Search

International audience; Le réseau Internet offre un gisement d'images numériques considérable. La localisation de ces images passe le plus souvent par l'utilisation d'outils de recherche spécialement dédiés à la recherche d'images. Si la littérature est assez riche en ce qui concerne les outils de recherche de pages web, peu d'articles ou d'études se sont attachés jusqu'à présent à décrire le fonctionnement des outils de recherche d'images. L'objectif de cet article est donc de comparer les fonctionnalités de ces outils de recherche, en distinguant ceux qui dérivent directement d'outils de recherche de pages web de ceux qui ont été spécialement créés et développés pour la recherche d'images sur Internet.

International audience; Purpose: Artificial intelligence (AI) has entered the field of medicine, and ophthalmology is no exception. The objective of this study was to report on scientific production and publication trends, to identify journals, countries, international collaborations, and major MeSH terms involved in AI in ophthalmology research.Methods: Scientometric methods were used to evaluate global scientific production and development trends in AI in ophthalmology using PubMed and the Web of Science Core Collection.Results: A total of 1356 articles were retrieved over the period 1966-2019. The yearly growth of AI in ophthalmology publications has been 18.89% over the last ten years, indicating that AI in ophthalmology is a very attractive topic in science. Analysis of the most productive journals showed that most were specialized in computer and medical systems. No journal was found to specialize in AI in ophthalmology. The USA, China, and the UK were the three most productive countries. The study of international collaboration showed that, besides the USA, researchers tended to collaborate with peers from neighboring countries. Among the twenty most frequent MeSH terms retrieved, there were only four related to clinical topics, revealing the retina and glaucoma as the most frequently encountered subjects of interest in AI in ophthalmology. Analysis of the top ten Journal Citation Reports categories of journals and MeSH terms for articles confirmed that AI in ophthalmology research is mainly focused on engineering and computing and is mainly technical research related to computer methods.Conclusions: This study provides a broad view of the current status and trends in AI in ophthalmology research and shows that AI in ophthalmology research is an attractive topic focusing on retinal diseases and glaucoma. This study may be useful for researchers in AI in ophthalmology such as clinicians, but also for scientists to better understand this research topic, know the main actors in this field (including journals and countries), and have a general overview of this research theme.

International audience; The purpose of this paper was to assess the presence of researchers on two author identi-fier services (ORCID and ResearcherID) and to compare the results with two academic social networks (Academia.edu and ResearchGate) using the categories of discipline, career advancement, and gender in a medium sized multidisciplinary university in France (University of Caen Normandy). Metrics such as number of publications per researcher, h-indexes, and average number of citations were also assessed. Of the 1,047 researchers studied, 673 (64.3%) had at least one profile on the four sites, and the number of researchers having multiple profiles decreased as more sites were studied. Researchers with only one profile numbered 385 (36.8%), while 204 (19.5%) had two, 68 (6.5%) had three, and only 16 (1.5%) had four. ResearchGate had by far the highest number of researchers present , with 569 (54.3%), whereas presence on the other sites was about 15%. We found that, apart from Academia.edu, researchers in Sciences, Technology, and Medicine (STM) were over-represented. Overall, experienced male researchers were over-represented on the sites studied. Our results show that, because of the numerous profiles lacking publication references (particularly on ORCID) and a low presence of researchers on the four sites studied (except for ResearchGate), assessing the number of publications, h-indexes, or average number of citations per article of individuals or institutions remains challenging. Finally, our data showed that French researchers have not adopted the use of the two author identifier sites (i.e. ORCID and ResearcherID). As long as French researchers remain reticent, these sites will not be able to provide the services for which they were created: addressing the problem of author misidentification, consequently providing exhaustive access to scientific production and bibliometric indicators of individual researchers and their institutions.

