It's a monitoring. In military terms , it's a sentry in night duty
1.1 Definitions
"Know, plan and do ", detection of weak signals The different stages of web watch :
Collect information
Sort out, process information
Spread to a chosen audience
Decide according to the collected information
But according to the kind of watch, all stages are not compulsory (the spreading, in the case of an individual web watch for instance ) or formalized (most often the decision making).
There are different kinds of watch which vary according to the topic (legal, informative,...) and the objective (strategic watch where an eye is kept on competitors in order to position one's firm and products, creative watch on designers works in order to detect new trends, ...)
In all cases, watch is an iterative and continuous process
iterative : round trips between results and sources (and tools) so as to improve the watch's results with regard to its objectives. This round trip also allows to develop the watch compared with the development of or compared with the evolution peculiar to the topic of the watch (emergence of new themes in the watch's field for instance).
continuous : watch is not an inventory of fixtures at a given time or a situation of art but a continuous process following current events or the development of topics.
1.2 An informative … and collaborative web watch
Within the framework of this course,we'll be interested in "informative " web watch. Examples of watched contents :
The theme news
Calendar of events, colloquium
Similar group initiatives but in another environment (that brings in some fresh air)
Research news
Watch on new publications (bibliography)
Collaborative initiatives
Regulatory web watch (light)
....
The stages of collaborative web watch are the same as those of individual watch but they are done by all its participants.
Collect information (together)
Sort out, process information (together)
Spread to a chosen audience (to the group and beyond)
Decide according to the collected information (rarely formalized for informative watch).
These tasks are done by all the people and are not distributed).
The most important point is the spreading : we agree on the content, on the selection of criteria of what is going to be brought to the collaborative watch, with all the participants. It may be the watch which interests the group and/or the watch aiming a specific audience.
It's the difference between the shared watch and the collaborative watch :
shared watch : I do my watch according to my own criteria and I share it,
collaborative watch : the group agrees on the topic of the watch and only feeds the collaborative watch about this topic.
2. What does web watch change ?
Watch is the processing of flow
2.1 Before the Internet
Flows: radio, television, newspapers
Stages of watch :
Collect : Subscription to newspaper and newsletter
Sort out, process : press review, documentary file
Spread : press review, provision of files
Decide : summaries, reports
2.2 Now
New flows of information which add in to old ones :
Alarm system (on line services which alarm via an email or a RSS feeds when a new article comes out on chosen key-words)
Website watch (enables to follow news from sites without feeds).
And often one finds himself alone : no selection, no publishing
3. Set up a web watch
3.1 Define field and themes
3.2 Find sources
Finding the right sources is attempting the impossible in front of the profusion of information available on the Internet !
Canvassing is done through browsers, social bookmarking tools (Diigo, Pearltrees,...), newspapers, recommendations from colleagues, etc... But there is afterwards a constant revaluation of sources (addition, deletion) according to the watch's results : constant toing and froing.
In the model proposed by Jean-Michel Cornu : abundance/rarity/opportunity/constraint, web watch is located on the abundance's side, the abundance of information on the Internet.
(Crédit : Marc Lanssens - SupAgro Florac - CC By-Sa )
The winning strategy is therefore the opportunity strategy and not the planning strategy : I can't know in advance what information I will find, because the watch is exactly the supervision of news and weak signals. One must "try " sources and according to results modify them.
The struggling is to know where to start, where to find the right end of the thread. Indeed, as the watch gets structured and becomes more and more relevant, the different sources refer to other relevant sources. It is as if a bonus to the winner was given : the more evolved and structured the watch is, the easier it gets to improve it ! And suggestion algorithms proposed by watch tools (Google reader, Twitter) strengthen this paradox : the more relevant the sources already composing our watch are, the more numerous the sources automatically proposed get. In order to limit this pernicious effect, it is more interesting to start spotting sources used by watch on close topics rather than starting from a search engine. That is all the interest of shared web watch. This sentry has put on-line a RSS feeds aggregator on a theme close to mine, what are the sources he used ? I think relevant the web watch used by this user of Twitter, who is he following ? Who is he followed by ? So, gradually, groping along, one makes his own selection benefitting from the selection work already done by other sentries and adapting it to his own needs.
