WHICH CONTRIBUTION OF THE WEB SERVICES IN THE
IMPROVEMENT OF WEB SEARCHING?
A BEHAVIOURAL STUDY OF THE NET SURFERS
Christian Belbèze and Chantal Soulé-Dupuy
IRIT, Université Toulouse 1, 2 Rue du Doyen-Gabriel-Marty, 31042 Toulouse Cedex 9, France
Keywords: Information access, Web searching, Internet surfer behaviour, behavioural study, Web Services.
Abstract: The difficulty of finding information on the Web grows and this even for the most experts of us. In order to
better understand how the Net surfers search, we observed five adults and four children. Two different
protocols of observation, both presenting imposed and free searches, were defined for the children and the
adults. Thus, we have defined a certain number of behaviours, attitudes and difficulties, and then we
classified them. The result of these observations as well as the analysis of the behaviours are presented in
this paper. The main goal of this paper is to introduce the contribution of Web services in the improvement
of information access and Web searching.
1 INTRODUCTION
While the World Wide Web contents offer more and
more each day, finding an answer or a piece of
information becomes increasingly difficult. Before
thinking about modifying the solutions of research
suggested to the Net surfers, it appeared significant
to observe them. The search engines are classically
tools not guiding the Net surfer. According to his
knowledge of search engines mechanisms and of his
own capabilities (at the same time on the required
subject and on the information format), Net surfers
provide different reference frames of keywords. In a
query without result, for example, an experienced
Net surfer who is speaking English will quickly
widen his research by the use of keywords in this
language. But this extreme case is rare in a general
public context. The search engines accept any
reference frame of keywords and generally turn over
thousands or million sites and it is of course there
that the difficulties for the Net surfers start. Which
adequacies and which relevance represent these
proposals? How do Net surfers decode the few lines
proposed to determine the contents of the page?
How do they read then these pages once they
selected them? To answer these questions, to
imagine the Web of tomorrow and its information
retrieval systems, we selected nine people from eight
to sixty twelve years old, including five female,
from different sociocultural backgrounds and we
observed them.
According to these observations, our goal is to
propose some solutions to improve information
access. Thus, our work aims at considering the
contribution of Web services to help Net surfers in
Web searching.
In this paper, we provide some related work.
Then, we report the experimentation and the results
of the observations. And in the last section we
present the architecture that we propose to help Net
surfers in Web searching by using selected services.
2 RELATED WORK
Since the beginnings of the documentary systems,
and in particular since the work undertaken by
Gerald Salton (Salton, 1970) (Salton & McGill,
1983), information retrieval enormously evolved.
The evolutions were not carried on the models of
information retrieval themselves, but rather on the
contexts of use, the Net having upset the practices
and the needs as regards access to information.
Many studies have lead to the same conclusion
that information retrieval systems (more particularly
on the Web) provide too disappointing results
(Jansen & Al., 2000) (Savoy & Picard, 2001),
(NachMias & Gilad, 2002), (Laurie, 2005).
129
Belbèze C. and Soulé-Dupuy C. (2007).
WHICH CONTRIBUTION OF THE WEB SERVICES IN THE IMPROVEMENT OF WEB SEARCHING? A BEHAVIOURAL STUDY OF THE NET
SURFERS.
In Proceedings of the Ninth International Conference on Enterprise Information Systems - SAIC, pages 129-137
DOI: 10.5220/0002387901290137
Copyright
c
SciTePress
Traditional information retrieval models prove to be
ineffective because of the growing mass of
information. Even the valuation campaigns, such as
TREC (http://trec.nist.gov/), CLEF (www.clef-
campaign.org), show the limits of experimentations
feasibilities. Improvements might then come from
new architectures including new technologies such
as Web services and from researches led in the
context of the Semantic Web: http://www.w3.org.
3 THE EXPERIMENTATION
This section describes an observation of nine users
searching the web: All measurements, details and
videos are present on the site http://sissiprojet.free.fr.