International audience; The emergence of the ‘‘Open Access’’ (OA) movement and OA repositories has upset (and still upsets) the economy and access to scientific publications. The aim of this article is to actualize and complement the results of previous studies that have attempted to quantify the importance of OA in the field of biology/medicine, through a focus on the PubMed bibliographic database. An analysis of OA publications in PubMed, based on the geographical localization of authors, was also conducted (countries and continents) and some parameters related to OA (evolution of the OA journals, the number of mandates and OA repositories established by countries and continents) have also been studied. The results show that thepercentages of articles available in OA are still growing and reach 39.1% of articles available in the PubMed database in 2015. The geographical analysis of the 25 most productive countries and continents shows a great variability in the percentage of articles in OA (from 21.9% for Italy to 42.08% for the USA and from 22.80% for Oceania to 40.84% for North America). Furthermore, our results show that the number of mandates and OA repositories is not significantly correlated with the percentage of articles in OA at the national and continental levels, confirming that the public policies or mandates in OA have had an influence below expectations. The establishment of more coercive mandates may succeed in obtaining more significant effects in the medium or long-term. The steady increase of the number of OA journals, with the acknowledged growth of the number of citations of articles in OA, will certainly boost the attractiveness of the authors for the OA; L’apparition du mouvement « open access » (libre accès, LA) et des archives ouvertes a bouleversé (et bouleverse encore) l’économie et l’accès aux publications scientifiques. L’objectif de cet article est de réactualiser et compléter les résultats des études antérieures qui ont tenté de quantifier l’importance du LA dans le domaine de la biologie/médecine, par le biais d’un focus sur la base de données bibliographiques PubMed. Une analyse des publications en LA dans PubMed en fonction de l’origine géographique des auteurs a également été menée (pays et continents) et un certain nombre de paramètres liés au LA (évolution du nombre de journaux en LA, nombre de mandats et d’archives ouvertes par pays et continents) ont également été étudiés et mis en perspective. Les résultats mettent en évidence que les pourcentages d’articles dont le texte intégral et disponible en LA ne cessent de progresser et concernent en 2015, 39,1 % des articles disponibles dans PubMed. L’analyse géographique des 25 pays les plus productifs et des continents montre une grande variabilité concernant le pourcentage d’articles en LA (de 21,9 % pour l’Italie à 42,08 % pour les États-Unis et de 22,80 % pour l’Océanie à 40,84 % pour l’Amérique du Nord). Par ailleurs, nos données montrent que le nombre de mandats et d’archives ouvertes n’est pas corrélé de manière significative au pourcentage d’articles en LA au niveau national et continental, confirmant ainsi que les politiques publiques successives ou les mandats relatifs au LA n’ont eu qu’une influence, sinon secondaire, du moins inférieure aux attentes. La mise en place de mandats plus coercitifs parviendra peut-être à obtenir des effets plus significatifs à plus ou moins long terme. L’augmentation régulière du nombre de journaux en LA, concomitante à l’augmentation avérée du nombre de citations des articles en LA, amplifiera certainement encore l’attrait des auteurs pour le LA.

International audience; This study examines the role of the academic affiliation in research evaluation. The increasing structural complexity of the French research landscape has a massive influence on the construction of reliable indicators for the purposes of research assessment, one of the most damageable downside being the heterogeneity of the bibliometric indicators. The study aims to make the inventory of practices and to suggest solutions, sometimes relying on technology, to overcome the visibility issue of the French research and higher education institutions.; La problématique des affiliations dans les publications scientifiques se pose de manière de plus en plus aiguë, à l’heure où le paysage qui structure la recherche se complexifie et alors que les enjeux autour de l’évaluation donnent la mesure de l’hétérogénéité croissante dans la construction d’indicateurs fiables pour les établissements de l’enseignement supérieur et de la recherche.

International audience; Background - The problem of access to medical information, particularly in low-income countries, has been under discussion for many years. Although a number of developments have occurred in the last decade (e.g., the open access (OA) movement and the website Sci-Hub), everyone agrees that these difficulties still persist very widely, mainly due to the fact that paywalls still limit access to approximately 75% of scholarly documents. In this study, we compare the accessibility of recent full text articles in the field of ophthalmology in 27 established institutions located worldwide. Methods - A total of 200 references from articles were retrieved using the PubMed database. Each article was individually checked for OA. Full texts of non-OA (i.e., "paywalled articles") were examined to determine whether they were available using institutional and Hinari access in each institution studied, using "alternative ways" (i.e., PubMed Central, ResearchGate, Google Scholar, and Online Reprint Request), and using the website Sci-Hub. Results - The number of full texts of "paywalled articles" available using institutional and Hinari access showed strong heterogeneity, scattered between 0% full texts to 94.8% (mean = 46.8%; SD = 31.5; median = 51.3%). We found that complementary use of "alternative ways" and Sci-Hub leads to 95.5% of full text "paywalled articles," and also divides by 14 the average extra costs needed to obtain all full texts on publishers' websites using pay-per-view. Conclusions - The scant number of available full text "paywalled articles" in most institutions studied encourages researchers in the field of ophthalmology to use Sci-Hub to search for scientific information. The scientific community and decision-makers must unite and strengthen their efforts to find solutions to improve access to scientific literature worldwide and avoid an implosion of the scientific publishing model. This study is not an endorsement for using Sci-Hub. The authors, their institutions, and publishers accept no responsibility on behalf of readers.