3.3 Evaluate sources
3.3.1 Problematic of the evaluation
3.3.1.1 The hardware world : a known world where authority prevails
Materiality of the hardware medium (paper, magnetic tape, DVD,...) implies the resource's "rarity " : a billion books can't be published each time ! Information goes through define and limited channels (by materiality and/or regulations : the CSA attributes radio and television frequencies because they are in limited numbers, an efficient broadcasting of a book needs to be done by an editor (who will take a financial risk) who will go through a distributor to carry the books in well defined point of sale. And the work will need a legal deposit to the BNF. In conclusion, the low number of channels added to an information scene which has not changed for 30 years allows to have marks to easily evaluate information : name of the editor, of author, of the radio station, the kind of periodical and even the model of the periodical are often enough.
Evaluation is done from the authority : whether it is the one granted by the author or more often the publisher, the periodical which broadcasts or the radio or television channel to which it belongs 3.3.1.2 Web 2.0 or dilution of authority and classical marks
There is no publishing left for most contents. Hierarchization of information is often left to algorithms : hierarchy of results with search engines, linked videos on Youtube, etc.
More over, the data carrier, out of reality, is not restricted to only one identified source. Information coming automatically from several external sources (RSS feeds, nested pages) may be found on a single web page without it being obvious for the reader.
The data carrier enables collaborative writing, where the author is not know any more (Wikipedia) but also contents which are automatically generated through data bases : for a same research, an Amazon page will be different according to former researches, purchases already done, purchases done by other users, etc.
Opportunities given by the copy-paste function contributes to this dilution of authority. Some texts are made up of a collection (more or less successful) of other texts.
3.3.2 Evaluation criteria of sources on the web 2.0
3.3.2.1 Evaluation of primary sources Primary sources : sites which contains articles. They are watched on through the site's feeds, web watch, subscription to newsletter publishing new publications.
Evaluation criteria :
Authority : the author, the site editor : I keep an eye on the blog of such author who is specialist in …, I follow the feeds of the Whatnot Institute.
External criteria (when neither the author or the editor of the site are known)
intrinsic quality of the text
consistence with what is already known
vocabulary, text construction
sources
When evaluating external criteria, we need to be careful to the " credibility look " consisting in giving a website all the appearances of credibility. (see Fogg, B. J. 2002. Persuasive Technology: Using Computers to Change What We Think and Do. 1st ed. Morgan Kaufmann) 3.3.2.2 Evaluation of secondary sources Secondary sources : spotted and indicated articles through Twitter, Facebook, a shared web watch, a watch bulletin...
Evaluation criteria : the confidence given to the person who indicated the link.
(Little historical and informative digression:
we have been used to trusting information officer for the choice of selected and available documents : we rarely check the liability or the serious of a work from an information center, assuming that the document has been selected. When the Internet became mass appeal in the mid 90's, the pattern was applied to the web under the form of a directory listing and classifying web sites which were selected by humans (and nevertheless information officers). In front of the scale of the task and the improvement of selections done by search engine, non-specialized directories disappeared for the benefit of search engines. It's the beginning of the Google era when the confidence formerly given to human beings has been transferred to a research algorithm. Then Web 2 with social bookmarking or data curation tools but also web watch through social networks (particularly Twitter) reaffirmed the profit of selection.)
In this environment, shared space and/or collaborative web watch shows all its strength, because selection of information is done within the community of interest and then besides the relevance, it's the level of confidence that is high and which endows all its value to that kind of watch.
3.4 Classify and order one's watch
We are showered with information flows, sources are multiplying thanks to web 2, there is an overflowing, a flood, so to avoid drowning, we need to do a little plumbing...
Spotting relevant sources thanks to tools, filtering, gathering and bringing them to us.
3.4.1 In practice : aggregate sources in the same spot : Padawan learner
Tools to aggregate feeds and/or flows from social networks :
all the flows : iGoogle, Google reader
automatic publishing : Paper.li, Feedly
This allows to optimize the time dedicated to the watch and to extract it from email reading, because the time spent reading emails is not necessarily the time dedicated to web watch.
