the custom development is nicely
constrained.
I expect that it would not require much more than an additional database

field to record whether or not the posting has already been emailed,
along
with some code to periodically run through new postings and send them
over
to the list processor (Mailman).

Besides, once a community gets to the level of having 10,000 committed
participants/observers, it really is financially viable to formally
become a non-profit organization and own assets. Considering how many
other NOGs there are in other regions, I wonder how long it will be
before someone tries to organize them all in a mutual-benefit society 
of some sort.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] MLC post-mortem]

[michael.dillon]

> But, I don't think any reasonable amount of enforcement is 
> going to solve what I perceive as the biggest cause of the 
> current low signal to noise ratio on the mailing list.  At 
> ~10,000 people with posting access, the list is just too big, 
> and there are too many views among those subscribers about 
> what the list should be.  A tiny percentage of people who 
> just want to talk about random stuff is still a whole lot of 
> people, who can make a lot of noise.  Groups of people who 
> think the list exists to discuss their area of expertise can 
> quickly make enough noise to thoroughly annoy people with 
> different specialties, who think the list is for their 
> subject area instead.  I really hope that as part of the 
> mailing list repair process, some further thought will be 
> given to breaking it up into smaller lists for discussions of 
> different subjects.  That way we could have some clear 
> guidelines on what is and isn't acceptable to post to each 
> list, rather than getting into endless arguments about the 
> purpose of the One True NANOG List.

Personally, I think that all posting access to the list should be
removed. Yes, nobody should be able to post messages to the list.
Instead, posting would be via a blog website, and the blog messages
would then be echoed into various NANOG mailing lists based on
which blog category was being used. Subscribers could then construct
their own personal NANOG list by subscribing to the blog categories 
that they are interested in. Threading would work very well since 
on a blog, you have to identify the thread before you post.

There are literally hundreds of blogging software packages out
there, many of them blog construction tool kits with dozens
of plugins available. Given the fact that the list runs on Mailman
which is written in Python, I suggest that folks have a look at
this list of Python blogging software:
http://wiki.python.org/moin/PythonBlogSoftware
My sense is that PyBlosxom and the Django-based stuff are the
more popular. It's too late to do this as a Summer Of Code project
but the same principal could be applied, i.e. find a university
student, put them together with a technical mentor, and give them
access to the server needed to test and trial the blog.

If the SC isn't interested in testing this right now, then I suggest
that the new MLC give serious consideration to taking this on as 
a project.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] MLC post-mortem

[michael.dillon]

> there is too much reliance on process and trying 
> to document and plan in advance for every eventuality rather 
> than the good judgement of the MLC (which hopefully is why 
> they were appointed in the first place!).

And too much secrecy. In the normal state of affairs where 
the MLC is acting as a list moderator, all the communication
should go to the list itself. After all, as RS stated, most
of the communication is just advice about what is and is not
appropriate.

> I'd be less than forthright to not point out that the job is 
> kind of thankless, but why don't you throw your hat into the 
> ring anyway

I think that if the bulk of the work is done in the open, then
it isn't so thankless.

Maybe it's time to change the structure of the NANOG organization
and farm out the MLC enforcement work to the SC. It should only be
necessary in exceptional cases therefore should not place undue load
on the SC. That way the MLC becomes the LMC (List Moderation Committee).

Nash drug, Alex, has been doing a great job on the stuff that has been
visible, such as the thread summaries. The SC really should invite him
to form the new LMC and continue the work of gardening the list and
increasing the signal. NANOG still has an overall positive reputation
out in the world, which causes people to recommend it to friends and
coworkers. 

--Michael Dillon


___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] [NANOG] [NANOG-announce] Mail List Committeeannouncement

[michael.dillon]

> Trust me when I say that some of the things being proposed as 
> list policy behind the scenes was quite ridiculous.

Sorry, but I don't trust you. In this statement you seem to be
condoning "in camera" discussions about what is supposed to be
a public mailing list that is supposed to be run by and for the
community. Why are all these discussions so secret? 

I'll make my own judgements about what is and is not ridiculous.
And I'm not foolish enough to mistake brainstorming ideas for
serious proposals.

> I second Mr. Seastrom's calls for the 
> SC to do the 
> honorable thing and commit seppuku.

Can't say I disagree about this. There is too darn much arguing
going on about something that simply should not ever descend into
this level of discord. It's a conference and a mailing list.

> All I can offer as an observance is that there are multiple 
> commercial conflicts with members of the SC, myself, other 
> members, and the community. Creating a political atmosphere 
> in NANOG is not working because of the commercial conflicts.

Why should there be a political atmosphere? It's a conference
and a mailing list.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] reply-to

[michael.dillon]

> If the NANOG list sprouted a reply-to header, I'd be very unhappy.

To echo an earlier comment, how much development work would be
involved in allowing list members to individually specify whether
they want:

Tags in the subject line,
Additional message footers,
Reply-to headers

Anything else?

Yes, I realize that this might require some changes to a batched
emailing process, but in these days of big bandwidth, powerful CPUs and
a SPAM flood dwarfing real email, I would say it is a reasonable request
to make this all possible.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] Subject line Tag and footer

[michael.dillon]

> Having run several mailing lists in the past and present, you 
> would be amazed at how many people don't know what what 
> mailing list they are on, and how to unsubscribe.  Most 
> people don't have a clue how to check headers.  Even quite a 
> few of router experts don't have a clue about email headers.  
> If you must get rid of the subject tag, I would really 
> recommend that the footer stays.

I agree.
This list is archived on several websites. It is useful to have
the URL to the subscribe info on the last line of every message.
However, we could lose the underscores, the list name and even
the list address.

The URL is even of use to current subscribers because it is the
quickest way to find the mailman web page to change settings.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] IPv6 or not?

[michael.dillon]

> 2) offer constructive feedback about what you would rather 
> see (bonus points for who might present it);

On the ARIN IPv6 wiki here

there is an article from April 2nd, that tells a bit about the work
various
carriers are doing to get IPv6 into production. It might be interesting
to
contact these companies and get one or more of them to present on the
problems
that they are running into in their test labs. I suspect that they will
tend
to simply refer such problems to vendors, but since IPv6 is all about
interoperability
it would be useful to give these problems a more public airing.

There has been a tendency for people to try out IPv6, find some
stumbling block, 
then put it on the backburner without publicising what the problem is.
Since we
are now at the stage where 99% of us have to deploy IPv6 commercially,
whether we
like it or not, it is important to air the specifics of these stumbling
blocks.

The 99% figure is based on assuming that pure IPv4 network services will
soon
be a niche market, and the majority will have to have at least some IPv6
services.
And I said that we have to deploy whether we like it or not, because I
am assuming
that IPv4 address space will be exhausted in two to three years which
seems to me
to be the minimum time needed to test and plan and deploy it as a fully
functional
Internet access service.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] Opinions requested: NANOG-worthy or not?

[michael.dillon]

> Of course, given how pervasively evil NetSol has 
> turned out to be, it's getting increasingly difficult to keep 
> track of all they ways they've attempted to turn our Internet 
> into their personal profit center.

"Our Internet"!?
You sent your email from gsp.org which is in a TLD managed by PIR, not
Netsol. And you replied to an email from labrats.us which is in a TLD
managed by NeuStar, not Netsol.