Purpose To perform a bibliometric analysis in the field of dry eye disease (DED) research to characterize the current international status of DED research and to identify the most effective actors (journals, countries, authors) involved in this field. Methods Scientometric methods were used to evaluate global scientific production and development trends in DED research, using the Web of Science Core Collection. Results The growth of the literature related to DED averaged 12.18% over the last 10 years. A total of 5522 original and review articles, published in 821 different journals, were identified. The USA was the most productive country with 34.53% of the overall articles studied and 46.10% of the overall citations. The Ocular Surface published a very high percentage of articles related to DED relative to the total number of articles published (31.87%). The most productive institutions and the most frequently cited articles were from the USA and Japan. A network visualization map for country collaboration revealed that most European countries developed most of their collaborations with countries belonging to their own continent, which was not the case for the USA or Japan. A total of 41,956 KeyWords Plus were found with an average of 7.6 (SD = 3.15) KeyWords Plus per article. Conclusions This study provides a broad view of the current status and trends in DED research and may help clinicians, researchers and policy makers better understand this research field and predict its dynamic directions.

International audience; Over time, academic social networks are more and more used by researchers, especially thanks to the possibilities of sharing articles they offer. The objective of the present study was to evaluate the proportion, the typology (pre-print, post-print author/publisher) and the legality of the full-text publications deposited by researchers on ResearchGate which is widely used by the medical and biological community, using a sample of 1,500 randomly selected articles in PubMed and published between 2013 and 2015. To compare, the access to the full-text of the 1500 articles via PubMed and PubMed Central has been assessed, putting into evidence the important role ResearchGate plays for providing full-texts of articles in biology/medicine. It also puts academic social networks into perspective in relation to open-access repositories and open access.; Les réseaux sociaux pour les chercheurs s’imposent au fil du temps auprès des communautés académiques, notamment grâce aux fonctionnalités de dépôt et de partage d’articles qu’ils proposent. L’objet de cette étude est d’évaluer la proportion, la typologie (pre-print, version auteur non expertisée ; post-print auteur/éditeur, versions auteur et éditeur expertisées définitives) et la légalité des textes intégraux d’articles qui y sont déposés par les chercheurs, à partir d’un échantillon de 1 500 articles publiés entre 2013 et 2015 sélectionnés aléatoirement dans PubMed, en prenant pour exemple ResearchGate qui est très largement utilisé par la communauté médicale et biologique. À titre de comparaison, l’accès au texte intégral des 1 500 articles via PubMed et PubMedCentral a également été évalué, mettant en évidence le rôle important que joue ResearchGate dans la fourniture de textes intégraux d’articles en biologie-médecine. Cet article met également en perspective le positionnement des réseaux sociaux par rapport aux archives ouvertes et au libre accès.

International audience; Over time, academic social networks are more and more used by researchers, especially thanks to the possibilities of sharing articles they offer. The objective of the present study was to evaluate the proportion, the typology (pre-print, post-print author/publisher) and the legality of the full-text publications deposited by researchers on ResearchGate which is widely used by the medical and biological community, using a sample of 1,500 randomly selected articles in PubMed and published between 2013 and 2015. To compare, the access to the full-text of the 1500 articles via PubMed and PubMed Central has been assessed, putting into evidence the important role ResearchGate plays for providing full-texts of articles in biology/medicine. It also puts academic social networks into perspective in relation to open-access repositories and open access.; Les réseaux sociaux pour les chercheurs s’imposent au fil du temps auprès des communautés académiques, notamment grâce aux fonctionnalités de dépôt et de partage d’articles qu’ils proposent. L’objet de cette étude est d’évaluer la proportion, la typologie (pre-print, version auteur non expertisée ; post-print auteur/éditeur, versions auteur et éditeur expertisées définitives) et la légalité des textes intégraux d’articles qui y sont déposés par les chercheurs, à partir d’un échantillon de 1 500 articles publiés entre 2013 et 2015 sélectionnés aléatoirement dans PubMed, en prenant pour exemple ResearchGate qui est très largement utilisé par la communauté médicale et biologique. À titre de comparaison, l’accès au texte intégral des 1 500 articles via PubMed et PubMedCentral a également été évalué, mettant en évidence le rôle important que joue ResearchGate dans la fourniture de textes intégraux d’articles en biologie-médecine. Cet article met également en perspective le positionnement des réseaux sociaux par rapport aux archives ouvertes et au libre accès.