3.1 Protocols
Two protocols have been defined, one for the five
adults and one for the four children. Both protocols
mix imposed research subjects and free research
subjects. A research is a set of reading tasks and a
set of search requests using search engines. A new
request will be considered to each search engine
search button click.
3.1.1 Adults Protocols
The adult protocol is constituted of five researches:
Song lyrics “All blues”;
The distance between two French town;
A music score (Romania hymn);
Two free research subjects.
3.1.2 Children Protocol
This protocol is constituted of two researches:
The dolphin (life expectancy);
One free research subject.
3.1.3 Notes and Measurements
The nine users give four notes (from 0 to 10), at the
end of each research.
Capacity to understand the met information;
Interest of the visited site;
Subjective Time length;
General feeling.
For each research we measure a list of indicators
Subjective Time length;
General feeling.
Research duration;
Number of visited sites;
Number of supplementary read pages;
Success or failure of the research;
Number of requests (number of times the search
engine has been requested).
And for each request we recorded
Used keywords;
Number of keywords;
Number of found sites;
Whether the request multi-lingual or not.
3.2 The Users
Each user introduced himself/herself completely on
the site http://sissiprojet.free.fr. We try to choose
users with different ages, genders and social levels.
Users have accepted the use of their images in a non
commercial use and in the limit of the
experimentation and analysis of this research.
3.2.1 The Five Adult Users
Jean
Age: 16
Use: Daily
Experiment: > 5 years
Maximum acceptable search time: 15 minutes
Trade: student or school
Comment: He is the referent. He uses internet more
than often to search, to mail or to use instant
messaging programs, since 1995! He is part of the
Internet generation!
Annie
Age: 72
Use: Monthly
Experiment: > 5 years
Maximum acceptable search time: 25 minutes
Trade: Director (retired)
Comment: Annie only used Internet with a coach.
Here are her first steps in solo.
George
Age: 49
Use: Never
Experiment: None
Maximum acceptable search time: 15 minutes
Trade: Craftsman
Comment: George is really a complete beginner but
his interest for motorbikes and the brand Moto-
Guzzi is a good motivation.
Marie
Age: 37
Use: Rare
Experiment: lower than one year
ICEIS 2007 - International Conference on Enterprise Information Systems
130
Maximum acceptable search time: 15 minutes
Trade: Liberal profession
Comment: “When I do not find I ask my children. I
sometimes have big trouble to find”. On the other
hand for my emails, it is impeccable! ”. Marie is an
artist by passion and scientist by profession.
Yasmina
Age: 20
Use: Monthly
Experiment: superior to five years
Maximum acceptable search time: 15 minutes
Trade: Employee
Comment: Is Yasmina a girl who only uses Internet
to make research on lyrics? Yasmina in spite of her
engagements will not continue the protocol; thus we
will have only the imposed requests.
3.2.2 The Four Children
Guillaume
Age: 11
Use: Weekly
Experiment: between one and five years.
Maximum acceptable search time: 15 minutes
Trade: student or school
Comment: Guillaume loves music and computers.
He is a data processing passionate. Patient, posed, he
will find what he seeks.
Celine
Age: 11
Use: rare
Experiment: less than one year
Maximum acceptable search time: 15 minutes
Trade: student or school
Comment “I never went on Internet to seek all
alone…” Celine is a dreamer but a child who
however does not miss tenacity.
Lou
Age: 7
Use: Weekly
Experiment: between one and five years
Maximum acceptable search time: 15 minutes
Trade: student or school
Comment: Lou works well at school. She always
wants to do her best.
Paul
Age: 7
Use: Rare, he is a real beginner.
Experiment: Never
Maximum acceptable search time: 15 minutes
Trade: student or school
Comment: Paul is often ”on the moon”.
3.2 Measurement
The 9 users completed 21 researches composed of
113 requests for a total of 829825484 returned sites.
3.3.1 Research Measurement
Table 1: The research: Time and number of requests by
research
*/* Time Requests
Max 01:03:30 20
Average 00:18:45 4.5
Min 00:01:30 0
3.3.2 Request Measurement
Table 2: The search request Number of keywords and
number of sites returned by request.