3.4.2 In practice : filter, reconstitute sources : Jedi apprentice
Gather in one several flows
Filterone or several flows to reconstitute another one (Yahoo Pipes)
Create an new flow from a shared web watch (Delicious, Diigo)
Connect several web 2 services (Itttf)
3.4.3 Recreate a human scale informative space in the web ocean
The risk of this sorting is to reach only what is know, comfortable without facing other opinions or visions. But here again, collaborative web watch helps us to avoid this pitfall :
sources are many (all the sources from all participants)
But information is selected
It's the opportunity with sources multiplicity (outer intelligence, aeration) without the weight of infobesity!
Same thing with social networks but with less relevance (and therefore more noise)
A serendipity that increases more on the internet than in the real world. And we must accept it !
3.5 Select to avoid drowning !
To avoid drowning and still find your way around, one must learn to evaluate information coming from secondary data : You must not read everything !
3.5.1 Selection criteria
Secondary data : title, comments, etc.
Hence the necessity to make them as clear as possible
Redundancy
It's no longer an enemy(noise) but an alert on the importance of the information
I read or not / I keep or not / I transfer or not
3.6 Spread without drowning others !
Beware of the "God will recognise his own people" syndrome. Don't send everything to everybody thinking : "they will select" or "there will surely be someone interested". The content and the audience of what is spread must be selected. The opposite is counterproductive. That is the great advantage of collaborative web watch, the audience is well known and identified. Only what is of interest for the group must be spread.
What is of interest for us should be spread on social networks and not what could potentially interest those who follow us because choice is subjective, this publishing is the capital gain of our web watch. Social networks are the place for an assumed subjectivity, its actual strength.
How to spread your web watch ?
PDF bulletin automatically generated from a flow : Zinepal
Avoid newsletter-shaped email
Collaborative web watch : use of a key-word specific to the group to redistribute the information which is especially dedicated to the group : Diigo, Scoop.it
4. Capitalize
4.1 Everything must not to be kept ! Change in position
Two logics :
Logic of collection : in a world of rarity, we try to gather everything in the same place (library). It's the world of paper backup.
Logic of flow : in a world of abundance, tools and methods are implemented to spot interesting information. It's the world of online backup.
Furthermore, the value of writings has changed. In the real world, we speak more than we write. Most of the time the written work is a trace, a protection. On the web, we write more than we speak but it is a written work which is equivalent to the words we used to say, it's particularly the case with emails, far less formal and far more fleeting than a written mail. You don't keep an email saying "Are you coming Saturday? Neither do you record the voice of the person who is asking directly. The value of writings on Twitter is more like the value of information on the radio : you don't record the news from France Inter because you did not ear it. If I don't go on Twitter for a week, I am not going to read all the tweets of the week, nor feel guilty because I missed them. And conversely, the value of audio information changes too. With podcast, I can listen to a radio programme months after.
4.2 How ?
We know perfectly how to manage information coming from "former " channels : written mails, telephone, etc. We don't know yet how to manage the disproportionate e-arrival of information. A knowledge managing method has then to be implemented. They are many of them, the GTD method for instance can help. Must choose one method, whichever one. Once chosen, tools enable to implement it.
5. But I don't have time !
Automate all that can be.
Machines are there to relieve humans from repetitive tasks and enable them to benefit from more time for really intelligent. For example, take time to settle filters on your electronic mail to get rid of spams or to tidy up in files emails from newsletters which you will read later. Chain your web 2 apps to post your web watch in only one spot so that it can be sent automatically to other media. Most services offer it but you can do it more widely with services like IFTTT. Example: I had a piece of news to my shared watch on Diigo, it is automatically published on Facebook, Twitter and on my watch's feeds.
Define a time in your schedule for watch and respect it ! (i.e turn down if possible any meetings during this slot: "no, I am not available") If you decide to do your watch just when you don't have anything left to do, you'll never do it.
Take advantage of others watch thanks to shared web watch !!
Focus on collaborative web watch
Gathering : indicate
While doing your watch, spot what could be of interest for the community
Key-words: Delicious, Diigo
Hashtag : Twitter
Sharing with a group : Facebook
Use redundancy : Yahoo Pipes
Spreading
RSS Feeds : aggregator, web site
Scoop.it, Paper.li
Facebook page
Email (can be automatically generated by the feed)