Digital object identifiers (DOIs) were launched in 1997 to facilitate the long- term access and identification of objects in digital environments. The objective of the present investigation is to assess the DOI availability of articles in biomedical journals indexed in the PubMed database and to complete this investigation with a geographical analysis of journals by the country of publisher. Articles were randomly selected from PubMed using their PubMed identifier and were downloaded from and processed through developed Hypertext Preprocessor language scripts. The first part of the analysis focuses on the period 1966–2015 (50 years). Of the 496,665 articles studied over this period, 201,055 have DOIs (40.48%). Results showed that the percentage of articles with DOIs began to increase for articles published in the 2000s, with spectacular growth in the years 2002–2003, then reached a peak in 2015. Data on countries showed that some countries gradually implemented DOIs over the period 1966 to 2015 (the United States, the United Kingdom, and the Netherlands), while some did not (Russia, the Czech Republic, and Romania). The second part of the analysis focuses on the year 2015 and includes 268,790 articles published in 2015, randomly selected to evaluate the current implementation of DOIs. In 2015, 86.42% of articles had DOIs. The geographical analysis of countries of publishers showed that some countries (Russia, Thailand, and Ukraine) still assigned few DOIs to articles in 2015. Thus, if the scientific community aims to increase the number and the usefulness of services rendered by DOIs, efforts must be made to generalize their use by all persons involved in scientific publication, particularly publishers.

Background: The purpose of this study is to provide a report on scientific production during the period 2010–2014 in order to identify the major topics as well as the predominant actors (journals, countries, continents) involved in the field of eye disease.Methods: A PubMed search was carried out to extract articles related to eye diseases during the period 2010–2014. Data were downloaded and processed through developed PHP scripts for further analysis.Results: A total of 62,123 articles were retrieved. A total of 3,368 different journals were found, and 19 journals were identified as “core journals” according to Braford’s law. English was by far the predominant language. A total of 853,182 MeSH terms were found, representing an average of 13.73 (SD = 4.98) MeSH terms per article. Among these 853,182 MeSH terms, 14,689 different MeSH terms were identified. Vision Disorders, Glaucoma, Diabetic Retinopathy, Macular Degeneration, and Cataract were the most frequent five MeSH terms related to eye diseases. The analysis of the total number of publications showed that Europe and Asia were the most productive continents, and the USA and China the most productive countries. Interestingly, using the mean Five-Year Impact Factor, the two most productive continents were North America and Oceania. After adjustment for population, the overall ranking positions changed in favor of smaller countries (i.e. Iceland, Switzerland, Denmark, and New Zealand), while after adjustment for Gross Domestic Product (GDP), the overall ranking positions changed in favor of some developing countries (Malawi, Guatemala, Singapore).Conclusions: Due to the large number of articles included and the numerous parameters analyzed, this study provides a wide view of scientific productivity related to eye diseases during the period 2010–2014 and allows us to better understand this field.