*/* Keywords Found sites
Max 8 230000000
Average 3.44 7343588.
Min 1 0
Table 3: The navigation – Number of read sites, number of
read pages in these sites, and number of search engine
consulted pages.
*/* Sites Pages Search engine pages
Max 44 17 9
Average 8.73 3.53 2.6
Min 1 1 0
Table 4: Search success This tab is done by users in
concordance with usage and experience.
Users Usage Experience in year %found
Annie
rare
>5 40
Céline
monthly
<1 50
Georges
never
0 40
Guillaume
weekly
between 1 and 5 100
Jean
daily
between 1 and 5 60
Lou
never
between 1 and 5 50
Marie
rare
< 1 year 60
Paul
never
0 50
Yasmina
monthly
> 5 33
Average
N/A N/A
53,3
3.3.3 Keyword Measurement
With 3.44 we can notice that the number of
keywords increased from precedent observations.
But even with an increase of the number of
keywords, the average number of found sites is
enormous. The average number of returned sites,
using mainly Google was up to 7 millions. So we
can affirm that keywords used in the observation are
WHICH CONTRIBUTION OF THE WEB SERVICES IN THE IMPROVEMENT OF WEB SEARCHING? A
BEHAVIOURAL STUDY OF THE NET SURFERS
131
not efficient enough and compel to filter the mass of
information in the web.
Requests number / keywords number
1
10
100
0246810
Number of keywords
Request Number
Figure 1: Groups of dots – number found sites/ keywords
number.
Most requests are using between two and 6
keywords (Figure 1).
Figure 2: Groups of dots – Request number / Keywords
number.
The found sites number decrease with the
increase of keywords until the number of 6
keywords, we can notice an inversion of this
tendency for highest value (Figure 2).
3.4 Users Feeling
With an average general satisfaction note at 6.25 the
feeling on internet as information base and on its
capability to find information could be improved.
Results are very different from a user to another
(
Table 5).
The exercise was based on the research of
information. The users play the game so they were
very impacted by the fact that they found the answer
or not. The real time spent on the research does not
match the general note.
Table 5: Average satisfaction note by user (from 1 to 10).
Users Readable Interest Time General
Annie 7.4 6 7 6.3
Céline 7.5 6 4 4
Georges 5.5 5.5 6.5 6
Guillaume 7.5 7 7.5 7
Jean 8.4 6.5 5.8 6.3
Lou 2 3.5 5 5.5
Marie 4.8 5.4 5.2 5.7
Paul 7 10 5 10
Yasmina 7 5.75 7.25 6.5
Max
10
10 10 10
Average
6.41
6.05 6 6.25
Min
0
0 0 0
Figure 3: Groups of dots – Real Search Time / Note.
Even the time perception does not match the real
time. But the fact to find or not is in correlation with
all notes.
Figure 4: Groups of dots – Real Search Time/Time cost.
Search result/ General Note
-0,5
0
0,5
1
1,5
2
2,5
3
3,5
0246810
Note
Success / Fai l ur e
0 : faillure - 1 : some elements - 2 : more elements - 3 : succes
Figure 5: Groups of dots – Search Success or Failure
/general Note.
ICEIS 2007 - International Conference on Enterprise Information Systems
132
Time perception and site interest seem to be
more in correlation with the success or the failure of
the research than anything else. We have to read the
users notes very carefully. Users were obsessed with
finding answers. If they found them, it was short and
interesting if not it was too long and boring.
Figure 6: Groups of dots – Search Success or Failure /
Subjective Time Cost.
Search result/ Sites interest
0
2
4
6
8
10
12
0246810
Sites Interest
Success/Fai lure
0 : faillure - 1 : some elements - 2 : more elements - 3 : succes
Figure 7: Groups of dots – Search Success or Failure
/Subjective site interest.
3.5 The Users’ Difficulties
The users’ difficulties are located on each basic task
that composed the internet information search task.