I wonder what relevance this has to nanog-futures? Maybe we need a
workshop on how to transition your domains away from the commodity .com
TLD?

Or maybe we need to petition ICANN for a .tube TLD for all of us
Intertube operators...

--Michael Dillon


___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] default routes question or any way to do the rebundant

[michael.dillon]

> Since when does "the sharing of knowledge for the Internet 
> operations community" extend to "home user questions"?

Let's start by refreshing everyone's memory:

   ls it possible to have 2 default routes?
   or how can I do the rebundant when the route is still working either
eth1 or eth2 down?

   [[technical details omitted]]

   i am using freebsd router

That is the question. Nowhere does the person say that they are a home
user. The technical details that I omitted show that the person has some
low-level technical understanding, i.e. mention of ping, diagram of
network fragment.

It seems to me that this person commits only two crimes. 
First, English is not their native language, which means
that people have to stop and think a bit in order to 
understand the question. And secondly, this person does 
not use the status-symbol brand of router that many
chest-thumpers on the list use. Therefore this person
is an OUTSIDER who must be chased away to preserve the
integrity of the tribe.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] default routes question or any way to do the rebundant

[michael.dillon]

> For the sake of discussion, let's come up with some semblance 
> of boundaries so we know whether we'll get scolded for 
> leading a guided tour or not.  If someone's rude, we can 
> handle that separately.

One strategy that works well is to make a habit of
leading a guided tour and just ignore those who scold 
you. Basically, if you don't have the time to write
a proper guided tour reply, complete with tested and
working URLs, then don't bother answering the question.

For an example of a good way to reply to seemingly naïve
beginner questions, look at what Paul Vixie wrote.

There was a time when NANOG was part of an activity which
was spreading the good news of the Internet to the whole
world, and educating people in other countries how to 
implement Internet technology in their own environment.
That NANOG didn't sneer at small networks, rather it helped
the small operators implement their networks using best practices
so that their networks ran smoothly and they had a chance at
scaling up relatively smoothly.

Funny thing is that if you read the first sentence in section 3 (Mission)
of the NANOG charter, it seems to include that goal. 

   The purpose of NANOG is to provide forums in the North
   American region for education and the sharing of knowledge
   for the Internet operations community. 

--Michael Dillon

P.S. does everyone here still read "Internet" as "global public network"
or has the popular meaning shifted to "American public network"?

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] default routes question or any way to do the rebundant

[michael.dillon]

> We were all beginners at once, and limiting the participants 
> to the 5 or
> 6 North American networks that would fit into the category of 
> large would just be completely pointless, and a sure thing to 
> signal the end. 

NANOG has always been a forum *IN* North America *FOR* all people
worldwide who work in Internet network operations. If you look at the
first sentence of section 3 (Mission) of the NANOG charter, it reads
similarly.

> But, lets be serious.  Most of the large networks probably 
> don't care about NANOG.  They have smart people, and access 
> to smarter people, and have very large pull with the vendors. 
>  I doubt that they are going to learn much here that they 
> don't already know.  NANOG is really for everyone EXCEPT the 
> large networks.  It's for the people that are still learning, 
> and companies that are still growing, small to medium size networks.

Which is why it is still appropriate to revisit beginner topics
from time to time. It also helps the archives because people are
less likely to read, or trust, older messages.

Of course, we could build up a NANOG FAQ, but maybe that is too
much like work.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] [BULK] Re: Mailing list procedures for reviewby the NANOG community

[michael.dillon]

> > MLC is not the list police - we expect
> > community to self-moderate
> >
> >  and:
> >
> > Guiding the conversation without stifling
> > the discussion
> >
> >  and:
> >
> > New initiatives
> > - Thread summaries - time-intensive
> >
> >  But your document about MLC procedures doesn't seem to 
> cover  those 
> > things but focuses on POLICING activities.

> This isn't my document. It's yours.

If part of the MLC's task is to guide the conversation, provide
thread summaries, and other positive cultivating actions, then
why not write that down in the MLC procedures guide.

If the MLC procedures are the property of the community,
as you have just said, then why not write that down too.

Is the MLC fundamentally, the list police?
Or are they the list shepherds/gardeners?

A list policeman issues warnings, but a list gardener would
ask the miscreant to write a thread summary in lieu of a 
warning, thus turning the situation around. Think Lao-Tzu,
not Machiavelli.

I realize that the MLC doesn't issue a lot of warnings, but that
is partly because they don't want to feel like policemen. Perhaps
they would intervene more often, if they had some carrot-like
tools in their arsenal as well as the big stick.

Of course, if your only goal is to document the status quo, then
don't listen to me. I always view these things as an opportunity
to improve things and move beyond the status quo so I'm just 
raining on your party.

--Michael Dillon

P.S. one way to write a DRAFT document, is to start with a brain 
dump of everything one could possibly think of including, no matter
how silly. Rather like brainstorming. I think that the NANOG steering
group situation is fluid enough at present, that this would be a
good way to approach documenting the MLC procedures. But brainstorming
works better with a wiki than with a PDF document. 

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] [BULK] Re: Mailing list procedures for review by the NANOG community

[michael.dillon]

> http://www.fugawi.net/~hannigan/nanog-mlcp1-1.pdf
> http://www.nanog.org/mtg-0802/presentations/community-pilosov.pdf
> 
> If you compare those, I think that's night and day, don't 
> you? 

Yes, I agree. If you look through Alex's presentation it says
things like:

MLC is not the list police - we expect
community to self-moderate

and:

Guiding the conversation without stifling
the discussion

and:

New initiatives
- Thread summaries - time-intensive

But your document about MLC procedures doesn't seem to cover
those things but focuses on POLICING activities.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] Mailing list procedures for review by the NANOGcommunity

[michael.dillon]

> http://www.fugawi.net/~hannigan/nanog-mlcp1-1.pdf
> 
> Please reply here or privately.

Law Enforcement and DMCA Designated Agent
NANOG Mailing List Committee
c/o Merit Network, Inc.
1000 Oakbrook Drive
Suite 200
Ann Arbor, Michigan 48104-6794
Telephone: (734)764-9430
Email: [EMAIL PROTECTED]

This is just plain weird. Is this some kind of legal requirement in the
USA? Has anyone ever posted any copyrighted material to the NANOG list?
Given the short size of email messages, wouldn't any copyrighted
material be considered to be "fair use" quotations?

Since the MLC consists of several individuals, only one of whom works at
Merit, this seems to be making Sue Joiner the official legal contact for
the mailing list. Is that her phone number?

The extra paragraph added to each AUP item seems rather forced, often
restating the obvious with slightly different words. Is there an issue
with understanding of AUP items? If so, then perhaps it would be worth
addressing those specific misunderstandings. If not, then why add the
extra words? I wonder what is the point of saying that NANOG is neutral
in matters of politics.

It wouldn't hurt to clarify the AUP with specific examples but the added
text in this document is wishy washy and detracts from the short punchy
nature of the AUP itself.

On the other hand, a statement like:
   No Differentiation
   The MLC will consider all policies equal.
isn't punchy at all. It's just meaningless and I wonder why it is in the
MLC processes.