International audience; BackgroundOver the past two decades, the world wide web has changed researchers' habits. These changes were further extended when “web 2.0” appeared in 2005, bringing tools and platforms facilitating user collaboration, user-generated content and data sharing. These tools have gradually influenced the world of research, especially in medicine. Bibliometrics is a helpful and widely used tool for describing patterns of publication and interpreting temporal evolutions and the geographical distribution of research in a given field. Few studies employing bibliometrics, however, have been carried out on the correlative nature of scientific literature and web 2.0.ObjectivesThe aim of this bibliometric analysis was to provide an overview of web 2.0 implications in the field of Medicine. The objectives were to assess growth rate of literature, key journals, authors and country contributions, and to evaluate whether the various web 2.0 applications are expressed within this biomedical literature, and, if so, how.MethodsA specific query with keywords chosen to be representative of web 2.0 applications has been built for PubMed database. Articles dealing with web 2.0 were downloaded in XML and were processed through developed PHP scripts, then were imported to Microsoft Excel 2010 (Microsoft) for data processing.Results1347 articles were included in this study. The number of articles dealing with web 2.0 has been increasing from 2002 to 2013 (average annual growth rate was 106.30 % with a maximum of 333 % in 2005). The USA was, by far, the predominant country for authors, with 514 articles (53.99 %). The second and third most productive countries were the UK and Australia, with respectively 87 (9.14 %) and 44 articles (4.62 %). Distribution of number of articles per author showed that the core population of researchers working on web 2.0 in the medical field can be estimated to approximately 75. 614 journals were identified during this analysis. Using Bradford's law, 27 core journals were identified, among which 3 (Studies in Health Technology and Informatics, Journal of Medical Internet Research and Nucleic Acids Research) produced more than 35 articles related to web 2.0 over the period studied. 274 words in the field of web 2.0 were found after manual sorting of the 15878 words belonging to title and abstract fields of articles. Word frequency analysis reveals “blog” as the most recurrent, followed by “wiki”, “web 2.0”, ”social media”, “Facebook”, “social networks”, “blogger”, “cloud computing”, “Twitter” and “blogging”. All categories of web 2.0 applications were found, indicating the successful integration of web 2.0 into the biomedical field. Remarkably, apart from wikis, applications specifically developed for science, biology or medicine were rare (e.g. PatientLikeMe,) or not represented (e.g. researchblogging.org for blogs, Researchgate and Academia for social networks).ConclusionsThis study shows that the biomedical community is engaged in the use of web 2.0 and confirms its high level of interest for these tools. Therefore, changes in the ways researchers use information, generated by the arrival of the world wide web and web 2.0, seem far from over.

Purpose To calculate the growth rate of the biomedical literature on eye neoplasms and to assess which journals, countries, and continents are the most productive. Methods PubMed was used to search for articles published from 1966 to 2012. Total number of articles per year was fitted to a linear equation as well as an exponential curve. To identify the core journals and predict the number of journals containing articles related to eye neoplasms, Bradford's law was applied. For each country and each continent, the gross domestic product (GDP) index (publications per $1 billion USD of GDP) and the population index (publications per million inhabitants) were calculated. Results A total of 27,943 references were retrieved. The growth in the number of publications showed a linear increase with a yearly average growth rate of 2.08%, which was lower than for the whole PubMed database (3.59%). Using Bradford's law, 17 core journals were identified, among which 2 journals produced more than 1000 articles (JAMA Ophthalmology and American Journal of Ophthalmology). Europe was the most productive continent, followed by North America and Asia. The United States was by far the predominant country in number of publications, followed by Germany and the United Kingdom. However, population and GDP indexes showed that absolute production did not reflect the production per capita or economic efficiency. Conclusions This bibliometric study provides data contributing to a better understanding of the eye neoplasm research field.

International audience; In the last decade, the applications known as "web 2.0" (Facebook, Youtube, Twitter...) have changed our daily life and gradually influence the field of research. This article aims to make a critical overview of these new services, and emphasizes the changes induced for researchers (practice of scientific publication, sharing and mutualization of research data and discussion between researchers...) especially in the field of biology/medicine. A focus is done on the limitations that deter most of scientists from using these services in their common practice (lack of knowledge about these tools, time-consuming, fear of sharing data and ideas). Despite these restrictions, some mutations affecting researcher's information uses are unavoidable, and these new tools may rapidly contribute to scientific advances.; Les nouvelles applications du web dites " web 2.0 " (Facebook, Youtube, Twitter...) ont envahi notre quotidien et s'immiscent peu à peu dans le domaine de la recherche. Le présent article se propose d'en faire un panorama critique, et de pointer les évolutions qu'elles induisent : pratiques de publications scientifiques, mise en débat des idées, partage et mutualisation des données de la recherche... Un certain nombre de freins à l'adoption de ces nouveaux outils par les chercheurs persistent cependant, obstacles qui, s'ils ne sont pas levés, limitera leur utilisation et leur participation aux avancées scientifiques.

Édition numérique : ISBN 9782375460450.; National audience

Some websites are specialized in the referencing of educational resources and their evaluation, as in the peer-review of scientific publications. But can their description, indexing and evaluation be done without knowing the context of use of the resource ? if so, what are the criteria ? Through a comparison of a few web sites offering educational resources, this article proposes an analysis of the functioning and the evaluation criteria in order to draw models of “description-evaluation” of these resources., Certains sites web se sont spécialisés dans le repérage de documents pédagogiques et leur évaluation, à l'instar du peer-review des publications scientifiques. Mais peut-on réaliser la description ; l'indexation et l'évaluation sans connaître le contexte d'usage de la ressource ? Si oui, selon quels critères ? La comparaison de quelques sites proposant des documents pédagogiques validés permettra d'analyser leur fonctionnement et leurs critères d'évaluation, pour modéliser des scénarios de « description-évaluation » de ces ressources.

