Copy Link
Add to Bookmark
Report
Netizens-Digest Volume 1 Number 211
Netizens-Digest Thursday, November 19 1998 Volume 01 : Number 211
Netizens Association Discussion List Digest
In this issue:
[netz] Re: Is Netscape filtering sites?
[netz] Re: Is Netscape filtering sites?
Re: [netz] Re: Is Netscape filtering sites?
[netz] Re: Is Netscape filtering sites?
[netz] Re: Is Netscape filtering sites?
Re: [netz] Re: Is Netscape filtering sites?
----------------------------------------------------------------------
Date: Thu, 19 Nov 1998 15:32:03 -0400
From: kerryo@ns.sympatico.ca (Kerry Miller)
Subject: [netz] Re: Is Netscape filtering sites?
Carsten,
> What I was saying was that if a (whatever) change in the IP #s occurs,
> this would mean reorganising a few things and would be a good
> opportunity to modify the name system as well (on a psychological more
> than on a technical basis).
>
Yes, it would be a grand opportunity -- if everyone would just agree to shut
the present system down! I think we have to keep thinking of how to
*change, not replace -- and how to keep the *ability to change* as a
functional aspect of whatever system results from a change.
The commercial world likes to boast of its 'innovations' but a) most changes
are very miniscule, b) a good many began with small fry who were *not
commercial, and c) there are examples galore of practical possibilities which
are not developed because the _marginal_ benefits are not as big as maybe.
The *systemic* effect is that corporatism is fundamentally a conservative
factor, because the 'bottom line' is not how to 'do things better' (DTB) but
how to 'increase market share' (IMS) which is a derivative value.
The proposal to privatize the DNS creates a monopoly. With 100% market
share, what intrinsic reason does ICANN (e.g.) have to make changes?
None. They say it will be 'responsive' to net 'stakeholders' interests - so we
should then ask what reasons/ interests those may be: I'm sure, for every
DTB lobbyist there will be 100 for IMS.
> > Those TLDs are simply one way of divvying up the implicit .us
> > domain.
>
> No. If they were, there wouldn't be any problem. But many European
> sites now have .com or .org addresses. This also adds to the
> confusion.
>
The confusion is only ours, who still think in terms of national entities...
> I'm not sure what ou mean with "wildcards", sorry. I just meant those
> to be replaced by anything, just like I would write *.com...
My fault, I thought I recognized the * from RFC1480 (see next msg)
> No. A "hop" would be "on" the route. What you have in mind, if I
> understood you well, is more like what Monolith do today, just with
> a choice of several IP #s per address (BTW: how would that
> choice be made, if no physical person is available for that purpose ?)
As far as I know, nobody but us humans has the slightest use for names.
But you're right, the menu would not be a hop.
> > b) nothing whatsoever stops a user from plugging in the IP number from
> > the get-go -- then or now.
>
> Sure. But, then, we're not talking DNS or your system anymore :)
> BTW: what would happens if "someone" decided not to expand IP numbers
> but to use dynamic ones for everyone ?
>
Better yet, what if somone introduced a browser which bookmarked by IP
addresses?
kerry
------------------------------
Date: Thu, 19 Nov 1998 15:32:02 -0400
From: kerryo@ns.sympatico.ca (Kerry Miller)
Subject: [netz] Re: Is Netscape filtering sites?
Greg,
> >Those TLDs are simply one way of divvying up the implicit .us domain. Even
> >in 1986 or whenever, it was obvious that a grad student wasnt going to be
> >able to catalog every last Stateside URL. Postel's problem then is our
> >problem now, with a decade of neglect/ inertia/ oppportunism on top.
>
> I don't see it that way. The Internet grew up very quickly.
> Decisions were made because it was considered important to keep the
> infrastructure going, as many people had come to depend on it. I
> think, as Dave Farber has pointed out, that honest mistakes were
> made.
>
Mistakes were made, and I repeat, people have been capitalizing on those
mistakes ever since. Now the tail is wagging the horse.
> I don't think anyone foresaw the rampant commercial takeover. You
> have to keep in mind that the Internet was designed as part of an
> open-ended research project.
>
You're telling me!
> I don't really have any alternatives, if by "GS" you're referring to
> me. These are complex problems that I think will take years to
> solve.
>
Therefore the first item of business is to erect a problem solving system,
not something that will operate as if all problems are already solved.
> > Physical schmysical ;-) it'll simply look like one more hop to a
> > trace router.
>
> No, not really, because it is not just another "hop" in IP routing;
> it's an additional reference at the application level.