For that matter, given that this is an MLC operational procedure
document, why is it repeating the AUP at all?

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] The Peering BOF and the Fallout?

[michael.dillon]

> I did swing by Radio Shack. It can be done, but then I 
> thought about it and the professional queue system was < 
> $1500. I think that Merit should make an investment in it to 
> improve the conference and speaking experience. It would be 
> well worth it in terms of making things run smoother.

$1500!?

Go to ebay and pick up an old PocketPC for a tenth of that 
and install this speech timer freeware:


Or get an old laptop (386 should do fine) and set it up
to flash cues to the speakers.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] Objection: RE: [admin] Re: EU Official: IP Is Personal

[michael.dillon]

Some typical Hannigan posturing below. Joe Greco posts a long opinion 
piece about privacy policy, IN DISAGREEMENT WITH MY MESSAGE, and 
Hannigan launches into a diatribe against me.

For the record, I *HAVE* referred to the mailinglist AUP which is *NOT*
at the URL referred to by Hannigan. The 1st line of the AUP says:

  Discussion will focus on Internet operational and technical 
  issues as described in the charter of NANOG

To repeat my objection, the first line of the charter says:

  The purpose of NANOG is to provide forums in the North American
  region for education and the sharing of knowledge for the
  Internet operations community.

Therefore, Hannigan's message to the list is OUT OF ORDER since
it is telling list members that European-oriented discussion is
off topic. For the record, I received some private messages thanking
me for posting my objection.

Sadly, it seems that nobody much cares about NANOG any more. Nothing 
is discussed on the futures list. There is no community any more except
for the people who use NANOG as an element of their social calendar.

> -Original Message-
> From: Martin Hannigan [mailto:[EMAIL PROTECTED] 
> Sent: 28 January 2008 23:00
> To: Dillon,M,Michael,DMK R
> Cc: nanog-admin
> Subject: Re: Objection: RE: [admin] Re: EU Official: IP Is Personal
> 
> Michael:
> 
> Please refer to the (mostly) community accepted NANOG AUP as 
> the document that mandates how we manage the list and what is 
> and isn't on
> topic:
> 
>   http://www.nanog.org/mailinglist.html
> 
> I would urge you to post on topic on the nanog list where 
> applicable or risk being formally warned to do so or even 
> banned from posting to the NANOG list as prescribed.
> 
> You should feel free to discuss charter/AUP changes or issues 
> in nanog-futures.
> 
> Martin Hannigan
> NANOG MLC Member
> 
> 
> 
> 
> On Jan 28, 2008 5:28 AM, Joe Greco <[EMAIL PROTECTED]> wrote:
> >
> > > > Folks, we'd like to ask that this thread die a quick 
> and painful 
> > > > death. It's gone off topic and it seems to have run 
> whatever short 
> > > > course that it tried.
> > >
> > > I agree.
> > >
> > > > While what Europe does
> > > > is interesting to us as network operators, this is 
> European policy 
> > > > and off topic for NANOG.
> > >
> > > Whoa there! You need to re-read the first line of the 
> NANOG mission 
> > > statement
> > >
> > >   The purpose of NANOG is to provide forums in the North American
> > >   region for education and the sharing of knowledge for the
> > >   Internet operations community.
> > >
> > > In other words, the NA part of NANOG refers to the 
> location of the 
> > > forums, *NOT* the scope of the discussions. The Internet 
> operations 
> > > community is global in scope and it is natural for our 
> discussions 
> > > to also be global in scope.
> > >
> > > Since many North American network operators have 
> infrastructure in 
> > > Europe (PoPs, colocated servers) they have to be aware of 
> uniquely 
> > > European Internet issues.
> > >
> > > And when it comes to solving a domestic problem, nothing 
> puts things 
> > > in perspective more than comparing how others approach 
> the problem.
> >
> > I had already commented in some other private messages something to 
> > the effect that whether or not this was "operational" 
> depended largely 
> > upon the ethics of and legal requirements imposed upon 
> network operators.
> >
> > In the same manner that most butchers care little for the manner in 
> > which their products are handled while alive, as long as the meat's 
> > good when it gets here, many network operators care little for the 
> > implications of these sorts of things on the privacy of 
> their users, 
> > so long as the users keep paying the bills.  This is related to the 
> > attitude which got the telcos into the warrantless 
> wiretapping problem ...
> >
> > However, for those of us who actually care (either because we feel 
> > morally/ethically bound, or because we are legally obligated due to 
> > local politics), it doesn't even have to involve operations in the 
> > European theater.
> >
> > Privacy policy may not at first appear to be directly 
> related to the 
> > art of entering an enable password, but for some network 
> operators, it 
> > will involve questions such as "do we provide a way for a user to 
> > become more anonymous," which could include things ranging from 
> > esoteric stuff such as running a Tor proxy and SOCKS 
> gateway (we do), 
> > to more mundane things, such as DHCP lease strategies, PTR 
> > assignments, etc.  These are all issues relating to the design and 
> > operation of your network - and I've only touched on a few issues.
> >
> > Respecting the privacy of your customers *ought* to be a BCP, and 
> > *ought* to be relevant.  Whether or not this particular discussion 
> > falls under a reasonable level of relevance remains to be 
> seen, and is 
> > not the point of my message to you.  Some of it clearly 

Re: [Nanog-futures] What should NANOG be?

[michael.dillon]

> Why can't the NANOG community identify and agree to adopt 
> some BCPs that the group generally feel are good ideas?

Why? I think the reason why NANOG cannot do this is that
this kind of work is better done outside of a mailing list
and the NANOG community has not shown much interest in
doing it. NANOG folk set up a blog but virtually nobody
contributed. There was (and still is) a best practices wiki
out there but not much in the way of contribs. There was
another wiki that was set up as part of the NANOG-FUTURES
work, but that quickly fizzled.

> For example, as was discussed on the nanog list, why not 
> identify a BCP for identifying the appropriate contact at an 
> organization?

If you can't get a significant level of participation in 
writing such BCPs, then people won't buy into them and
won't follow them.

> The first step towards making things better is to identify 
> things that are broken and *a* generally agreed upon good solution.

Agreed, but NANOG has repeatedly backed away from doing this
kind of work.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

[Nanog-futures] Dean is watching you! was: Status of Investigations

[michael.dillon]

I though y'all might want a heads up to Dean Anderson's latest
initiative. Although his investigation, at present, is focused on ARIN,
it might spread to NANOG, MERIT, etc.

For more details go here
http://lists.arin.net/pipermail/arin-discuss/2007-December/author.html
and read the three messages by Dean, especially the one posted at 6:20.

