[ih] History of Naming on The Internet - is it still relevant?

[John Day]

First, what does this have to do with Internet-History?  ;-)

Second, for DCS, it was a process and that is sufficient.

> On Aug 15, 2025, at 17:12, touch at strayalpha.com wrote:
> 
> It’s never that simple; every naming approach depends on a model and single model covers all possible perspectives.
> 
> Even thinking of a service as a process is flawed; it assumes a kind of “locus of work” rather than a “locus of information”, i.e., it’s not about the process, it’s about getting the information you need, regardless of who/how it’s determined.

Please explain, how an information generates a response.
> 
> Right now, the model is supposedly “service on a machine”, which is suboptimal a few ways:

a) what model where?

> 	- services can span more than one machine
> 	- it’s really interface, or more specifically interface address, of which there can be more than one of each
> 	- but that makes it hard (impossible) to have more than one instance of a service per IP address (yeah, there’s mDNS but it doesn’t go outside a subnet broadcast domain)

If you don’t do it right, it won’t work.
> 
> So let’s say we pick “name just the process”. What’s a process? Many services are composed of multiple components, providing multiple capabilities, each of which could have multiple instances. And how do you ensure that the names don’t collide, if they’re not dependent on something that someone, somewhere manages as an exclusive resource (e.g., global IP addresses)?

Yep, all worked out decades ago.
> 
> The reason it’s hard to find a single answer is that there isn’t one. Naming is always inherently multidimensional (time, system, service, location, etc.) and nearly always includes some aspect of “relativity” (as in ‘every perspective is relative’, there is no absolute frame of reference). 

Funny how that works, isn’t it?

Thanks for explaining the perspective.

Take care,
John

> 
> Joe
> —
> Dr. Joe Touch, temporal epistemologist
> www.strayalpha.com
> 
>> On Aug 15, 2025, at 5:31 AM, John Day via Internet-history <internet-history at elists.isoc.org> wrote:
>> 
>> Which is as it should be.  Farber got it right.
>> 
>> The only reason for naming ‘machines’ is for network management and even then it isn’t the machine but the process responsible for managing the machine.
>> 
>> Take care,
>> John
>> 
>>> On Aug 14, 2025, at 21:24, Dave Crocker via Internet-history <internet-history at elists.isoc.org> wrote:
>>> 
>>> On 8/14/2025 2:49 PM, vinton cerf via Internet-history wrote:
>>>> Dave Farber's distributed DCS system at UC Irvine effectively wrote into
>>>> the memory of the receiving computer.
>>> 
>>> As I recall, naming in DCS was to processes, not machines. So processes could move around transparently to other processes.
>>> 
>>> d/
>>> 
>>> -- 
>>> Dave Crocker
>>> 
>>> Brandenburg InternetWorking
>>> bbiw.net
>>> bluesky: @dcrocker.bsky.social
>>> mast: @dcrocker at mastodon.social
>>> -- 
>>> Internet-history mailing list
>>> Internet-history at elists.isoc.org
>>> https://elists.isoc.org/mailman/listinfo/internet-history
>>> -
>>> Unsubscribe: https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
>> 
>> -- 
>> Internet-history mailing list
>> Internet-history at elists.isoc.org
>> https://elists.isoc.org/mailman/listinfo/internet-history
>> -
>> Unsubscribe: https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
>