Determine the efficient keywords in charge of
filtering the web and use words in non mother
tongue Languages ;
read and choose in the list results including URL,
acronyms and different languages mixed;
extract the information from an object;
manage research time and time perception
3.5.1 Difficulties to Find the Efficient
Keywords
With an average of found sites number up to 7
millions, we can certify that the 3.44 keywords are
not efficient enough to filter the Web. Users have
difficulties to use more keywords and to choose
these words. This is a problem of search nature user
competences: i.e. users do not know the subject well
enough to enrich the request. It will also be a mix
between indolence and confidence in search engine.
The user does not manipulate words in an
unknown language. Looking for rumanian music
information, nobody has used rumanian keywords
(even if translation tools were known). Only Jean
will use the song words to found the lyrics in
English (Jean is bilingual French-English).
Jean has mixed French and English, using the
word “resume” in French to mean abstract.
Keywords are selected in concordance with her
or his knowledge of the subject, and the knowledge
of the search engine. We also have to consider that
keywords could be use with different spellings.
More and More young people use SMS. During
the observation children will find information using
SMS language (by mistake?) in their request.
3.5.2 Choose in the list of Proposed Sites
The list returned by search engines is constituted of
some elements, Here Google is the sample: the site
title; the snippet (a part of sentences including the
keywords) and the URL.
Site title: In the site the title is very often non
present. In this case Google chose other meta tags
as "title" "author” and build a sort of title. If the title
is too long Google chooses to cut and present only a
part.
Snippet: This is just a few words around
keywords extracted from the site. The extract can
include any words from any part of the site. There is
no global sense in the snippet when code is not
mixed in it.
The URL: Uniform Resource Locator. See
“RFC 1630
We can imagine, a beginner looking for this
information: The title has no sense, the Snippet is
just a list of sentence extracts, and the URL Looks as
binary code. For example, Annie who is 70 years has
asked Google on « natura 2000 marais de
gabarret » . Natura 2000 is a French ecologic
organisation, “marais” means marsh and “gabarret”
is a small village in France. Google sends back this!
Figure 8: Google typical sites list.
WHICH CONTRIBUTION OF THE WEB SERVICES IN THE IMPROVEMENT OF WEB SEARCHING? A
BEHAVIOURAL STUDY OF THE NET SURFERS
133
In figure 8, the title is in English, the snippet in
French speaking about the Eiffel tower and the URL
is an IP address. Annie had chosen to burst out
laughing.
3.5.3 Extract Information from the Object
From the page: It is difficult for the users to get
information from the page. All the users have missed
information on the page. It could be because:
they stop reading the page before the information
was displayed , the page is too long;
they just read the posted part of the page;
the page was too loaded, especially with
advertising;
they used a search function on a site instead of
using a Web search engine (Quid, Amazon, …);
they always used the same link instead of read
the read the full page;
they stopped reading if the page or part of it is in
an unknown language.
From a file: Jean was looking for a music
partition. He found a midi file. But he did not know
that it is possible to extract the partition from it.
3.5.4 Manage Research Time and Time
Perception
The majority of users have declared that the
maximum research time is around 15 minutes. Annie
has given 25 minutes. With an average search time
of 18 minutes and 45 seconds and 50% of the
research up to 15 minutes this is a reason of
disappointment. If research time is less important
than success or failure in the average general note on
the research up to 15 minutes is only 5.5.
3.6 Observation Conclusion
Internet users are disappointed by the search
process. They have difficulties to communicate
efficient keywords and to extract information from
the found sites list and from the site itself.
Following the observations, the questions and
the difficulties of the Net surfers, it appears that:
the choice of the keywords in a semantic field
does not seem sufficient to bring back a
reasonable number of sites and to guarantee a
scheduling related to this semantic field
the Net surfers cannot carry out requests in
foreign languages;
the Net surfers do not understand the texts turned
over by the search engine;
if the turned over texts are in a foreign language;
the situation can be even lived as disappointing;
the Net surfers cannot extract information from a
starting format containing it.
