[ih] infrastructure history [was: who invented the Internet]

[Vint Cerf]

ARPA and NSF especially are NOT laboratories; they are funding
agencies. They have strong technical leadership. The involvement of
individuals working for the USG and funding seems to me
incontrovertible in the Internet story. This is not to diminish the
essential role of the contractors (mostly academia but also private
sector - think of BBN, IBM, MCI among others)  involved in ARPANET and
NSFNET and Internet. Collaboration theme is strong here along with
very open processes and institutions.

vint


On Thu, Jul 26, 2012 at 1:05 PM, Dave Crocker <dhc2 at dcrocker.net> wrote:
>
>
> On 7/26/2012 9:44 AM, John R. Levine wrote:
>>>
>>> Looking at the details of research, development and operations for
>>> packet-switching and interworking, represented in today's Internet, I
>>> believe there was primarily government funding until the latter '80s,
>>> when commercialization started.  I believe the three notable
>>> exceptions were PARC's ethernet, Digital's routing work, and the
>>> interplay between PARC and ARPA reserarch folk.
>>
>>
>> I was more thinking of big help in that it was built largely by
>> non-government organizations out of non-government parts, albeit paid
>> for by government money.
>
>
>
> Well, that phrasing gets some objective details correct but, I think, still
> understates government's role.  (Mind you, as a resident of Silicon Valley,
> I consider private initiative a major religion; but it's not the /only/
> religion...)
>
> In the case of Arpanet and NSFNet, the criticial infrastructure efforts were
> government initiatives, not just in terms of funding but in terms of
> strategic, tactical and many technical formulations.
>
> Just to beat this into the ground, while I was at UCLA, I seem to recall
> hearing that Larry Roberts (head of ARPA's IPTO that directed Arpanet work)
> was driving the substance of the modeling work every bit as much as
> Kleinrock, et al.  (As I understand it, this was and probably still is a
> common characteristic of ARPA program management, since it tends to hire
> technical folk.)
>
> The organizational model for NSFNet -- including its forcing a move towards
> non-government sustain able funding -- were definitely government work,
> including the "market research" of the model with the predecessor effort,
> CSNet.
>
> I can imagine that the details of TCP/IP and internetworking conform more to
> the model of "non-government initiative benefiting from government funding"
> but I don't know enough of the particulars.
>
>
> d/
> --
>  Dave Crocker
>  Brandenburg InternetWorking
>  bbiw.net