>
But only at the application level (if that's what we call the 'user' now), and
not 'additional' in any more oppressive way than having to hold the control
and alt keys 'in addition to' delete.
> It's unwise to rely on IP addresses being fixed. One could argue
> that it's unwise for DNS names to be fixed as well, I suppose, but at
> some point you have to have some name that is at least assumed to be
> fixed for the time period you want to use the resources associated
> with that name. Thus, there is always going to be a need for some
> mapping from a high-level name to a network resource, whether that
> name is a DNS name, a Real Names object, etc.
>
That may be the way we think now, but I like Postel's Proposition:
While ease of use to the end user is desirable, a higher
priority must be placed on having a system that operates.
Whatever structures people have had to deal with, there has always been
some ways that are 'easier' than others. There is no reason whatsover that
suddenly makes it important to *design* in terms of ease of use. If movable
IP addresses are a solution to some problem, then by all means lets have a
system that accomodates that, but I dont see that ' there is always going to
be a need' for *any* thing in particular, and any organization that commits
itself to such an axiom is d-doubleO-med.
Btw, earlier you wrote,
>There *is* a shortage in "attractive" names -- short, easily
> remembered names, and the names that have already been established as
> brands, companies, or products. However, in the old days, when an
> "attractive" name was in use, someone would just choose another name.
In the old days, there were multiple markets, both geographically and
functionally. You could have 'Best, Inc' in your town (or making shoes) and I
could have 'Best, Inc' too, in my town (or refilling candlesticks), and this kind
of ambiguity was perfectly tolerable (why should anybody in my town go to
yours anyway? ;-)).. That we are now trying to move from such discrete
functions to a single unified system is *exactly* the conceptual
(epistemological) problem, and it's appalling that all the scrambling for 'easy
names' (and the apparent readiness of a DNS admin to support it) stems
from the local-isolation frame ('business as usual' you might say) without
admitting the fundamental absurdity.
kerry
------------------------------
Date: Thu, 19 Nov 1998 13:28:53 -0800 (PST)
From: Greg Skinner <gds@best.com>
Subject: Re: [netz] Re: Is Netscape filtering sites?
In article <19981119193521.AAB22479@LOCALNAME> Kerry Miller wrote:
>Mistakes were made, and I repeat, people have been capitalizing on those
>mistakes ever since. Now the tail is wagging the horse.
I don't think I've written anything to the contrary.
>Therefore the first item of business is to erect a problem solving system,
>not something that will operate as if all problems are already solved.
Like I've said before, I don't think the way ICANN has set itself up
is the best way to go about solving the problems. From what I read on
the ifwp list, lots of other people feel similarly.
>> No, not really, because it is not just another "hop" in IP routing;
>> it's an additional reference at the application level.
> But only at the application level (if that's what we call the
> 'user' now), and not 'additional' in any more oppressive way than
> having to hold the control and alt keys 'in addition to' delete.
But this reference would need to be employed by lots of applications,
not just web browsing. You need to realize that domain names are used
by a vast array of Internet applications and services. It is not
practical for them to be changed to require IP addresses. It is also
not practical for them to be changed so that they must require user
intervention to make distinction between entities that might appear in
the same menu. Many of these applications and services (for example,
mailing list redistributors) are automated.
>Whatever structures people have had to deal with, there has always been
>some ways that are 'easier' than others. There is no reason whatsover
>that suddenly makes it important to *design* in terms of ease of
>use. If movable IP addresses are a solution to some problem, then by
>all means lets have a system that accomodates that, but I dont see
>that ' there is always going to be a need' for *any* thing in
>particular, and any organization that commits itself to such an axiom
>is d-doubleO-med.
I don't understand what point you're trying to make here.
>That we are now trying to move from such discrete functions to a
>single unified system is *exactly* the conceptual (epistemological)
>problem, and it's appalling that all the scrambling for 'easy names'
>(and the apparent readiness of a DNS admin to support it) stems from
>the local-isolation frame ('business as usual' you might say) without
>admitting the fundamental absurdity.
I never claimed that the DNS solved all types of naming problems or
that there weren't (potentially) other, better ways of doing name
resolution. All I am saying is that the DNS is designed to solve a
set of specific problems have to do with mapping network names to
network resources in a large, distributed networked environment.
There will most definitely be a need to transition to other types of
naming services as the Internet grows, and some people in the IETF
community are working on this sort of thing. However, this is a very
long process, requiring time *and* money for research, development,
testing, and deployment. It's not the sort of thing you can just plug
in to a large, operational framework. This is the reality that the
Internet is now; even if it were totally noncommercial, its size and
scope requires that changes be made slowly and carefully, because so
many people use it and depend on it.