> -Original Message-
> From: Dean Anderson [mailto:[EMAIL PROTECTED] 
> Sent: 01 January 2008 19:41
> To: Dillon,M,Michael,DMK R
> Cc: [EMAIL PROTECTED]
> Subject: Re: [arin-discuss] Status of Investigations
> 
> [ARIN agrees that discussion to recall Board members, to 
> conduct investigations of ARIN expenditures, and to stop 
> certain ARIN expenditures does not violate the ARIN AUP.]
> 
> On Sun, 30 Dec 2007 [EMAIL PROTECTED] wrote:
> 
> > > 2. Truth is an absolute defense to claims of libel.
> > 
> > Not here it isn't! You have to go to a court of law in order to use 
> > the "truth" defense.
> 
> You have to go to court to hold someone accountable for libel. 
> 
> > On this list, if you constantly annoy us then we don't have to pay 
> > attention to you any more or read any of your messages.
> 
> On this list, ARIN is subject to US Law governing 
> corporations, and other US laws; There are laws governing 
> what corporations can do, and corporations cannot prevent 
> members from investigating the activities of a membership 
> corporation. Of course, as a member, you are free to read or 
> not read anything you wish.  You are also free to vote anyway 
> you wish.
> 
> > > Question: Why is ARIN paying to have the rest of these 
> people learn 
> > > to configure BGP and configure spam filters, and other technical 
> > > network operation tasks? This seems to be recoverable.
> > 
> > Question: If you haven't even got the faintest clue of what 
> NANOG is 
> > and what kind of things are discussed at NANOG, then why are you 
> > wasting our time with this nonsense?
> 
> Well, I have recently analyzed attendence at NANOG over the 
> last 10+ years. I have read the NANOG list for over 10 years, 
> and participated at times. I have worked in this industry for 
> 20 years. I have some idea of what NANOG does and what NANOG 
> doesn't do. But the best evidence is
> objective: Let's see what NANOG says about itself:
> 
> From NANOG's charter:
> 
>   "The purpose of NANOG is to provide forums in the North American
>   region for education and the sharing of knowledge for the 
> Internet operations
>   community.
> 
>   NANOG is a small venue in which technical matters 
> pertaining to network
>   operations and network technology deployment in Internet 
> providers may
>   be discussed among experts. "
> 
> 
> In fact, examining the attendence records of NANOG I found that:
> 
> 2 in 3 of the attendees dropped (never came back) after only 1 meeting
> 4 in 5 dropped after only 2 meetings
> 7 in 8 dropped after only 3 meetings
> 9 in 10 dropped after only 4 meetings
> 
>   NANOG 41 was the first meeting attended by 145 attendees 
> (33% of total).  Statistically, 100 or so will never return.
> 
>   NANOG 41 was attended by 79 people (18%) who had attended 
> more than 10 meetings. There are only 184 people in the world 
> who have attended more than 10 meetings. 
> 
> Indeed, the facts I found indicate that NANOG is essentially 
> a small club; if you are pals with about 184 or so core 
> members (those who have attended more than 10 meetings), you 
> come back repeatedly. Those who aren't, never come back.
> 
> 
> 
> Contrast the NANOG charter with the jobs of these ARIN employees:
> 
> 13|Ray PlzakExecutive President & CEO
> 17|Richard JimmersonExecutive Chief Information Officer
> 2|Nate DavisExecutive Chief Operations Officer
> 1|Therese ColosiHuman Resources Executive Assistant
> 1|Ray Stark Engineering Windows System Administrator
> 1|Abram Thielke Engineering Software Engineer
> 3|Tim Christensen   Engineering System Architect
> 1|Matt Rowley   Engineering Unix Systems Administrator
> 2|Ming Yan  Engineering Database Administrator
> 7|Michael O'Neill   Engineering Network Administrator
> 2|Cathy Murphy  Engineering Principal Software Engineer
> 1|Darren Kara   Engineering Database Administrator
> 6|Matt Ryanczak Engineering Systems Operations Manager
> 20|Leslie NobileRegistration Services Director of 
> Registration Services
> 4|David HubermanRegistration Services Technical Specialist
> 1|Jon WorleyRegistration Services Senior Resource Analyst
> 4|Cathy ClementsRegistration Services Principal 
> Resource Analyst
> 4|Jason Byrne   Member Services Membership Operations Manager
> 6|Susan Hamlin  Member Services Director of Member Services
> 4|Einar Bohlin  Member Services Policy Analyst
> 2|Erin Centanni Member Services Meeting Planner
> 2|Erika GoedrichMember Services Membership 

Re: [Nanog-futures] Cisco outage

[michael.dillon]

> The AUP does not apply in this case.  However, it does seem 
> that Marty (and
> whomever) are between a rock and a hard place with the 
> options left if private mail is not deliverable.

Only if Marty is Lord High Poobah of the mailing list. But
he is not. There are *FIVE* other MLC members who presumably
are not wholly dependent on gmail.com for their email. One
wonders why Marty could not have passed this case on to other
MLC members to deal with.

--Michael Dillon

Somebody needs to write a NANOG soap opera and put on
an evening performance at one of the meetings.


___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] Cisco outage

[michael.dillon]

> Should we remove this person or not?
> 
> ---begin
> 
> Delivery to the following recipient failed permanently:
> 
> [EMAIL PROTECTED]

Seems to me that the best way to solve this squabble is 
to punish both parties. I vote to unsubscribe all addresses
at infiltrated.net and gmail.com

--Michael Dillon


___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] [admin] RE: Creating a crystal clear and pureInternet

[michael.dillon]

> MLC was discussing blogs recently. I don't think I like your 
> idea - I doubt many people will post to blog, but frankly, as 
> long as on-list traffic becomes operational, I'm all for that!

The point is that one of the list features is a searchable web archive.
By changing the software used to provide the archive to some blogging
package, we won't subtract anything from what we have now. But we could
have a more open AUP for blog comments than for list postings. People
who feel more comfortable with newer Internet tools can subscribe to an
RSS feed from the blog rather than the mailing list. The gateway would
be one-way, i.e. no blog comments are fed back into the list. This would
mean that every list member needs to have a username (could be list
email address) and a password, but that is not unusual in the world of
mailing lists.

> > What we have here is a failure of the imagination. 
> (paraphrased from 
> > the
> > 911 commission report)
> Why don't you volunteer for MLC? (serious question).

Probably because of my perception that it is a police force,
not a team of gardeners. And because I am not one of the
Internet old-timers, i.e. I only joined NANOG in 1994 which
makes me one of the 3rd wave of newcomers. And I really don't
approve of the cliquish approach that dominates a lot of the
debate.

This reminds me of an incident heading to a NANOG meeting
in Tampa Florida. I was on the shuttle bus headed to the
hotel and some of the engineers sitting right near the driver
were proclaiming with pride that they were the people who 
actually run the Internet. At that time, it was probably 
true that the people on the NANOG list were the key players
who ran the Internet's networking infrastructure and I think
the cliquishness helped because it caused people to copy
best practices from each other and grow the network faster
than otherwise would have been possible.

However, I don't believe that it is true any longer that
NANOG members are the key players in running the Internet's
networking infrastructure. Vendors have a much bigger role
with their training and books and certifications. I think
it is a mistake for NANOG to try to recover past glories and
have such a narrow focus on so-called "Internet operations".

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

[Nanog-futures] Marching to a different drummer

[michael.dillon]

> I believe that many can contribute in 
> different aspects that might not have been visible or even a 
> notion to others. Some people tend to think outside of the 
> norm. 

Isn't this the whole point of NANOG? It started with getting
vendors, researchers and operators all together in the same
place. By their very nature these three groups see the world
differently and that is where the strength of NANOG came from
and that is what helped the Internet achieve exponential growth
and elbow aside the Information Highway concept.

