past vs future use

[Jeff Williams]

Stephen,

Stephen Sprunk wrote:
> 
> At 16:33 29-06-97 +0100, you wrote:
> >  I agree that 80% or more ISP's do fail in the first year.  But if you
> >look a bit deeper you will find that 96% of these fail due to lack of
> >adaquate funding.  The case that I am refering to in this discussion
> >does not suffer form that problem.  We already have $12m in initial
> >cash on hand, in addition a $8m line of credit, and currently a
> >$4m in residual partnership funding down stream.  So I don't think
> >we will be facing a failuer due to funding at any rate.
> 
> Good for you (golf clap).  If you have this much capital, I assume you also
> have a fair number of clients lined up.  If you have enough clients (which
> I will assume you do), you may be able to qualify for a /19 immediately, if
> you can meet the requirements under RFC 2050.  SWIP out 80% of the
> allocation and make sure each customer uses 25% immediately and 50% within
> the first year.

  Well first let me say, thanks for the "Golf clap".  >;)  Next is see
that you have not read RFC2050 very closely.  If you had in section
2.1 (4-7) partians to this area.  In those sections, there is no 
spicific pervision stated for this type of situation.  Not only that
if you had been following the discussion you would, of course already
know that.

  My question to you, is, are you speeking authoritativly?

> 
> >> The expected procedure for a new ISP is thus:
> >>
> >> 1. Connect to an upstream provider
> >> 2. Obtain some PA IPs from that provider
> >> 3. Efficiently assign those IPs to your customers OR do a bunch of
> >>    fake SWIPs that make it look like you're efficient
> >> 4. Repeat 2 and 3 until you have ~8192 PA IPs
> >> 5. Trade in your PA IPs for a /19 allocation
> >> 6. Make every customer you have renumber
> >
> >  I here what you are saying here.  But this method is too pacarious due
> >to point #6 [renumber], #2 [get PA IPs], and #4 [lather, rinse, repeat].
> >Been there done that!  Or the three ISP's
> >that I have been directly associated with only one did we need to
> >renumber.  And that was mainly due to this sort of planning.  The other
> >two we got alot smarter, and certianly didn't use this method, as we
> >pretty much did in the first one.  Hence, back to my original
> >question....  >;)
> 
> Would you care to enlighten the rest of the world as to the method you used
> for the latter two businesses?
> 
> >  Not likely!  This plan or method is definatly flawed and of course
> >very likely to create a failier senerio.
> 
> I never said I liked, suggested, endorsed, or otherwise felt anything
> positive about this plan.  That's just how it is (now).

  I am sure this is NOT true.  In fact I KNOW that it isn't in all
cases.  Yes it would be true for SOME small start-up ISP's.
> 
> If you don't like ARIN/RIPE/APNIC policies, become a member and put up a
> vote to change them.

  I intend to dojust that.  >;)
> 
> Stephen

Fondest regards,
-- 
Jeffrey A. Williams
DIR. Internet Network Eng/SR. Java Development Eng.
Information Eng. Group. IEG. INC. 
Phone :913-294-2375 (v-office)
E-Mail jwkckid1 at ix.netcom.com