The implementation of an “intelligent” system of
simplification of the services of Internet based on a
better knowledge of the Net surfer, of his tastes,
competences and resources could contribute to a
better adaptation of the turned over sites. The taking
into account of a machine translation of the
nonknown languages of the Net surfer towards
known languages, the indication of the contents and
nature of the sites by a system of icons would
reassure this one. The proposal for a change of
format or nature of information by Web services is
also a possible solution to help all one each one with
better exploiting the layer of knowledge of the Web.
4 SEARCH AND INFORMATION
AQUISITION INDIVIDUAL
HELP BY USING SELECTED
WEB SERVICES
4.1 Identify Individual Capabilities and
Knowledge Including Handicap
Today users are not included in the search process in
term of capabilities, knowledge and taste. Search
engines work in the same way for very different
people.
The general concept of this project is to propose
web services to a user. The goal is to help him or her
in the search process.
To include the user in the process, we define a
user as a set of resources. These resources recorded
in a profile, can be divided in two main parts:
Communication resources;
Knowledge resources.
Communication resources: these resources
describe the capability as sight or hearing of the
users.
Knowledge resources: these resources describe
the knowledge of the user like languages, music
reading or practice, math, Internet knowledge etc …
Handicap: These capacities and knowledge
could be less than the standard or average ones.
Can we speak of handicap?
Yes, but the handicap is not where we can think
it is. In searching the web a light sight or hearing
handicap could be less penalizing than the fact that
you don’t speak English, or you do not know how to
read URL.
ICEIS 2007 - International Conference on Enterprise Information Systems
134
In this way, we are all in trouble when facing the
internet. Who can say “I speak all the languages, I
know how to read music partition, I know enough
math to read all demonstrations, I understand the
SMS language and I can read PDF documents from
binary. So I do not need any software to do it”?
In reality when facing massive and diverse
information we are all handicapped.
We will add in this profile a list of proffered
subjects or user tastes. This is an important element
to help user to search filtering the web. The full user
profile including his/her tastes, experiences,
recorded researches; used keywords will be saved, in
anonymous way, to have the have the best
knowledge of the user.
4.2 Offering Services to Help User to
Be More Efficient and Less
Disappointed
We can propose solutions to help users in their
difficulties. These solutions have to be individual.
Each user needs adaptive solution. These solutions
could be very complex to implement. The idea is to
create or use for each solution an independent Web
service(s). Trying to start from this observation we
can imagine a list of services to help our users.
4.2.1 General Services to Help Users
In chapter 3.4 we identified ours user’s difficulties.
We can propose solutions to help then. These
solutions have to be individual. Each user needs
adaptive solution.
Which services to help users to determine the
efficient keywords in charge of filtering the web?
a service to build smart search requests based on
user keywords, on previous requests and on
semantic space. The semantic space is
determined by the knowledge of the user taste
and previous researches. These requests could
filter the web more (efficiency) efficiently.
a service translating Keywords from SMS to
official orthographic and from official
orthographic to SMS ;
a service to translate keywords to a specific
language and an automatic translation of the
results;
a service to collaborate with others users by
sharing request and resources;
a service to correct and create spelling mistakes
in order to widen researches.
Which services to help users to read and choose
in the list results?
a service to give an explicit reading of URL;
a service to give a more detailed snippet;
a service to add icons to characterize the site
contents (shop, forum, encyclopaedia,..)
Which services to extract the information
from an object?
a service to highlight key-word.
a service to hide pictures and entertainments ;
a service to display a picture (map) of the page to
help to the localisation;
a service to extract files from any archive type a
web service to send by email the content.
Which services to help user to manage
research time and time perception?
A service to measure and control time with the
possibility to save and reuse the research key-
words and results.
a service to analyse found sites to help users to
modify the request;
4.2.2 Offering Services to Help
“Handicapped” Users
These web services could be use to help people with
accessibility troubles. For example we can imagine
web services:
for people with myopia trouble a service
guaranteeing a minimal font size to display
pages;
For colour-blind people a service correcting
colours to get the best readable pages;
for people having understanding handicap a
service placing a link to definitions for all
detected words as “difficult words”;
for blind people a service reading out loud the
site.