I have already said that I am not happy with many things the ICANN is
doing or purports to do. Others have expressed similar reservations.
I am not an ICANN board member. I have no authority over what they
decide to do. Why haven't you been sending your complaints to them
directly? They are not on this mailing list, as far as I know.
However, they are on the ifwp list.
- --gregbo
------------------------------
Date: Thu, 19 Nov 1998 20:32:00 -0400
From: kerryo@ns.sympatico.ca (Kerry Miller)
Subject: [netz] Re: Is Netscape filtering sites?
Carsten,
> What I was saying was that if a (whatever) change in the IP #s occurs,
> this would mean reorganising a few things and would be a good
> opportunity to modify the name system as well (on a psychological more
> than on a technical basis).
>
Yes, it would be a grand opportunity -- if everyone would just agree to shut
the present system down! I think we have to keep thinking of how to
*change, not replace -- and how to keep the *ability to change* as a
functional aspect of whatever system results from a change.
The commercial world likes to boast of its 'innovations' but a) most changes
are very miniscule, b) a good many began with small fry who were *not
commercial, and c) there are examples galore of practical possibilities which
are not developed because the _marginal_ benefits are not as big as maybe.
The *systemic* effect is that corporatism is fundamentally a conservative
factor, because the 'bottom line' is not how to 'do things better' (DTB) but
how to 'increase market share' (IMS) which is a derivative value.
The proposal to privatize the DNS creates a monopoly. With 100% market
share, what intrinsic reason does ICANN (e.g.) have to make changes?
None. They say it will be 'responsive' to net 'stakeholders' interests - so we
should then ask what reasons/ interests those may be: I'm sure, for every
DTB lobbyist there will be 100 for IMS.
> > Those TLDs are simply one way of divvying up the implicit .us
> > domain.
>
> No. If they were, there wouldn't be any problem. But many European
> sites now have .com or .org addresses. This also adds to the
> confusion.
>
The confusion is only ours, who still think in terms of national entities...
> I'm not sure what ou mean with "wildcards", sorry. I just meant those
> to be replaced by anything, just like I would write *.com...
My fault, I thought I recognized the * from RFC1480 (see next msg)
> No. A "hop" would be "on" the route. What you have in mind, if I
> understood you well, is more like what Monolith do today, just with
> a choice of several IP #s per address (BTW: how would that
> choice be made, if no physical person is available for that purpose ?)
As far as I know, nobody but us humans has the slightest use for names.
But you're right, the menu would not be a hop.
> > b) nothing whatsoever stops a user from plugging in the IP number from
> > the get-go -- then or now.
>
> Sure. But, then, we're not talking DNS or your system anymore :)
> BTW: what would happens if "someone" decided not to expand IP numbers
> but to use dynamic ones for everyone ?
>
Better yet, what if somone introduced a browser which bookmarked by IP
addresses?
kerry
------------------------------
Date: Thu, 19 Nov 1998 20:32:00 -0400
From: kerryo@ns.sympatico.ca (Kerry Miller)
Subject: [netz] Re: Is Netscape filtering sites?
Greg,
> >Frankly, I do not see why a menu site should become 'popular';
> >wouldn't you rather have a name shared by *fewer* parties than be
> >tagged at the bottom of a long list?**
>
> A menu site that hosts Columbia Broadcasting would become quite
> popular, simply because there are a lot of people who want information
> about them.
>
Arent we confusing two senses of the word? The popularity of a given final
destination is entirely distinct from the popularity of the menu which *people
who don't know how else to get to that destination* will use. San Jose may
be a popular vacation spot, but we dont say the roadmaps and transportation
schedules that reference SJ are popular for that reason.
> >Isnt it interesting that the fundamental conceptuial problem in all
> >of this has to do with the knot that binds all the distributed
> >net-links together? That organizationally speaking, this fantastic
> >*net* doesnt begin to compare with the electricity transmission grid,
> >or even your city water supply reticulation?
>
> I'm not sure what point you're trying to make here.
>
Your uncertaintly points up the difficulty of the issue: people have utilized
networks of all sorts for years, but they did not involve individual expression,
nor have they had to 'scale' indefinitely. If we had an electrical engineer
involved in this discussion, however, I'm sure she would stand amazed at our
trying to 'reinvent the wheel,' and not getting any further with our stone axes
than the simplest model of all -- top-down centralized hierarchical domain
administration -- and then scratching our heads, admitting all sorts of
problems on one hand, and denying that there could be any alternative on
the other.
