Re: Surfing the Web Backwards

Dear Kevin McCurley:

I am pleased to inform you that your paper named above has been accepted by the WWW8 Program Committee for presentation at the conference and inclusion in the Proceedings. You will be receiving more information on how to prepare the final version shortly.

Reviews for your paper are attached. Please be sure to read them carefully and address all the points raised in the reviews in your final version.

Congratulations on behalf of the WWW8 Program Committee. I look forward to seeing you in Toronto.

-Alberto Mendelzon
Program Committee Chair.

447

Surfing the Web Backwards

Soumen Chakrabarti, David A. Gibson, and Kevin S. McCurley

Report 1 for paper 447

Relevance: Yes
Correctness: Yes
Length and Content: Yes
Presentation: Accept (Comparable to papers in the best conferences)
Originality: Weak Accept (I vote acceptance, but won't argue for it)
Technical Depth: Accept (Comparable to papers in the best conferences)
Practical Impact: Accept (Comparable to papers in the best conferences)

Comments for author

o finish your references (eg. first paragraph!)
o need some work explaining the link between citations and the
bidirectional links. For example, they are *not* the same in that
you can't "trace backward in time through the evolution of ideas
that leads to the current work" nor forward in time sensibly (I
would argue) as links can be created and deleted so easily. I'd
like some more work linking the two concepts if you're going to use
it as a reason for the usefulness of what you're doing. I agree
with what you're saying, just that you need to strengthen that
link if you are going to use it.
o You gloss over the barriers too easily I fear. The the
"...scientific community generally endorses open commentary..."
says nothing about the problems of getting it to happen!
o This is probably more of a position paper to W3C rather than a
conference paper per se. However, in the past some of the more
useful changes have come about in ways other than submissions to
W3C so I won't hold this against it...
o The paper is also too long (IMHO) for this conference. I'm not
sure what I would recommend cutting so I'll leave it to the authors
but it should be culled a bit.

Report 2 for paper 447

Relevance: Yes
Correctness: Yes
Length and Content: Yes
Presentation: Accept (Comparable to papers in the best conferences)
Originality: Accept (Comparable to papers in the best conferences)
Technical Depth: Accept (Comparable to papers in the best conferences)
Practical Impact: Accept (Comparable to papers in the best conferences)

Comments for author

A very interesting and thought provoking paper!

My only concern is that there is quite a bit of related work missing (mainly from the hypertext community). Although you have mention Xanadu you get the historical perspective wrong. Ted Nelson wrote in detail about Xandau in his book "Literary Machines" which he first published in 1981 (although the ideas pre-date this considerably). Also, although ZigZag can be said to have been derived from his ideas on Xanadu, it is certainly not a prototype of it, as I understand it is a new system. But more importantly you've missed out a whole family of WWW distributed link services (eg. Southampton's DLS, Pitkow and Jones' Atlas) from your introduction. These also provide backlink functionality, albeit by different means.

References:

Leslie Carr, David De Roure, Wendy Hall and Gary Hill, "The Distrubuted Link Service: A Tool for Publishers, Authors and Readers", in Proceedings of the Fourth International WWW Conference, 1995.

James E. Pitkow and R. Kipp Jones, "Supporting the Web: A Distributed Hyperlink Database System", in Proceedings of the Fifth International WWW Conference, 1996.

Finally, you mention bidirectional linking, but fail to mention the Hyper-G/Hyperwave system, in which all linking is bidirectional, and if you use a Hyperwave Web server, back linking comes free!

Report 3 for paper 447

Relevance: Yes
Correctness: Yes
Length and Content: Yes
Presentation: Accept (Comparable to papers in the best conferences)
Originality: Accept (Comparable to papers in the best conferences)
Technical Depth: Strong Accept (As good as the top 10% papers in the best conferences I know)
Practical Impact: Strong Accept (As good as the top 10% papers in the best conferences I know)

Comments for author

A well-written and very readable paper about adding "backlink"
functionality to Web servers (and, thus, Web sites). Provides
reasonable and feasible implementation information, explaining
required HTTP extensions and additional MIME types. In all, a very
pleasing and seemingly comprehensive treatment of the topic.

The obvious privacy issues (e.g. relating to the use of the "referer"
field) are adequately discussed. The usefulness of backlink information
is illustrated with "anecdotal" evidence from user studies.

A few minor corrections or suggestions for clarification:

In the second line of the infroduction, the editorial remark
"(reference to Xanadu?)" should be replaced with the actual
reference.
In the paper, RDF is transcribed "Resource Description Framework",
while thje references section refers to the "Resource Definition
Framework".
In section 3.2, the syntax on the defition of x-backlinks information
does not seem to match the syntax of the examples given (different
order of mime-type and URI).