If we think like gardeners trying to nurture NANOG, wouldn't we
actively encourage such contributions, and *NOT* try to force
everyone to conform to the cliquish ideal of the American Internet
Operator? NANOG is not a clique, not an old boys club, not a
secret society. It may sometimes have elements of all of these
but that doesn't mean there isn't place for topics which fall
outside the interests of the insiders.

> I also don't expect someone to get bent out of shape and 
> attempt to throw in an aura of "entitlement" because they 
> believe they're higher than someone else.

This is the fundamental problem with what Hannigan has been
doing recently. I also got a message from him, and like
J. Oquendo, I also misunderstood the tone of the message t
be an official warning. Also like J., I had not been active
on the list for a time, in my case 8 days, so the warning
seemed like an unfair bolt out of the blue. Having seen that
I am not the only recipient of Hannigan's bluster, I suspect
that there is a pattern here and others have also received
them. Like J. I don't agree that these messages are appropriate.

In the past, I have occasionally received personal messages
from the MLC, mostly Rob Seastrom but perhaps others as well.
The tone of those messages was worlds apart from Hannigan's
style and I never misunderstood them as official warnings and
always took them on board and dropped threads or whatever.
Given the fact that email is notorious for generating misunderstandings
about the tone of a message, I don't understand why the MLC
doesn't use some kind of form for any message that they send
in an official capacity. Even if it is a gentle nudge to
end a thread or steer back on topic, it could still be done 
with a template message.

I'm beginning to think that an AUP and a charter are simply
the wrong way to manage the content on a mailing list. 
A basic AUP is needed to establish what are the norms of
behavior, but when it comes to topicality, there has got
to be a better way to explain what is fodder for the list
and what should be discussed elsewhere. 

--Michael Dillon



___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] [admin] RE: Creating a crystal clear and pureInternet

[michael.dillon]

> Should politics/culture/society be on-topic? Or should we 
> maintain this list as *internet operations* list.

What is *internet operations*? 

In the telco companies that operate most of the Internet, operations
does not include network design, choice of hardware, etc. It does
include things like Trouble-to-Resolve, Service Management and the
provisioning parts of the Lead-to-Cash processes. In these companies
network designer who are not in operations, make decisions about the BGP
architecture (communities, confederations, peering, etc).

I always thought that NANOG's remit was broader than that, so if the
term "internet operations" does not fit, and we want to have a list
where people know and understand the AUP and rarely violate it, then we
need to have a much clearer definition of things. It is not good enough
to say that the MLC members understand it. Everybody needs to understand
it.

The original charter and AUP, which I had a hand in wordsmithing, were
created way back when the Internet was run by ISPs, small
entrepreneurial outfits in which people wore several hats. Some of those
outfits were companies, some were embedded in universities or telcos or
other large companies like IBM. NANOG tradition has been to have
discussion that wandered over many areas analogous to the way a job
description in an entrepreneurial outfit tends to wander over many areas
of human endeavour.

Why can't we be more open about this and discuss things like the
definition of the terms we use? How can we allow discussion to be
reasonably broad as long as it is relevant and doesn't overcrowd other
issues? Why can't we be more creative in the use of technology and do
something like copy all message threads to a blog and have the
moderators cut off wandering threads on the list while allowing the
discussion to continue on the blog?

What we have here is a failure of the imagination. (paraphrased from the
911 commission report)

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] [BULK] Re: Fascist police force or team of gardeners?

[michael.dillon]

> > Will NANOG become a better list with a team of fascist 
> police officers 
> > scrutinizing every posting?
> > http://www.infiltrated.net/nanogpolice.jpg
> 
> clearly not but lets begin with the evidence :)

That was one of the reasons why I posted my message. I know that these
messages are being sent to various individuals and Hannigan himself
noted that there is some discussion amongst the MLC about them, but
there has been nothing openly discussed here.

Seems to me that when changes are made, the outcomes deserve to be
discussed here as well.

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

Re: [Nanog-futures] Creating a crystal clear and pure Internet

[michael.dillon]

> ITU anti-botnet initiative
> 
> http://www.itu.int/ITU-D/cyb/cybersecurity/projects/botnet.html

If nothing else, pick up a copy of this document
http://www.itu.int/ITU-D/cyb/cybersecurity/docs/itu-botnet-mitigation-to
olkit-background.pdf
and read starting at page 10 entitled "Lack of Coordination among
Stakeholders". Is NANOG just another silo that fails in the objective of
being a forum for Internet operators?

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

[Nanog-futures] Fascist police force or team of gardeners?

[michael.dillon]

Looking at the recent drop in list traffic which coincides with Martin
Hannigan's campaign of stern warning letters to various people, I wonder
whether things are heading in the right direction.

Will NANOG become a better list with a team of fascist police officers
scrutinizing every posting?
http://www.infiltrated.net/nanogpolice.jpg

Or would it flourish better with a team of gardeners intent on
cultivating a healthy discussion?

I remember a session that I attended back at ONE ISPCON in 1996, the
first ISP conference that Jack Rickard operated. There was someone there
from Prodigy, a very successful social networking blogging service
although they didn't call it that. They had many forums, each one
focussed on some area of interest, and people participated by reading
and posting in these forums. Each forum was managed by one or more
sysops (what we would call moderators today) and the site's revenue
depended on having healthy traffic levels. Prodigy put a lot of effort
into building up user participation at a time when the huge buzz and
marketing push of the Internet did not yet exist.

The ONE ISPCON presenter was telling us their secret of success and that
was the fact that they CULTIVATED sysops and those sysops CULTIVATED
users. They actively attempted to ensure that there was a sufficient
volume of relevant discussion as well as blocking abusive users. In
other words, they approached their forums as a team of gardeners.

I think that the current NANOG SC et al. are unfortunately headed down
the path of policing which may very well reduce the weeds in the garden,
but will not necessarily increase the garden's yield. When you damage
the roots of your crop by forcibly yanking out weeds, then yield
suffers. When you make it hard for weeds to grow by nipping them in the
bud early, and gently, then yield prospers. Note that the NANOG Program
Committee does seem to take the approach of cultivating in managing the
NANOG meetings. This makes the contrast with the list all the more
stark.

Again, I don't have any answers. I'm just asking questions.

But I'm curious whether anyone has considered this analogy as an
approach to the issue, or whether anyone has looked to other historical
reference points when developing the plans for managing the NANOG list. 

--Michael Dillon

___
Nanog-futures mailing list
Nanog-futures@nanog.org
http://mailman.nanog.org/mailman/listinfo/nanog-futures

RE: [admin] Errors to NANOG list subscribers take II

[michael.dillon]

> Given the serious impact this is having on operations, does 
> this have a master ticket number or escalation id of some 
> type? Has the vendor been involved yet? When can we expect to 
> see a post mortem/RFO?

Even though this suggestion was made with tongue-in-cheek,
it would still be a good idea to see a post-mortem
Reason For Outage report to the -futures list.

--Michael Dillon

RE: Jared's network operator contact list

[michael.dillon]

> (What potential motivations might Jared or PCH have in renaming?)

Renaming? No good reason that I can see.

