webcaching.guerrerotome.com

Dear Carlos Guerrero,

********************************************
This email contains missing referee comments about your paper named below.
We kindly apologize for the missing referee comments at the previous email you recieved.
********************************************

We regret to inform you that your paper

Proposed Architecture for a Dynamic Web Fragment Caching System
Carlos Guerrero
Carlos Juiz
Ramon Puigjaner

has not been accepted for publication in the proceedings of the 22nd International Symposium on Computer and Information Sciences (ISCIS-07). Referee comments to your paper are at the end of this email.

We had a high number of high quality submissions and due to limited space and the outcome of the reviewing process
we were not able to include your paper in the program. However, you can still submit your work as a poster presentation
details of which is announced at the following link:

http://www.iscis.metu.edu.tr/callforposters.htm

Please find enclosed the reviews of your paper. We hope that you will find them useful in revising your paper
for future submission. Thank you for choosing to submit to ISCIS 07.

Best regards,

ISCIS 2007 Organization Committee

********************************************
Referee Comments to Your Paper:

Reviewer 1:

Recomendation : Reject: <i>Content inappropriate to the conference or has little merit</i>

Category : Tends towards theoretical
Values :
* Not much

Referee Familiarity : High
Best Paper Candidate : No
Length Appropriate : Yes
Difference From Previous Works on the Same Area : Don’t know

Referee comments to Author:
Except for the parts copied from the web, this paper is poorly written, and includes multiple grammatical errors. In times, the witing style makes it very difficult to understand.

The idea presented is interesting, but not a breakthough. It is anyway just a position statement, not an actual result. As such, one would expect much more discussion and deliberation.

It is unethical to copy parts of other public documents into your paper (this is called plagiarism).
———————————————–

Reviewer 2:

Recomendation : Clear Accept: <i>Content, presentation, and writing meet professional norms; improvements may be advisable but acceptable as is</i>

Category : Tends toward practical
Values :
* New perspective, issue, or problem definition

Referee Familiarity : Moderate
Best Paper Candidate : No
Length Appropriate : Yes
Difference From Previous Works on the Same Area : Don’t know

Referee comments to Author:

———————————————–

Reviewer 3:

Recomendation : Marginal Tend to Reject: <i>Not as badly flawed; major effort necessary to make acceptable but content well-covered in literature already</i>

Category : Highly practical
Values :
* Clarity to present understanding
* Not much

Referee Familiarity : Low
Best Paper Candidate : No
Length Appropriate : Yes
Difference From Previous Works on the Same Area : Don’t know

Referee comments to Author:
The authors propose an architecture for caching of web documents using ESI, which fragments the content of dynamically created web documents.

While the architecture is described in detail, it is unclear what the actual contribution of this paper is. Additionally, the paper is not well written and the English grammar and spelling should requires some rework. Furthermore, the paper is in my opinion not well structured and the benefits of improving the web performance through this method is not shown in any way by simulation or measurements.
———————————————–

Reviewer 4:

Recomendation : Probable Reject: <i>Basic flaws in content or presentation or very poorly written</i>

Category : Balanced theory and practice
Values :
* New perspective, issue, or problem definition
* Not much

Referee Familiarity : Low
Best Paper Candidate : No
Length Appropriate : Yes
Difference From Previous Works on the Same Area : Don’t know

Referee comments to Author:
Some parts of the paper is copied directly from esi web site without a reference.
The use of English in the paper is not satisfactory and makes it very difficult to read.
There are no performance results.

———————————————–

Dear Carlos,

We are sorry to inform you that your paper submitted for the AEWSE
‘08 workshop could not be accepted for presentation/publication. Reviews are inc
luded at the end of this message. We hope that the comments by reviewers can hel
p you evolve and improve your future research.

AEWSE’08 organizers.

REVIEW1:
O
verall rating:  0 (borderline paper)
Confidence:  1
Review:  I found the subje
ct of the paper interesting and novel
though not completely related to the core
of the workshop’s theme.
However this is not my main concern regarding this pape
r:
I found that most technical detail in the description of the architecture
is
needed, particularly it would be interesting that the authors
add some report on
experiments or potential scenarios of use
of the proposed architecture

REVIE
W2:
Overall rating:  -2 (reject)
Confidence:  3
Review:  The paper presents th
e idea of using ontologies together with reasoners to re-configure web applicati
on systems to improve performance. The adaptive part of the web system is theref
ore the configuration rather than any aspect of behaviour or information deliver
y. In this respect, I think the relevance to the workshop is borderline.

The ma
in problem with the paper is the lack of detail, especially as the main idea of
using ontologies to support adpative performance has already been published by t
he authors in other papers. On its own this paper, fails to really describe “why
” ontologies are needed, “what” form these ontologies take and “how” these ontol
ogies are used. I comment on each of these in turn. The authors do not really mo
tivate why they use ontologies rather simply than using databases to store the d
ata that they use. While they refer to it as a “knowledge base” and write about
the ability to use reasoners, they give no information about the precise nature
of the data stored and the processing of the data used. Hence the key feature of
their system that perhaps distingusihes it from other systems that adapt automa
tically to meet performance requirements which is the use of ontologies is not w
ell justified. Further, there is really no information about the ontologies them
selves. Statements such!
as “Each of theses ontologies has he information how
the system works.” are far too vague. The authors would be better to write less
about general background in terms of the fact that web sites are dynamic nowaday
s and the semantic web etc. and instead use the limited space to really explain
the forms of knowledge stored. Last but not least, they state that “reasoners co
uld be used” but don’t explain what they would do or how they would work. What e
xactly would be the output of such a system? The impression given is that it is
all an idea and they haven’t really implemented anything or even done the detail
ed design. This may not be the case - but then they have to provide more detail
to be convincing.

There are also many mistakes in the language of the paper and
the presentation would need to be improved significantly. THe general impressio
n is that the paper was written in great haste and is far from being in a state
acceptable for publication even at a workshop.

REVIEW3:
Overall rating:  -1 (
weak reject)
Confidence:  4
Review:  The aim of this paper is to create a tool
that changes the web system setup and configuration by
using the global informa
tion of the system (e.g., users, servers, middle-tiers, performance, behavior).

I think the idea is interesting, however, the paper is poorly written, contains
many English errors.

On the scientific level, the paper lacks a motivation for
the chosen approach, a related work section and a critical evaluation. Actually
all three things are essential for any scientific paper to be acceptable.

Afte
r reading the paper the reader is still unclear about the technical motivation f
or the chosen architecture, i.e., why it is superior to others, where are the re
al gains and what lessons are to be learned from this approach. What about issue
s such as scaleability, performance, etc.?