Reliability of sources and quality of knowledge on the Web

To what extent is what we read on the Web reliable? There was a time, not too long ago, when the phrase “It said on the television” counted as a pledge of reliability for any information. Today, the web tends to take its place (think of the use of social media in the debates between politicians and reporters) and, despite lacking much of the credibility that television gained during its phase of greatest success, it is the source to turn to for both practical questions and any kind of news. This very article is based on many sources from the Web, whose only partial guarantee of reliability stems from cross- checks. That is to say that it is good practice to critically question how true what the network offers is what the network offers.

Easy to say and hard to do: to doubt is a lot more tiring than to believe. Moreover, the area to which methodical doubt applies is so extensive that it tends to escape any systematic investigation. The issue of source reliability has always existed and does not only relate to complicated academic issues: one of the world’s most famous apocryphal documents, known as the Donation or Edict of Constantine,[1] resulted in a significant part of history from the eighth to the nineteenth century and beyond making an indelible mark on the destinies of countless people. When you say a well-told lie…

Going back to the web, if you try to explore the topic of reliability you are faced with a multitude of themes, roughly grouped into two broad categories: documents (textual or otherwise) produced by humans, which can distort reality in a more or less voluntary way, and those developed by machines according to algorithms, exposed to significant risks of misunderstanding and “stagnation” of knowledge, as we shall see later.

The first of these categories poses questions such as How can we distinguish reliable information on the net from those manipulated for partisan interests? How can we know whether topics are real or pure and simple inventions made up by some hoaxer? The second promises dilemmas worthy of a short story by Philip K. Dick: Are the items we read on the various wiki websites written by men or robots? To what extent are translations by linguistic robots reliable? For those who want to expand the horizon a little: what implications arise from entrusting information processing to machines that cannot understand their meaning?

However, we should set some limits on this survey and agree to deal with it in part and try to get the clearer picture of the complexity we have to cope with when we surf the Web. Some examples, although limited, can give an idea of how the quest for truth (well let’s forget about beauty) is now more than ever a difficult business.

Example 1: Wikipedia and the hoaxes:

The starting point of this strange trend probably has to do with the desire to impress, although in some cases real disinformation was intended.

Some time ago Marta Serafini, a Corriere della Sera[2] blogger, reported the strange story of a Wikipedia entry describing the war of Bicholim, fought between 1640 and 1641 – according to the online text – by the Portuguese and the Indian empire of Maratha for the control of the island of Goa. I hate such entries because they make me always think : «How ignorant I am».

But this time my pride suffered less than expected. Last December user and detective Wikipedia ShelfSkewed sifted through the story and – says Serafini – stated the following: “After a careful evaluation and some research, I came to the conclusion that this entry is a lie, an intelligent and elaborate hoax. “Of course, Wikipedia has removed the item.

“It will be the isolated act of some jester” I imagined. Wrong this time too. The phenomenon is well known to Wikipedia goes by the name of hoax, or falsehood (bufala in Italian). There is an amazing number of hoaxes, often well designed and written in great detail even with references, to be found on Wikipedia[3] and which necessarily correspond to a comparable number of hoaxers (at best) who invest time and energy to come up with rumours, urban legends, pseudo-sciences or April Fools’ Day events sometimes very hard to break.[4] The Wikipedia record holder is someone called Caio Flavio Antonino, alleged murderer of Caesar, who remained so for 8 years and 1 month. Congratulations to the imaginative inventor!

The hoaxes are not limited to characters who never lived and events that never took place but tend to contaminate all fields of knowledge. A quick leaf through the pages of the Museum of Hoaxes[5] gives a good idea: you will come across marshmallow plantations, bacon mouthwashes bacon, prehistoric fish with the head of a rooster, cotton candies and other similar wonders: in short, there is much to enjoy.

I think that this a satisfactory explanation to the existence of websites dedicated to dealing with identifying and cataloguing the hoaxes that populate the Web and expose some of the hoaxers who pull our legs (or worse). A consultation of these pages and, above all, a comparison of independent sources can be helpful, but despite our best effort, the Web offers a lot of information whose degree of reliability cannot at first be determined. To the wide range  already available prior to the birth of the Internet, new material has been added, increasingly numerous and often carried out in a professional manner. Today, thanks to the Web 2.0 functions, anyone can publish anything on the Web and this has naturally stimulated the proliferation of tall tales of all kinds. It turns out to be so easy for those who have some know-how to legitimate a further suspicion: could it be that some “official” information, trusted by the media system, is actually equipped with horns and tail? (again, the Italian translation for hoax is “bufala”, that is to say buffalo). Since candidates abound, ranging from politics to science, economy and sport to virtually any field of knowledge, I leave the reader to cast doubt on whatever he prefers. In any case, best be on guard.

Author’s note

I wish to inform the readers that both the following chapters and the other themes of this one – writing robots, bootleg pictures and video, etc.. – are part of an editorial project which will lead to the publication of an e-book, and if I find a publisher, also a hardcover book. I will keep you informed on the progress of the project.