> I'm not sure I understand the rationale behind your suggestions.

NANOG as a conference and mailing list was running just fine when there
was no governance structure in place. Some complaints about unfairness,
but this did not impact most people.

Now there is a governance structure in place and many people have put a
lot of effort into that. If this is a GOOD governance structure for the
inter-Netops community, then it seems to me that NANOG should
investigate ways in which they can support and improve the work of other
groups. 

Case in point is that Jared's site and routeviews have suffered from
outages. That is to be expected with a labor of love and says nothing
bad about the maintainers of these sites. But if these people and groups
would hand over their projects to NANOG, then there should be enough
critical mass to cover the cost of making sure that the sites don't go
down. And if there is worry that NANOG doesn't have the money to do
this, I'll note that one of the teams that I mentioned, cymru.com, have
managed to cross that hurdle. 

My suggestion is all about synergies, critical mass, and opening a
discussion with the people and groups working in this space. No doubt
they will have their own ideas about how to work with NANOG, so please
consider my suggestions as just the opening ideas for a discussion.
Would some sort of alliance with NANOG raise the profile of all of these
projects and make them more useful, and more valued in the community?
Would having a governance structure for these projects lead to more
technical improvements and more new useful projects?

In other words, I have no answers at all, merely some questions that I
think are worth asking.

--Michael Dillon

Oh yes, and one more

[michael.dillon]

And what about routeviews.org?

By pooling resources of all the small teams doing work in the
inter-Netop operations area, NANOG could accomplish more than the
individual teams working on their own.

I suggest that it would be worthwhile for the SC to host a meeting of
various individuals and teams in this area to see whether there is some
synergy there.

--Michael Dillon

Jared's network operator contact list

[michael.dillon]

Why doesn't the SC discuss with Jared Mauch, the possibility of hosting
his network operator contact list at NANOG.ORG?

http://puck.nether.net/netops/

This could also be integrated into the AUP, i.e. don't ask questions
that can be answered by this list of NANOG resources (NOC list, FAQ,
etc.).

For that matter, has anyone discussed with PCH about integrating
INOC-DBA into NANOG?

http://www.pch.net/inoc-dba/

In some ways, Team Cymru has been far more effective than NANOG. Is
NANOG fated to only be a mailing list and conferences even though it has
grown a governance infrastructure that could be used to do a lot more?

http://www.cymru.com/

--Michael Dillon

RE: mail operators list

[michael.dillon]

> I think Randy has hit on the solution to all problems.  I'm 
> waiting for the new Al Gore movie where he puts the solution 
> to genocide to the problem of global warming. 

Speaking of email spam, I believe the reverend Kris Korda has
already beat Gore to the punch by promoting the Church of Euthanasia
about 15 years ago. I wonder what happened to that church? Maybe
like the Shakers of New England, they all died out...

I wonder whether a better solution might be to create a second
NANOG list for all the non-core operational issues from mail
to P2p to botnets. People who are only interested in the traditional
NANOG core, can stay away. People who are interested in broader
operational issues can join the list and ignore threads that are
not relevant.

--Michael Dillon

RE:autoresponders

[michael.dillon]

> Mail between two users is mail between users. Mail that hits 
> the list is mail that is list mail. 

You've hit the nail on the head. Is there any way that the NANOG mailing
list can prevent such unwanted mail between two users?

Actually, yes there is.

The NANOG list could ensure that the From, To and Cc lines never have
any addresses in them other than the one instance of [EMAIL PROTECTED]
Any entries from those fields could be placed at the beginning of the
message text as:

  Orginal-From:
  Original-To:
  Original-Cc:

This would make it hard for people to send mail between users unless
they really wanted to do so. And those who really want to do so can
simply cut'n'paste from the message text to the To: header.

This is a technical possibility that could be implemented in the list
software and which would alleviate the situation. I suggest that the MLC
or SC should ask the list admins how hard this would be to do. And note,
if this is not currently in the list software, someone can get some
brownie points for contributing a bit of fairly straight-forward code.

--Michael Dillon

RE: proposed NANOG charter amendments

[michael.dillon]

> What 
> does the review of the PC hope to accomplish, anyways? 

Make sure it is not a marketing presentation or sales pitch.
Make sure that slides are reasonably clear and understandable.
Make sure that the topic has not been thrashed to death.
Make sure that the topic is relevant to network operations.
And other stuff like that.

Of course the detailed review performed by program committees of
academic conferences tends to dive deeply into the subject matter and
related research done by others. That kind of review makes sure that the
presentation is technically correct, is placed in the context of related
research, and is not a rip-off from some other researcher's work.

Really, what difference does it make how the program committee does
their work? They either produce good programmes or they don't. If they
really do badly then they can be replaced. There is a mechanism for
that. It's all the responsibility of the Steering Committee. Either let
them get on with their work, or file a complaint with them, and then let
them get on with their work.

--Michael Dillon

RE: proposed NANOG charter amendments

[michael.dillon]

>  Each member of the Program Committee must review all 
> presentations submitted for each meeting.

It doesn't say in what detail they must review it. Seems to be a good
idea for every member to give every presentation a cursory review. Also
seems to be a good idea for the committee to have detailed reviews done
by one (or a few) members who are specialists in a field.

If the committee has common sense then that line in the bylaws doesn't
really matter one way or the other. It neither promotes common sense,
nor does it prevent common sense from being applied.

--Michael Dillon

RE: proposed NANOG charter amendments

[michael.dillon]

>   >   by whom?  pc chair?  sc (which appointed them)?  michael
dillon?

>   I think  michael dillon is a fine choice. 

Vijay, it is rude to make fun of people who are showing their senility.
Stop picking on Randy.

In any case, I would think the committee should be able to remove its
own members by voting them out for cause. If the cause is in the bylaws,
then the process doesn't need to be there in every detail. And if a
committee did not feel it within their power to remove a member then
they should certainly feel it in their power to petition the appointing
body (sc) to do so. Again, this process is common sense and doesn't need
to be codified to the nth degree.

I say: NO TO THE BUREAUCRATIZATION OF NANOG! DOWN WITH THE RUNNING DOG
LACKEYS OF PROCEDURALISM!

Does anyone still hold to Common Sense as a virtue?

--Michael Dillon

RE: AUP modification - full first and last names

[michael.dillon]

> Personally, I have a hard time understanding much of the 
> opposition to including real names along with one's handle - 
> yes, future employers may look at the assorted NANOG 
> archives, but if one's content is good, that would be an 
> asset, not a liability.  

What!? Forcing people to use their real name would improve the quality
of NANOG content!?

We can't have that, now can we.

I disagree that there is such a thing as a well-known alias in the NANOG
community, where I define the community as the over 10,000 people who
read the list. The number 10,000 comes from the approximate total of
subscriber addresses (both per-message and digest subscribers). The
actual number of readers is probably larger than this because some
addresses are gated into private mailing lists and there must be some
people who read through the web archive.

But, maybe this whole alias issue is a red herring and we shouldn't
worry about it so much. After all, there is nothing magical about real
names, even if we tend to use a reserved vocabulary to form them. They
are merely labels for a person. The list AUP should be more about
content of the messages and less about which labels the writers use.