What does it take to get someone with your experience to *collaborate in
the description of a distributed administration? If you look back through the
archives of this list, you will see that I didn't *propose* anything -- surely by
now it's quite clear that I don't know enough of the details to offer any sort of
'model' -- but you insist on responding as if finding the compelling arguments
is up to me, and if I cannot rebut your criticisms then the status quo wins the
argument.
On the contrary, I simply *asked* why such and such wouldnt work. After
quite some time, it appears that we can boil down some particulars:
> *change* would be cumbersome and expensive.
> *Scalability*, [specif. the size of lookup tables becomes unmanageable]
> new applications... required more sophisticated *resolution* of names
and of course, *convenience* at the application level.
I thank you for these answers. Now I will try to find someone who is
interested in talking about them as if they themselves were *questions to be
answered* - as amenable to human ingenuity as the conditions which have
created them. Who knows where that may lead?
> >** Since alphabetization is moot (they're all www.cbs.com, remember),
> >obviously menu entries will build by a first-come, first-listed basis. And as
> >everybody knows, consumers find it remarkably inconvenient to look beyond
> >the first page of any list of hits.
>
> This is an example of how scalability affects use. In your model,
> people would tire of looking for the CBS they wanted if it was not on
> the first page. This would raise the sorts of objections we have now
> to the current DNS model, because every CBS organization would demand
> to be listed first.
I would say it's an example of equilibrium or homeostasis. If Consumers get
tired of looking, then the 27th or the 61st Producer will think twice about
choosing the name CBS. In other words, instead of holding the *value of
direct, instantaneous, unique and 'easy' access as a value around which
everything else must revolve, I let the value of identifying - or being identified
by - a certain name float. The first approach works, as long as the name
space is 'infinite', but as soon as the boundaries - in terms of desirabilty, or
number of hops or menu picks or anything else - come into view, then
behaviour pattterns based on the former assumption are no longer useful.
The problem is no longer how to give every site an easy name, but how to
set up a system so that names distribute themselves more or less equably.
This means not depending on the limits that have worked for 20 years, but
building on that experience, and stretching out just enough past the 'tried
and true' that we have a chance of being able to recognize whatever the *next
limitations will prove to be. In short, this is what I have, apparently futilely,
tried to convey by speaking of the net structure as a learning system.
> Search engines, like every other piece of code
> that requires network access, use name servers to resolve network
> names to network resources.
>
Yes, of course they do -- as long as the DNS is there for their use, they'd
be fools to duplicate it. But suppose the search engine sites were
registered 'mirrors' of the DNS - wouldnt their internal workings then give the
user an even faster and more convenient response? (And then, what if they
*were* the 'DNS', cooperatively mirroring one another... No, that's asking too
much, even if it is Xmas season!)
kerry
------------------------------
Date: 20 Nov 1998 02:34:36 +0100
From: Carsten Laekamp <lakamp@capway.com>
Subject: Re: [netz] Re: Is Netscape filtering sites?
kerryo@ns.sympatico.ca (Kerry Miller) writes:
> > > Those TLDs are simply one way of divvying up the implicit .us
> > >domain.
> > No. If they were, there wouldn't be any problem. But many European
> >sites now have .com or .org addresses. This also adds to the
> >confusion.
> >
> The confusion is only ours, who still think in terms of national
> entities...
>
I'm sorry but it's also that of the naming system. On one hand, you
have national domains and on the other the placeholders of what you
call "the implicit .us domain". This would work rather well if it were
really the case. But:
- - those TLDs also cover Canada
- - European (and Asian ?) companies and organisations are massively
switching to using .com and .org too.
With the current evolution of the Internet in these countries, the
effect on the availability of "simple" addresses is rather obvious,
IMO.
> As far as I know, nobody but us humans has the slightest use for
> names.
>
Hmmm... have you ever submitted a URL to a Web search engine or
directory ? You don't enter an IP# :) (of course, this wouldn't be too
difficult to change).
E-mail also uses names and not numbers... and this would be much more
difficult a problem to solve. BTW: e-mail is a good example where the
names are used by humans but are then handled remotely, without any
human interacting.
And how about "sites" that use DHCP ? How about servers that cannot
afford any interruption of operation and are therefore moved from one
machine to another when changes/upgrades are made ? All this is
handled much easier with DNS.
> Better yet, what if somone introduced a browser which bookmarked by IP
> addresses?
Well, that would solve *some* problems for the WWW (and create
different ones) -- but hardly for other connections.
- --
Carsten Läkamp
claekamp@mindless.com
------------------------------
End of Netizens-Digest V1 #211
******************************
