Apart from the structures or behind the screen technology of databases/search engines, I am trying to provide another dimension (purely user's  so Librarian's too). For a user, which one is more important for literature review and for research institutions? Just an article index/database or a citation index/database? As we know, article databases usually allow us to search and find journal literature (now many of them includes book chapters, theses and many OA documents). These were introduced by aggregators or compiled by individual libraries, especially when we were in 100% print based. When the scenario changed to online, the records provide direct hyperlinks to articles. Then we have online indices/databases (Web of Science/Web of Knowledge, Scopus etc) and search engines (Google Scholar) that provide the citation details of articles, in addition to direct links, enabling us to gauge the influence and impact of that work, plus links to related articles and cross references. Citation indices/databases are used for a "comprehensive literature search/review", where the quality of your references is important for your assignment/paper/thesis. In addition to the literature search and the bibliometric/scientometric studies (individual performance analyses metrics carried out by using Citation index/database), recently Institutions started using them to measure the impact of their research and evaluating their programs/staff/funding etc. There are tools developed (Incites, Research-In-View, Scival, PURE etc) based on these databse/indexes. I also understand that these tools will also be utilizing new altmetrics components (like how many times articles downloaded, shared on social networking, news medias etc) . So my question is, having subscribed to just an article database (whatever the coverage be) is not going to serve the purpose of a researcher (literature review) or a new generation institution (research evaluation). As librarians, it is time to rethink, and do not miss the chance to get in to institutional wide efforts, with appropriate tools. We have to come out of the shadows of just "information retrieval" to a broader scope of "information analysis", "research impact analysis". Who else can do this for an organisation? Just my thought, many people will have different opinions. We have to move out of vendor defined tools, and look forward to user/institutions wanted tools....if we need to redefine ourselves. ---------- Dr. J. K. VIJAYAKUMAR, Manager, Library Research & Reference, King Abdullah University of Science & Technology,   -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.