--Michael Dillon

RE: AUP modification - full first and last names

[michael.dillon]

> I'm very well aware that there are persons whose alias is far 
> better known in the community than their real name,

Which community is this?

I would suggest that in the community of 10,000 email subscribers to the
NANOG list, there is nobody whose alias is better known than their real
name.

> The firstname/lastname is in continuance of that dilemma - 
> "[EMAIL PROTECTED]" 
> (for example) complies with the letter but not really with 
> spirit. How should MLC distinguish between 
> [EMAIL PROTECTED] where this person is known by 
> everyone as "something.com peering coordinator" and someone 
> else who is really an anonymous alias?

What is wrong with signature blocks like this one?

--Michael Dillon

RE: AUP/autoresponders, rehashed

[michael.dillon]

> NANOG reserves the right to remove any member of the mailing 
> list who has configured automatic response software to 
> respond to any address other than the envelope sender, 

What is the "envelope sender"? 
Policies should be in plain English.

> As far as CR being rude...  I find it rude for people to 
> assume that you want to talk to them. 

This is the problem with taking the discussion to the level where words
like "rude" are used. Small differences become disputes, people take
sides, both sides get emotional and the namecalling escalates.

--Michael Dillon

RE: AUP/autoresponders, rehashed

[michael.dillon]

>8.  Challenge/response sender whitelisting software which 
> requires interaction by any party to validate a post to the 
> NANOG mailing list as non-spam shall be treated by the list 
> administration team like any other condition that generates a 
> bounce message.  Subscribers with software (such as but not 
> limited to TMDA) that is (mis)configured in this fashion are 
> subject to removal from the list without notice, and are 
> welcome to resubscribe at such time as their software is fixed.

8. If a subscriber operates a challenge/response system or vacation
notice system or any other automated system which sends messages to the
NANOG list servers then the list administrators will treat this as
equivalent to bounced email. Subscribers with software that is
(mis)configured in this fashion are subject to removal from the list
without notice, and are welcome to resubscribe at such time as their
software is fixed.

This is in response to RS's request to try and integrate my suggestions
into the proposed rule 8 wording from Alex.

By the way, Alex, kudos for the work you are doing. And further kudos
because you are discussing it openly on the futures list before you rush
into it. 

--Michael Dillon

RE: AUP/autoresponders, rehashed

[michael.dillon]

> Aren't you bothered by silly autoresponders whenever you post 
> to nanog-l?  

I rarely see such messages and you have to admit that I post a lot.
That's why I developed the theory that 99% of such C-R responses are
caused by people like you who don't edit their mail headers and send two
(or more) copies of a posting, one to the list and one direct to the
poster. Of course, I could be wrong. Maybe we just have a good mail
system that discards C-R messages before they get to me. In any case, I
have developed the habit of editing my headers to send only one copy of
the message to the list.

> Asking people to be nice doesn't work - people who have 
> autoresponders clearly don't know any better or don't care. 
> Waiting for them to come back from vacation to fix it - means 
> it can be few weeks for autoresponder to be running - why bother? 
> 
> Remove from the list, ask them to resubscribe when they come 
> back and it is fixed.

I agree with this, if it is triggered by the subscriber sending
automated responses to the listserver. But only if they go to the
listserver addresses.

--Michael Dillon

RE: AUP/autoresponders, rehashed

[michael.dillon]

> That said, one should not be surprised if his opinion is 
> marginalized if he simultaneously complains and refuses to be 
> engaged in the solution...

I tried the wiki for a period of several days, writing several
contributions. But when my edits were all deleted because they didn't
fit one other person's idea of a NANOG wiki, I walked away. Who needs
that aggravation? I don't believe in following the crowd and protecting
the status quo. And I don't believe that it does NANOG any good to
enforce the whacko cutesy narrowminded behavior of a certain vocal
minority. Running big internetworks is serious business and seriously
complex. Flippant banter doesn't help people learn what they need to
know in order to scale their networks. There should be less black and
white, less demonization and sainthood on this list. There are 10,000 or
more people on the NANOG list, and given the tiny percentage that write
postings, it seems to me that the number one goal of those people is
LEARNING.

That's why organizations like Wikipedia (for whom I have written and
edited articles) have developed their NPOV (neutral point of view)
policy along with clarity and other things. It helps make written
material more useful for learning. 

> I'm not on the SC and obviously don't speak for them, but I 
> believe the cluepon wiki is generally considered by the SC 
> members jointly and severally to be a good example of 
> community involvement and people stepping up and doing the 
> Right Thing.  

Maybe it has changed since when I was there, but I've lost interest. If
anyone thinks any of my list postings are useful, they will copy some of
the content to the wiki. 

> I can point to the lack of any serious 
> proposals for in-housing the wiki as supporting evidence.

Apathy is not a vote of confidence.

--Michael Dillon

RE: AUP/autoresponders, rehashed

[michael.dillon]

> I think the plain meaning of "post to the NANOG mailing list" 
> is completely and unambiguously clear to everyone who reads 
> it.  If you have an alternative wording that you believe 
> makes the intent even more clear, by all means suggest it.

How about:

NANOG reserves the right to remove email addresses from the mailing list
if list member creates technical problems for the mailing list servers
including, but not limited to, list members who run a challenge-response
server that replies to list postings or list members who send vacation
messages in reply to list postings.

That could probably be written better and more genericly by talking
about any system that generates automated messages in response to
messages send by the mailing list server. But it is good to include the
more common example scenarios in the rule. 

> That's a great idea for a page on nanog.cluepon.net, and I 
> encourage you to write one up.

Sorry, I gave up on that game ages ago. It may be running Wiki software
but it's being run like a personal vendetta, not like a Wiki. As a
contributor to the original wiki over 10 years ago, as well as a
contributor to Wikipedia, I find nanog.cluepon.net to be a remarkably
hostile environment. If NANOG is to have a wiki, it should be run under
the auspices of the NANOG Steering Committee with substantially the same
AUP as Wikipedia except where that conflicts with the mailing list AUP. 

--Michael Dillon

RE: AUP/autoresponders, rehashed

[michael.dillon]

>8.  Challenge/response sender whitelisting software which 
> requires interaction by any party to validate a post to the 
> NANOG mailing list as non-spam shall be treated by the list 
> administration team like any other condition that generates a 
> bounce message.  Subscribers with software (such as but not 
> limited to TMDA) that is (mis)configured in this fashion are 
> subject to removal from the list without notice, and are 
> welcome to resubscribe at such time as their software is fixed.

I object to this proposal on the grounds that it is unclear and it is
not fair in the form quoted above.

Thought experiment:
I write an email message addressed to [EMAIL PROTECTED] and send it. It is
processed by the mailing list software and some-nanog-server.merit.edu
sends it to [EMAIL PROTECTED] Joe has a challenge-response system
installed and it recognizes that this email comes from a mailing list to
which he subscribes and simply delivers it to Joe. Joe finds this
interesting and writes a reply to the list. Then Weeble Fuzzlebratz
takes exception to Joe's posting and writes a reply which he sends to
me, Joe Bloe and [EMAIL PROTECTED] When Joe's challenge-response system
sees the mail from Weeble, it sends him a challenge-response message.
Weeble fumes and fusses and forwards this outrage to the MLC demanding
that Joe Bloe be dropped from the list.

