Stupid server semantic argument (was: Non Linux but network tech question)

[VirginSnow at vfemail.net]

> Date: Sun, 24 Jun 2007 22:05:12 -0400
> From: "Ben Scott" <dragonhawk at gmail.com>

> On 6/21/07, Bill McGonigle <bill at bfccomputing.com> wrote:
> > Perception doesn't have any effect on the Comcast NOC, but it has an
> > effect on the ToS the company can get away with.
> 
>   Says who?
> 
>   The point I'm trying to make, and that most people seem to be
> missing, is that *Comcast controls the terms*.  This isn't a situation

Who controls the terms really isn't the issue.  The issue is Reality.
What *in reality* does Comcrap do when you violate their ToS.  In most
cases, I'd posit zip.

As much as I abhor philosophy, I want to break reality down, for you,
into four strata:

  (1) What the Terms of Service say.  This is the "theoretical level
      of reality".  This is where business analysts, lawyers, and
      language-picky Linux nerds live.

  (2) Official company policy.  This is where managers make the
      decision "which legally reserved rights of ours do we wish to
      enforce?" and "what policy will this company adopt for
      operations purposes?".  These policies are generally distributed
      to underlings in "Policy Manuals" which are never read and only
      briefly referenced during employee training.

  (3) Unofficial company policy.  This is where employees make the
      decision "what part of the policies in this stupid handbook do I
      want to comply with?" and "what on earth am are they saying I
      supposed to be doing, anyway?".  This is where most employees in
      a company operate.  In the particular case ISPs, this is MOST
      (if not all) of tech support staff fall into this category.

  (4) What's done in reality.  This is where all the Linux nerds and
      spammers live.  What you can get away with on a network often
      goes beyond even unwritten policy.  There are aspects of ISPs'
      networks with which the ISP is often unaware or unconcerned.  In
      some places, port 25 is firewalled.  In others, it's not.  It's
      not firewalled because (even though it may be policy), it's not
      "in reality".

So, there are many levels at which you can "allow" or "deny" your
subscribers to/from doing certain things with their feeds.  You seem
to be focusing on #1, the legal-verbal-business layer.  What I'm
concerned with is #3, the actual, unofficial, unwritten "policy" that
the company follows.  Not the ToS.  These are two different realities.

   "I reject your reality and substitue my own."

>   If you call Comcast trying to play semantic word games, in all
> likelihood, they're just going to summarily shut you off anyway.  See

I doubt it.  Most tech support won't understand or be interested in
the meaning of your ToS.  They'd most likely put you on eternal hold.
You'd probably (if, who knows why, you wanted to argue verbiage) have
to talk with management.  And in that case, you risk making enough of
a stink to get noticed and get beaten with a spool of hardline.

>   If someone comes up with some popular "client" application that
> needs to listen on port 25, or any other equally asinine scheme,
> Comcast will just allow that, but still deny the rest.  If need be,

That's the point.  Creating an IM client that interfaces with gmail
(substitute your own flashy, useless, fad application here) that
listens on port 25 would encourage ISPs to *not* block that port.
Then, those of us who use port 25 for its real purpose (email) would
be able to get on unencumbered.  It's insanity curing insanity.

> they can simply put a bandwidth meter on your modem, and sock you when
> your usage goes out of balance.  See #1.

That would require effort on their part.  ISPs generally want to
ignore as many people as possible and just collect monthly fees.  They
want to do as little as possible and make as much money as possible
doing it.

>   Note that I'm not saying you should immediately shut off your SSH
> server or port 25 listener or whatever.  Staying below the radar is
> fine if that's what you want to do.  I'm just saying that arguing

This is reality #4.  Living here is a little risky, because you never
know when any given part of reality #3 is going to start being
enforced in reality #4.

> Country *and* Western."  Whether competition be enabled by structural
> separation, many small local providers, distributed P2P wireless nets,
> or something else, I'd say this is the best bet for the whole, in the

IMHO, the best chance we have for reclaiming our Internet feeds (and
our telephones) is large-scale ubiquitious WiFi coverage.  With the
planned reallocation of the OTA TV broadcast spectrum, the WiFi
community is in the (historically) unique position to be able to
muscle its way - literally - across the country.  It's an obvious
freedom solution.  I believe we should be pursuing comprehensive-
coverage WiFi like our lives depended on it.  Indeed, our sanity may
depend upon it.

> seems more than a little far-fetched, to me.  This also destroys the
> anti-leech protocol arguments; implementing such just decreases the
> popularity of the protocols.

Anti-leech protocols?

So, I'm wondering.  If you try (for EXAMPLE ONLY) to connect to a
shared folder on your windoze box from a remote site, and call tech
support for help, what will they say?

I suspect they'd probably say that they firewall those packets for
your own protection.  They *probably* won't even mention their server
policy (or non-policy) at all.  I suspect that weapon's reserved for
spammers and residential dot-com'ers.

If anyone has first or second-hand experience with what, in fact,
Comca$t will do in a situation like this (trying to access your own,
password-protected, computer remotely), I'd love to know.

reality.level-=4; exit(0);