4.2.3 Offering Services to Modify
Information Format
These Web services can extract, rebuild, complete
information from a start-up document.
For people with a specific editor a web service to
transform files to the right editor format;
for musicians extract partitions from midi file;
4.3 Defining an Open Structure Project
to Record and Select Web Services:
SISSI
4.3.1 Fundamentals
The goal is to define an open structure project to
record and propose selected Web Services in charge
of helping users to search and to adapt and complete
WHICH CONTRIBUTION OF THE WEB SERVICES IN THE IMPROVEMENT OF WEB SEARCHING? A
BEHAVIOURAL STUDY OF THE NET SURFERS
135
data nature and format to the users competency;
knowledge and taste.
Articulated around a central service called SISSI:
Individualized System of simplification of Internet
Services (Figure 9). Web Services definitions and
users definitions are stocked in a database called
Sissidb.
Based on UDDI the Web services recorded in the
base Sissidb present their services to the user who
can select and parameter them in concordance with
his/her needs and tastes.
4.3.2 SISSI and Web Services
Communication
SISSI does not include any Web services. It is a
connector between us. SISSI could be viewed as a
generic web service client or as a generic web
service IHM.
Based on UDDI web services are described more
precisely in the base in the way that the user can
understand the result they are expecting, and SISSI
can communicate with then.
SISSI communication is based on SOAP and we
know that SOAP is a heavy protocol. But the Web
information manipulated by SISSI is mainly HTML
pages or active code used on the client side. The
complex objects as pictures stay on the Http server.
If a Web service needs one of these objects it will be
its responsibility to load it from the initial resource
server. (In this way) big objects manipulated by
SISSI are not to be loaded.
SISSI is here including search engine service but
it could be invoked directly on a document by
receiving an URL as argument.
5 CONCLUSION
The first feasibility works reveals a certain
complexity. If the architecture model seems to be
usable, we found some new challenges:
The “real life” HTML language is not
normalized. Working on it to modify a
document structure or presentation as generic
language is today a big work and due to this
HTML management complexity writing
efficient Web Services will be difficult.
SISSI is a service based on services. That
means that if a service falls down, SISSI falls
down. We will have to think “fault tolerance”.
Even if they use a pseudo, users have the right
to withdraw themselves from a base where
their competencies, knowledge, tastes and
handicaps are stored, especially if this base
records their activity.
However, as regard to the related works, the
observations and the experimentation carried on
with SISSI, we come to the conclusion that the use
of an open architecture based on Web services can
be a solution to help Net surfers in Web searching.
Indeed, it can make users aware from facing up to
certain situations and prevent them from giving up
the research, more particularly when they find some
difficulties to access or to exploit some retrieved
Figure 9: SISSI transactions architecture.
ICEIS 2007 - International Conference on Enterprise Information Systems
136
information (handicapped users, surfers who do not
speak any foreign languages …).
To conclude, services based architectures like the
one we have implemented (SISSI) must be freely
extensible in order to offer more and more. At
present we have to experiment this extensibility by
permitting to any user to add new services when he
judges it necessary. New observations could then be
carried on to analyse the capabilities of Net suffers
to interpret the results provided by the use of hidden
services.
Finally a better knowledge of the users’ tastes,
their competences, their contexts and their resources
will allow the implementation of communities to
share researches and results. This information could
be used to refine research by the creation of
semantic contexts of associates.
REFERENCES
Alexander, P.A., & Jetton, T.L. (2002). Learning from
text: A multidimensional and developmental
perspective. In M.L. Kamil, P. Mosenthal, P.D.
Pearson, and R. Barr (Eds.) Handbook of reading
research, volume III (pp. 285-310). Mahwah, NJ:
Erlbaum.
Berkowitz, B. (2002). Moving every child ahead: The
Big6 success strategy. Multimedia Schools, 9, 17-22.
Bransford, J.D., Brown, A.LO., & Cocking, R.R. (Eds.).
(2000). How people learn: Brain, mind, experience,
and school. Washington, DC: National Academy
Press.
