t***@datacaptureinstitute.com
1999-05-06 16:14:27 UTC
Greetings all;
I'm glad to see the WEIRD chartering process moving along...
Let me first post a disclaimer; this might be "coloring outside the lines"
of the charter as it is being written. But what the hey, this is an
exchange of ideas, right?
Picking up on Chris' idea about WEIRD putting IETF work in context with
related standards efforts, we could start by making up an indexed set of
abstracts. I am imagining structured, compact descriptions of the various
works-in-progress in IETF - and elsewhere (the "rest of world" area).
...You could imagine a table looking something like:
STD_BODY, STD_CMTE, STD_NAME, DATE, KEYWORDS, ABSTRACT...
The announcements on Internet-Drafts might be a good source to begin with,
as they usually include a reasonably succinct summary paragraph including
key terms. We'd just have to be careful to capture the semantic content of
"does not address" or "is out of scope" references (usually in the last
sentence).
This idea leads me to post two questions to the list:
1. Does one of our members have a Perl "crawler" handy, that could walk the
Internet-Drafts archive and extract/decompose these paragraphs?
2. Can we jointly compile a list of similar document archives from other
standards bodies that might be subject to the same approach?
The framework suggested above could then be "mined" to generate the view of
relationships and threads that is being discussed as the WEIRD value
proposition.
If we want WEIRD to relate IETF and "ROW" then it seems to me this calls
for a self-maintaining reference source. If we try to do this with the
"person-to-person" approach, we would need a voluntary contribution from a
member of every standards committee... On the other hand, the use of a
tool as described above would hopefully avoid the duplication of effort
involved in each of us digging through various archives.
I've taken a "bash" at the search.ietf.org tool; it presents a rifleshot
capability. You *can* find what you're looking for - if you know what it
is! But it's a place to start...
We could pull together a "candidate thread collection" with a few queries
from this resource. This might be a way to quickly generate an answer to
Chris's request, so succinctly described in the example "FOO" IAB
functional goal.
So, I submit that one foreseeable goal of a WEIRD information resource
would be to provide a flexible, but structured search capability (with
examples) to allow individuals to discover related topics as they emerge.
This would be possible given a...
"dynamically compiled searchable meta-standards archive"
"Sure, I'll tackle that over lunch"! ;^^)
-RTC
I'm glad to see the WEIRD chartering process moving along...
Let me first post a disclaimer; this might be "coloring outside the lines"
of the charter as it is being written. But what the hey, this is an
exchange of ideas, right?
Picking up on Chris' idea about WEIRD putting IETF work in context with
related standards efforts, we could start by making up an indexed set of
abstracts. I am imagining structured, compact descriptions of the various
works-in-progress in IETF - and elsewhere (the "rest of world" area).
...You could imagine a table looking something like:
STD_BODY, STD_CMTE, STD_NAME, DATE, KEYWORDS, ABSTRACT...
The announcements on Internet-Drafts might be a good source to begin with,
as they usually include a reasonably succinct summary paragraph including
key terms. We'd just have to be careful to capture the semantic content of
"does not address" or "is out of scope" references (usually in the last
sentence).
This idea leads me to post two questions to the list:
1. Does one of our members have a Perl "crawler" handy, that could walk the
Internet-Drafts archive and extract/decompose these paragraphs?
2. Can we jointly compile a list of similar document archives from other
standards bodies that might be subject to the same approach?
The framework suggested above could then be "mined" to generate the view of
relationships and threads that is being discussed as the WEIRD value
proposition.
If we want WEIRD to relate IETF and "ROW" then it seems to me this calls
for a self-maintaining reference source. If we try to do this with the
"person-to-person" approach, we would need a voluntary contribution from a
member of every standards committee... On the other hand, the use of a
tool as described above would hopefully avoid the duplication of effort
involved in each of us digging through various archives.
I've taken a "bash" at the search.ietf.org tool; it presents a rifleshot
capability. You *can* find what you're looking for - if you know what it
is! But it's a place to start...
We could pull together a "candidate thread collection" with a few queries
from this resource. This might be a way to quickly generate an answer to
Chris's request, so succinctly described in the example "FOO" IAB
functional goal.
So, I submit that one foreseeable goal of a WEIRD information resource
would be to provide a flexible, but structured search capability (with
examples) to allow individuals to discover related topics as they emerge.
This would be possible given a...
"dynamically compiled searchable meta-standards archive"
"Sure, I'll tackle that over lunch"! ;^^)
-RTC