How do we build a social web? - Information ranking: points of view and bumps of anger

One of my first managers at critical moments of development of a certain project liked to ask employees: “Will we work or feel?”. After publishing my “ duel ” note with “feelings”, I finally got stuck :). It's time to start working! A series of my notes "How do we build a social web?" Was aimed at this very work. It discusses the general approaches to the organization of an intelligent social resource (SRI), independent of who builds it and for whom, and what functions are laid into it. Below is the second article in this series.

And where is the first, you ask? Indeed, why in our collective blog, I start posting notes in this series not with an introductory note, in which I would need to be more streamlined and formulate the general principles for creating an intelligent social resource (IRS) in a refined form? Frankly speaking, the note with the principles is already ready, but I decided to refrain from publishing it. And I do it for "political" reasons.
An interested reader can find a more detailed answer in my iTech Bridge blog . I emphasize, interested, because the entire series is focused only on such readers - potential partners, members of the collective blog “Intelligent social web” and others who agree that all is not well in the kingdom of the social Internet. I recommend everyone not to waste time - the notes are very specific, and therefore, despite the concise nature of the presentation, they turned out to be quite long ..

So, I start with the first group of specific requirements for the IRS - requirements for ranking information.
')
What are we talking about

Information that, in one way or another, is more or less ranked in any social resource, is, first of all, a different type of message: basic (initial postings, topics / topics) and comments to them.

In general, all major messages (postings) can be divided into the following types:
• Original announcement , including:
- interview
- translation
• Original message - Questionnaire (questionnaire, vote)
• Link:
- without annotation
- with a brief annotation
- with expanded annotation (abstract)
- to a foreign language source:
- with a brief annotation
- with the essay.

Speaking about the ranking of any posts, we mean three groups of questions:
1. Toolkit for assessing information provided by the resource to users
2. Criteria for ranking information , which are used by the system (resource), which streamlines the main messages based on their direct and indirect user ratings. This also includes the question of how the system should respond to user ratings that do not directly affect the ranking.
3. Options for issuing ranked information from the system’s memory to the screen (listing, rss feed, etc.).

Sequentially consider all three groups of questions. Take for this purpose six real resources with more or less intersecting functionality (digg, reddit, news2, habrahabr, del.icio.us, bobrdobr.ru) and analyze how they answer questions from each of the three groups. And then for each group we formulate the requirements for an ideal (again, from my point of view), but so far, unfortunately, an almost mythical intellectual social resource (SRI). In this case, as I have already noted, the functionality of the WBS in our case does not play any role. In this article we start with the first group.

Hummocks anger

Let us return to the quote, which was the final drop that made me make a direct challenge to the little evil men. But now, as I promised, let's talk about something else entirely.

Taking into account the words covered up in the above quotation as “fig leaves” (I call them hints of anger ), such a comment would certainly have been classified as dirty, and its author was removed from the number of resource users. Even if we met in the message only one such bump . This is my deep conviction. Who and how should execute the sentence is a separate issue that deserves a special note. Here, running a little ahead, I can only say that weaning from a resource is a heavy punishment. Despite the possibility of incarnation. After all, at the same time, everything earned by hard informational work is lost. Your rating and related features are lost. Trust me through this :). By the way, this is why I offered my opponents such “tough” conditions of a duel.

In the meantime, let's see how the user can now, who agrees with my “punitive” point of view, respond to the appearance of such a comment.
In order to directly mark some message as containing bumps of anger, we have with you:

If this is the main posting :
digg - you need to consistently select bury and then spam
reddit - you need to select report
news2 - need to consistently choose a problem? and then spam
habrahabr - no direct opportunity
del.icio.us (bobrdobr.ru) - no opportunity

If she met in the comments :
digg - no direct opportunity
reddit - you need to select report
news2 - no direct opportunity
habrahabr - no direct opportunity
del.icio.us (bobrdobr.ru) - comments are not used

As we can see, the three resources offer some opportunity to point out inappropriate information. However, only reddit allows you to do this not only for the main posting, but also for the comment. I haven’t yet tested the reddit response to the report mark. As for the rest, the reaction is clearly slow. And even the "UFO", which sometimes sometimes “flies” in to the habrahabr and removes the “bad” comments, can be a way out, only if they are controlled by living beings, and not robots. After all, the latter do not know how the “bad” messages differ from the “good” ones. Thus, in the overwhelming majority of cases, you have three options: either to come to terms with the need to constantly stumble upon dirt, or stop using the resource, or put together a “funeral team” from friends and, eventually, “dig in” an obvious violator, using the legal means described below. Including, and calling for help "UFO", controlled by robots. True, reddit gives another opportunity - using the hide command, you can simply remove everything from your view that you consider unacceptable. But this is poor consolation.

Points of view

Now let's dissect our test subject a little bit. The result without the bumps of anger may look something like this (I hope that I won’t have copyright issues :)):

Sir, with all due respect to you, I cannot fail to attribute your position to the fascist ideology. Therefore, I strongly recommend that you, as soon as possible, repair your time machine and, using this mechanism, go to the 19th century, so beloved by you.

And now we move with you to the most interesting. What we read is already a point of view . You can agree or disagree with her. But nevertheless, anywhere from it you will not get to. What do you propose today to express your attitude to this message? Read about it in the extended version of the note , placed in iTech Bridge (Attention! Once again, I warn you, the note was quite long).

And, as always, read the full text of this note in iTech Bridge (Attention! Once again, I warn you, the note was quite long).