Personally, I don't see why Weeble is outraged at such a thing. 

But the larger issue is whether or not Joe's software is correctly
configured. It allows messages from the NANOG mailing list servers to
pass unchallenged. But when a list member, not previously known to Joe,
sends a direct message to Joe Bloe, the challenge-response kicks in.
This seems to me to be right and proper and the normal way a C-R system
should work. Joe has done nothing wrong. But Weeble has sinned against
the robustness principle "Be conservative in what you send, be liberal
in what you accept from others". A variant of this principle is found in
the Fidonet motto "don't offend, and don't be too easily offended" which
arguably is more relevant to a mailing list. Has Joe offended? Or is
Weeble too easily offended?

If there is to be such a statement in the AUP then I think that it needs
to be unambiguously clear that it only applies to messages fowarded by
the mailing list servers and not to private messages sent directly by
list members. The fact that many list members are unable to manage their
To: and Cc: headers is not sufficient reason for changing the AUP.

Also, one wonders whether this might be better handled by education than
by punishments. Do we make it clear how to configure a C-R system on the
NANOG mailing list subscription pages?

--Michael Dillon

RE: AUP enforcement, cont'd

[michael.dillon]

> > fact, I would go so far as to say that you should slam people for 
> > ungrammatical English, quoting too much of another message, top 
> > posting, unclear thinking, rambling, and anything else that 
> an English 
> > teacher or an editor would critique. If people try to continue a 
> > thread that has gone on too long, make them pay for it, and 
> help raise 
> > the quality of the list postings as well.
> I think you are reaching now. You don't want enforcement but 
> you want editor that hassles people for poor grammar, and 
> gets ignored because nobody has to listen to the editor? 
> Sorry, that doesn't compute.

I'm saying that when you get into the grey areas that are not clearly
covered by the AUP, then act editorial. You don't have a big stick, but
you can still act constructively.

Also, the whole discussion is about threads that drag on too long and
annoy people because they are repetitive. Well, if we get a few
well-written, researched postings with references to back up the facts,
then the thread becomes less annoying. In fact, the very best postings
could be at the tail end of the thread, if someone (not an MLC member)
volunteers to summarize the thread, tie up loose ends, clearly set out
the facts and point people to references for further reading. After a
message or two like that, only the most antisocial people would continue
the thread.

> > In fact, this editorial behavior could be done semi-publicy 
> by copying 
> > any critiques to the nanog-futures list as well as to the offending 
> > writer.
> This has been discussed by MLC and SC (whether warnings 
> should be public
> record) - and decision was that "name and shame" is 
> inappropriate for number of reasons. Thus, warnings will stay 
> private, but MLC will report statistics back to the community monthly.

I agree, warnings should be private. But when an MLC member publicly
asks for a thread to die out, then I think that it is fair game to
publicly critique the messages that follow, on nanog-futures. Unless the
messages are so well-written, accurate and useful, that they kill the
thread through kindness.

--Michael Dillon

RE: AUP enforcement, cont'd

[michael.dillon]

> I'm not sure what should we do in this case. Would adding a rule #8: 
> "Disregarding instructions from mailing list team to end thread." 
> be appropriate? Do we want to have broader rule: 
> "Disregarding instructions from mailing list team"?
> 
> Clearly, MLC does not want to be seen as dictator, however, 
> if MLC is to "guide" the discussion, we need to have some 
> teeth to bite the people who don't listen.
> 
> Opinions?

1. We don't need another rule.

2. We don't need a broader rule.

3. We don't need any rules in order to enable someone to "guide" the
discussion.

4. We don't need any teeth.

5. Biting people is rude and nasty.

6. If the MLC does not want to be seen as dictator, then it should not
ask for rules which will justify it acting as dictator.

7. You'll win more flies with honey than with vinegar.

I think 7 is enough...

--Michael Dillon

RE: Broadband routers and botnets - being proactive

[michael.dillon]

> I am wondering what, if anything, is being considered in this case. I
> get the same post from him on *four* mailing lists, and only on Full
> Disclosure (I am not on the botnet list) is it on topic. I believe
that
> he's been warned privately, and multiple times, and I hesitate to
speak
> out (I'm not interested in becoming the lightening rod this time), but
> really, outside of 5 or 6 people, I just don't see the support. 

The fact is that there is a mailing list committee which is supposed to
deal with these issues. It us not up to Albert Meyer, or you or I to
decide what is acceptable on the list. In fact, it is not really
something that we should discuss on this list, because we have a mailing
list committee to handle these things. Albert should have sent his
complaint to that committee. So should you. The fact is that there are
8000 people on the NANOG list and we will never all agree on anything.
Many people on NANOG are not on any of the other lists where you see
Gadi's posts, therefore they see Gadi as an OCCASIONAL contributor with
knowledge in a specialised area who writes interesting and useful posts.
You see more of the same-old same-old.
 
> > Etaoin Shrdlu wrote:
> >  > Between us, he is *not* respected. He is nearly universally
> > disliked, on

By this, you can only mean that the majority of the VOCAL members of
these lists do not like Gadi. If they are like NANOG, that means that a
small minority of the total list membership are moved to express their
dislike. Statistics like this are meaningless.

More meaningful would be the number of complaints that the mailing list
committee has received about Gadi and how many of those complaints, the
ML committee feels are justified, based solely on Gadi's NANOG postings.
Also, you must realize that the ML committee has the power to warn
people to lay low for a while and they do exercise this power from time
to time. 

> >> I know he has his defenders, but they are tending towards being
> >> nonoperational in many of their posts in any case...

Internet operations has become so diverse that stuff which is
operational for one company may be considered non-operational for
another company. We just don't have a very clear standard for what is
and is not operational.

Also remember that the whole steering committee and ML committee
structure arose because some people were concerned that NANOG moderation
was too heavy-handed. It is to be expected that the ML committee will
moderate with a light touch and lots of borderline messages will appear
on the list. If you don't like that, feel free to start a blog and
encourage your friends to do the same. Or set up a more tightly-focused
list and see if anyone comes to play. 

--Michael Dillon

RE: Broadband routers and botnets - being proactive

[michael.dillon]

> Frankly, from the sidelines, Fergie and Gadi both pale in comparison
to
> the often brunt and bold calls for their dismissal.  In those
instances,
> I do think that Fergie's and Gadi's posts are much more on topic for
> NANOG then the condemning onslaught that follows.

I agree, and this is the main reason why I defend Gadi's right to post.
I can't help but remember what happened when William Leibzon joined the
list and several people spoke out boldly and bruntly to condemn him as a
spammer. It turned out that William was doing research to help identify
spammers and that the people loudly condemning him were criminal lowlife
types. And here we have another situation where someone trying to root
out botnets is loudly condemned by a few people.

I don't know whether or not any of these complainers are earning their
living from building and running botnets, but I do know that there are
blackhats on the NANOG list. And quite frankly, I can't understand the
purpose of making such open complaints when we have an ML committee to
deal with this stuff.

As long as the MLC disregards the contents of any complaints and only
rules on the basis of the actual message content by Gadi or whoever,
then I'm happy.

--Michael Dillon