Search Engine News

Google	2 Nov. 01	Nouveaux formats indexés par Google En plus de ses 22 millions de fichiers PDF, Google vient d'annoncer qu'il indexait dorénavant les fichiers Word (.doc), Excel (.xls), Powerpoint (.ppt), Rich Text Format (.RTF) et PostScript (.ps). Google est le premier moteur de recherche majeur à prendre en compte ce type de fichiers, la plupart des moteurs se contentant d'indexer jusqu'à maintenant uniquement des fichiers HTML statiques ou dynamiques. Google renforce ainsi encore un peu plus son image d'outil de recherche innovant. Ceci dit, la masse de documents à ces nouveaux formats étant assez importante sur l'Internet, il risque de se poser pour le moteur un problème de masse critique à un moment ou à un autre. Et on peut penser qu'il faudra bien, très rapidement, penser non plus en termes quantitatifs mais plutôt qualitatif ("documents utiles" plutôt que "documents lambda") sous peine d'être noyé par le nombre de résultats pour un mot clé donné.... Source : SearchEngineWatch
	25 July 01	Google Image Search Grows This week, Google released a new, 66 percent larger index for Google Image Search. Google Image Search now enables users to search and browse 250 million digital images, 100 million more than the first index, which was released just a month ago.
	24 May 01	How to restrict your search to pdf files only Google now allows you to restrict your search to pdf files only: simply add to your search criteria the following: filetype:pdf or, to ignore all pdf files: -filetype:pdf
	5 April 01	Google Offers Translation Google has started offering machine translation of their search results. (Machine translation just means that the translation is done automatically, with a computer. Machine translation is nowhere near perfect, and should be relied on only to get the "gist" of a page you can't read.) This new feature, which is in beta, is integrated into Google in two ways. The first is in the search results. All search results in Spanish, German, French, and Portuguese will have a "translate this page" link beside them. Click on it and you'll get a translated page framed by Google. There's a link to a printable version as well. (If you don't see a "translate this page" link on your result page, go back and check your preferences. If you have "Search only these selected languages" checked, and none of the languages are the ones named in the last paragraph, you won't see the "translate this page" link.) Google's integrated the translation directly into their search results. Check out the preferences page. You can now specify that you want titles and page summaries in Spanish, German, French and Portuguese translated directly on the results page... that's the theory, anyway. Read more about Google's translation options.
	5 April 01	Google now supports 30 search languages. Go to http://www.google.com/ and click on "Google in your Language" at the bottom of the page.
	27 March 01	PDF searching Lots of valuable information on the web is not presented as HTML web pages, but as PDF files. These pages are snapshots of text and/or graphics documents and are viewed through special readers like Adobe Acrobat (available as a free download). Google now searches these Graphic files and makes them available to you within your search results. Each PDF file is labeled with a blue <PDF> at the beginning of the results listing. Clicking on one of these results will automatically launch a PDF viewer if it's on your computer or direct you to a site from which you can download one. Or you can chose the "text version" link to see the document's contents in an unformatted view. Read more!
	20 March 01	Google Adds Date Searching Google has finally added date searching to their Usenet archives. (Doogle? Deja Goo? Google Vu?) From the advanced search page you can choose from two dropdown menus at the bottom of the page. One will let you search messages posted anytime, within the last week, or within the last month. The second dropdown lets you specify a date between August 2000 and now. You may also sort your results (from 10 to 100 per page) by either relevance or date. That choice is at the top of the page near the text entry boxes.
	12 February 01	Google acquire Deja.com's Usenet Discussion Service On February 12, Google acquired Deja.com's Usenet Discussion Service. This acquisition provides Google with Deja's entire Usenet archive (dating back to 1995). Google is working quickly to make all this material available for browsing and searching. Already available at http://groups.google.com, is a new Usenet search feature that enables Google users to access a wealth of information contained in more than six months of Usenet newsgroup postings and message threads - more than was available with Deja's default setting previously. Once the full Deja Usenet archive is added, users will be able to search and browse more than 500 million archived messages with the speed and efficiency of a Google search. In addition to expanding the amount of searchable data, Google will soon provide improved browsing capabilities and newsgroup posting.
	3 July 00	Google's Hidden Special Syntaxes If you've heard about Google's new claim of a billion pages (about half of them fully-indexed pages) you may have decided it's time to take a look at Google. Before you head over to http://www.google.com, though, take a look at the secret hidden syntaxes. Google will say in its help file that it currently supports only one special operator, the link: operator (http://www.google.com/help.html#special). I know of at least two more. One is called cache:. If you use cache: and then a URL, you'll get the currently cached version of the URL that Google has in its search engine. For example, the query cache: http://www.cnn.com at this writing will give you CNN's front page from May 19th. The second is called site:. The site: syntax works like the host: syntax on AltaVista. For example, if you wanted to find every instance in Google's index where Jimmy Buffet was mentioned on the MTV website, you'd enter the query "jimmy buffet" site:mtv.com. You'll only get five results. Doing the search without the site:mtv.com gives you over 11,000 results. More hidden special syntaxes.
Altavista	5 April 01	AltaVista Changes Search Defaults and Has Phrase Problem Double quote phrases are broken in basic search at the moment and the Boolean defaults are downright confusing. Get the whole story!
	5 April 01	AltaVista's Babel Fish Learns New Languages AltaVista announced yesterday that their translation engine now supports Asian language characters and offers translation of Chinese, Japanese, and Korean. (Both traditional and simplified Chinese are supported.) If you're thinking, "My browser doesn't support Asian language characters, so this is irrelevant to me" -- it's not. Remember, with Babel you can cut and paste the URL of a page to be translated. So if you find a Japanese page via a search engine, cut and paste the URL into Babel Fish and read it that way. Two caveats: 1) Machine translation is not perfect. You'll get the gist but sometimes that's all you'll get. 2) Be sure to include http:// on the URL. The 'Fish, in my experiments, considered a URL without an http:// to be "malformed."