L’objectif de cet article est d’exposer le mode de fonctionnement des outils de recherche rencontres sur internet (moteurs, meta-moteurs et annuaires) et de proposer des pistes pour mieux maitriser leur usage. Les difficultes rencontrees lors de l’utilisation de ces outils de recherche d’information y sont exposees et l’interet potentiel des outils specifiquement dedies a la recherche d’information en biologie/medecine discute.

This article presents a pedagogical tool developed on the internet which aim is to facilitate the access to scientific information for graduate students and researchers in biology. This presentation is followed by a discussion on the interest of this tool, Cet article présente dans un premier temps un dispositif pédagogique mis en place sur internet dont l'objet est de faciliter l'accès des chercheurs et étudiants de 3ème cycle universitaire en biologie à l'information scientifique et technique et propose dans un deuxième temps une discussion sur l'intérêt de ce type de dispositif, tant du point de vue de sa structure (accès à distance) que de son contenu.

Pour citer cet article : Dassa M., Deniau A., Durand-Barthez M., Girard F., Pothier N., Séné A. (2014). Le millefeuille des affiliations françaises dans les publications scientifiques. Documentaliste - Sciences de l'Information, 51 (4), 12-16; International audience; À l’heure où l’évaluation et le financement sous-jacent de la recherche sont en jeu, l’affiliation des acteurs de la recherche publique française est une question essentielle. Le groupe Renatis/Urfist présente les conclusions de ses travaux.

[developpement] Le reseau Renatis du CNRS et l’Urfist de Paris menent une reflexion sur les affiliations dans les publications scientifiques dont l’enjeu est la visibilite du paysage scientifique francais. Cet article fait suite a un premier, paru dans le n˚4-2014 de Doc-SI sous le titre « Le mille-feuille des affiliations francaises dans les publications scientifiques ».

La recherche bibliographique représente, en quelque sorte, une des fondations de l’édifice des connaissances sur lesquelles les chercheurs s’appuient au quotidien dans leurs laboratoires pour réaliser leurs travaux de recherche. Parmi les nombreuses bases de données bibliographiques disponibles dans le domaine biomédical (Embase, Current Contents, Biosis…), Medline PubMed est de loin la plus utilisée en raison de la gratuité de son accès. Cependant, le bon usage des différentes fonctionnalités proposées par cette base (utilisation du vocabulaire contrôlé MeSH, réalisation de revues de sommaire, systématisation de recherches bibliographiques…), préalable indispensable à une recherche bibliographique de qualité, nécessite de posséder un certain nombre de notions qui sont exposées dans le présent article., To be informed about the last publications recently published or to produce a bibliography in a given thematic field is essential for researchers in the biomedical field. If the use of Internet information searching tools such as «Google» or « Alltheweb » makes possible to discover a great part of the grey literature, bibliographic databases like Embase, Current Contents, Biosis or Medline via PubMed are essential tools to locate scientific articles. Among these bibliographic databases, Medline PubMed, thanks to its free access, is the most used. However, a correct utilization of the various functionalities proposed (thesaurus MeSH, systematization of bibliographic searches…), and consequently the quality of bibliographic researches carried out in this database, requires to master elementary knowledge which are exposed in this article.

Le réseau internet révolutionne peu à peu les pratiques des chercheurs tant du point de vue de leurs recherches documentaires que du point de vue de la diffusion de leurs travaux. Si l’opportunité d’accéder à une masse d’informations colossale « d’un simple clic » est une perspective particulièrement séduisante pour tout chercheur en biologie, la découverte d’informations pertinentes dans cet « océan » s’avère en réalité relativement difficile, et cela malgré l’existence d’un nombre croissant d’outils de recherche mis à la disposition des internautes., The internet network revolutionizes little by little the practices of the researchers as well from the point of view of their information retrievals as from the point of view of the diffusion of their work. If the opportunity to reach a colossal mass of information very simply is a particularly attractive perspective for every researcher in biology, this study puts in evidence that the discovery of relevant information in the field of biology in this « ocean » turns out relatively difficult, and this in spite of the existence of an increasing number of search tools at the disposal of the Internet users. Furthermore, the results presented suggest that the use of non-specialized search engines and meta-engines seems preferable with that of specific search engines in the field of the biology and with that of non-specialized or specific directories in biology.