How To find Information On the Web

Article written by Denis Sureau on August 7, 2007.

To the most essential question that can be posed by a Net surfer, answers are numerous. Let me detail all means to find information or site on the Web, and at the same time to specify what the webmasters must make so that information which they provide has most chance to be found by those which it interests.

Search engines

The simplest mean to find information, is also the biggest activity for webmasters, the art and techniques of referencing and SEO.

A question of keywords

Finding a sentence in a text among 10 billions pages indexed, that is possible with main search engines (Google, Live, Yahoo), it is enough to put the sentence between quotation marks in a request! Most often the search is done on a list of keywords that one provides, according to a simplistic boolean search: + for a combination, - for an exclusion. The AND and OR operators are used in advanced search.
Google can return pages which do not contain the required words: these words are in fact in the links which point on the page, this one is thus in relation to the subject actually.

Facilitating the indexing

To improve the indexing of its Web site is really simple, it is due to the "anchors"?, the label of internal links, possibly supplemented by a "title" attribute?, which gives information about the link to visitors and also to engines. The anchor must contain a maximum of keywords while being significant for the visitor. One cannot control the anchors of external backlinks, but to inserting keywords in the names of files and sub-directories has the same effect.

Search engines for blogs

For the news, one will be better informed from blogs, and specialized search engine as Google Blog Search or Technorati will be more appropriate.

Universal Search

This new search service wants to combine various engines: text, videos, images in a single one, that requires the context of search taken into account. For example if one search for "scriptol"?, one finds information in "books"?, "groups"?, "code"? while an other research such as "Miami"? would relate to "Google Earth"?, "images"?, "text" instead?.

A good example to illustrate a such universal research, is a request on Steve Job.
Note that the installation of the Flashgot extension of screen capture for Firefox, by presenting a reduced image of the page in front of each result, adds to the multi-media aspect of searching.

Directories

Professional directories and directories of sites do not have the same role even when both list websites: contents of former one may be limited to a page of contact while latter as Dmoz requires that the site have extended and unique contents.
The interest to have a site listed in Dmoz is sometimes disputed by webmasters (that does not impeach them from insisting to appear there), but it seems real nevertheless. If one wants to find the best sites about a subject, the visit of the relevant category on the Open Directory can be profitable.

Sites of answers

To have an answer, why not ask directly the question? On these sites, there are not robots but voluntary human that will answer the question and the sites select the best answer, that which is voted by plebiscite by users. One never misses volunteers to answer, they find a profit in various ways there. That can be the opportunity to give the URL of their own site sometimes.

The Answers service from Yahoo, provides answers made by the volunteers for frequently asked questions. But Answertips service would return a better result. For example, if one types"How to find information on the Web"?, one obtains no answer! And one is redirected on the traditional engine using keywords.
Answertips is a service of Answers which helps search by supplementing the question according to a database, as if the question had been already asked, and that allow to translate the question to a close question under another formulation made by Net surfers.

The Answers.com site seems to use a traditional search engine with only 4 million page (against 10 billion for Google), but it provides a software that allow to start a search when one clicks on a word in a page.

Sites of news

The principle of digg-likes, is to associate a score to each article. They are the visitors by clicking on an image who give the points to denote their interest for the post, and thus lead attention from other visitors.

They are thus named because Digg has popularized the idea what contributed to its fortune, and then lot of other website followed, in particular Reddit and digpics which proposes astonishing images. The creation of the Pligg CMS now makes it possible for each webmaster to create its digg-like, and it is created almost a new one each day!

The problem on a digg-like for the blogger or webmaster is to arrive on the front page. Theoretically it is the number of clicks which brings an article in the home page, but in fact, that goes in the inverse direction: it is necessary to have contents of general interest, which is unique or that one is the first to publish and one go to main page, it is there that one obtains the marks!

Sharing bookmarks

Another idea is to share bookmark with Net surfers. Rather than to keep his list of favorites on the browser, one manages it on Del.icio.us, Blogmarks or Stumpleupon, the others have access there and can discover sites as you can yourself benefit of sites discovered by others. A relational network is created, you appreciate what mark another Net surfer, you add it to your list of friends.
Webmasters will not miss to use the system to promote their best articles, but to misuse it will have a negative effect, you will not have friends, not visitors on your list of bookmarks!

Forums

For a precise question, information, something which requires an experiment, forums bring an unequaled richness. Search engines include the contents of forums, also the forum is appropriate for a new question.

Search on site

Web sites are not always conceived like documentary tool, with a home page directing on each article or category. The new visitor will be often stray on an important site if it does not provide a sitemap in HTML format. Create it is easy.

The webmaster must multiply hyperlinks with the disadvantage to have visitors leaving the site, but a page which wants to provide complete information cannot avoid them, and such article it can go in lot of bookmarks (I have more than 1000 marks for some pages)!

What is upcoming... in brief

Search engines are directed towards e multi-media. An online encyclopaedia proposes Wikia to replace robots by human who will select the pages for users (a social progress!). The Quaero European project of search engine is also multi-media and multilingual but does progress quickly. Google Coop creates a network of confidence where members integrate the sites which they selected. A track for the future perhaps?