Coiro, J. L. (2003). Rethinking comprehension strategies
to better prepare students for critically evaluating
content on the Internet. The NERA Journal, 39(2), 29-
34.
Eagleton, M.B., Guinee, K., & Langlais, K. (2003).
Teaching Internet literacy strategies: The hero inquiry
project. Voices from the Middle, 10, 28-35.
Gay, L.R. & Airasian, P. (2000). Educational research:
Competencies for analysis and application. Upper
Saddle River, NJ: Prentice Hall.
Guinee, K., Eagleton, M.B., and Hall, T.E. (2003).
Adolescents' Internet search strategies: Drawing upon
familiar cognitive paradigms when accessing
electronic information sources. Journal of Educational
Computing Research, 29, 363-374.
Henry, L. A. (2005). "Information search strategies on
the Internet: A critical component of new literacies".
Webology, 2(1), Article 9. Available at:
http://www.webology.ir/2005/v2n1/a9.html
International Reading Association. (2002). Integrating
literacy and technology in the curriculum: A position
statement. Retrieved April 22, 2005, from
http://www.reading.org/downloads/positions/ps1048_t
echnology.pdf
Janesick, V. J. (1994). The dance of qualitative research
design: Metaphor, methodolarty, and meaning. In N.
K. Denzin & Y. S. Lincoln (Eds.), Handbook of
qualitative research (pp. 209-219). Thousand Oaks,
CA: Sage Publications, Inc.
Jansen, B. J., Spink, A., & Saracevic, T. (2000). Real life,
real users, and real needs: a study and analysis of user
queries on the web. Information Processing and
Management, 36, 207-227.
Laurie, A. H., (2005). Information Search Strategies on
the Internet: A Critical Component of New Literacies
University of Connecticut, Connecticut, U.S.A.
Leu, D.J., Jr. (2000). Our children's future: Changing the
focus of literacy and literacy instruction. The Reading
Teacher, 53, 424-431.
Leu, D.J., Jr., Kinzer, C.K., Coiro, J., & Cammack, D.W.
(2004). Toward a theory of new literacies emerging
from the Internet and other communication
technologies. In R. Ruddell and N. Unrau (Eds.)
Theoretical models and processes of reading (5th ed.).
Newark, DE: International Reading Association.
Nachmias, R. & Gilad, A. (2002). Needle in a hyperstack:
Searching for information on the World Wide Web.
Journal of Research on Technology in Education, 34,
475-486.
O’Sullivan, M., Scott, T., Ph.D March/April 2000
Teaching Internet Information Literacy: A Critical
Evaluation
http://www.infotoday.com/MMSchools/mar00/osulliv
an&scott.htm
Pressley, M, & Afflerbach, P. (1995). Verbal protocols of
reading: The nature of constructively responsible
reading. Hillsdael, NJ: Erlbaum.
RAND Reading Study Group. (2002). Reading for
understanding: Towards and R&D program in
reading comprehension. Retrieved March 3, 2004.
Available Online:
http://www.rand.org/multi/achievementforall/reading/r
eadreport.html
Salton, G. (1970). The SMART Retrieval System.
Englewood Cliffs, NJ: Prentice-HiII
Salton, G., & McGill, M.J. (1983). Introduction to Modern
Information Retrieval. New York: McGraw-Hill.
Savoy, J., Picard, J. (2001). Retrieval effectiveness on the
Web. Information Processing and Management, 37,
543-569.
Sutherland-Smith, W. (2002). Weaving the literacy Web:
Changes in reading from page to screen. The Reading
Teacher, 55, 662-669.
Williams, P, & Nicholas, D., (2006) Testing the usability
of information technology applications with learners
with special educational needs (SEN)
http://www.blackwell-synergy.com
/doi/full/10.1111/j.1471-3802.2006.00057.x
Yin, R.K. (1989). Case study research: Design and
methods (2nd. ed). Sage: Newbury Park, CA.
WHICH CONTRIBUTION OF THE WEB SERVICES IN THE IMPROVEMENT OF WEB SEARCHING? A
BEHAVIOURAL STUDY OF THE NET SURFERS
137