topbanner_forum
  *

avatar image

Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
  • Tuesday March 19, 2024, 3:50 am
  • Proudly celebrating 15+ years online.
  • Donate now to become a lifetime supporting member of the site and get a non-expiring license key for all of our programs.
  • donate

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - alex3f [ switch to compact view ]

Pages: [1]
1
Living Room / A like minded micro-donation community omidyar.net
« on: November 24, 2006, 08:19 PM »
I had joined the Omydiar network yesterday and find that it is like-minded project to DC. Here is a link:

http://www.omidyar.net

We believe every individual has the power to make a difference.
We exist for one single purpose:
So that more and more people discover their own
power to make good things happen.

People also can donate micro-credits to the authors of useful posts. It is easy to register, so I suggest you to try it out.

2
Warning: it's 1.5mb big and very short [editor note: but worth it - mouser]

2463_startrek-bluescreen.gif

from http://sake906.assau...rtrek-bluescreen.gif

3
Living Room / Re: How Digg Gets Everything Backwards.. And How to Fix It
« on: September 19, 2006, 11:34 PM »
Here is a case study of flickr that is somewhat related to this thread. It cites catering to the power users as a major success factor of flickr. It is quite likely that the same factor (or maybe strategy) played a major role in digg popularity.

http://www.startup-r...ut-tech-for-exit.php

Once power users realize their power to influence the project, they often become its evangelists and actively promote the project by recruiting new members. Promoting such a project means increasing their influence, which is a stong motivating factor. If the system is truly democratic, and hard to abuse, promoting the project only decreases the influence of each individual member. This might explain why truly democratic projects remain relatively small, while projects appealing to democracy, but allowing for abuse, grow very rapidly. However, this kind of rapid growth  doesn't necessarily lead to diversity of ideas and "wisdom of crowds" phenomenon as the project is not utilizing abilities of most of its members constructively.

4
Living Room / Re: How Digg Gets Everything Backwards.. And How to Fix It
« on: September 11, 2006, 04:35 PM »
You are welcome, housetier. I have a paper on the web explaining the mechanism these systems use. You may find it interesting: http://hbga.org/hbga.html

5
Living Room / Economics of virtual communities
« on: September 11, 2006, 01:52 AM »
This article by Andrea Ciffolilli looks into economics of virtual communities. Based on example of Wikipedia, he tries to understand how these communities can be far more efficient in producing public goods than the traditional institutional solutions. It seems quite relevant to the dc community.

http://firstmonday.org/issues/issue8_12/ciffolilli/index.html

Phantom authority, self–selective recruitment and retention of members in virtual communities: The case of Wikipedia by Andrea Ciffolilli

Virtual communities constitute a building block of the information society. These organizations appear capable to guarantee unique outcomes in voluntary association since they cancel physical distance and ease the process of searching for like–minded individuals.

In particular, open source communities, devoted to the collective production of public goods, show efficiency properties far superior to the traditional institutional solutions to the public goods issue (e.g. property rights enforcement and secrecy).

This paper employs team and club good theory as well as transaction cost economics to analyse the Wikipedia online community, which is devoted to the creation of a free encyclopaedia. An interpretative framework explains the outstanding success of Wikipedia thanks to a novel solution to the problem of graffiti attacks — the submission of undesirable pieces of information. Indeed, Wiki technology reduces the transaction cost of erasing graffiti and therefore prevents attackers from posting unwanted contributions.

The issue of the sporadic intervention of the highest authority in the system is examined, and the relatively more frequent local interaction between users is emphasized.

The constellation of different motivations that participants may have is discussed, and the barriers–free recruitment process analysed.

A few suggestions, meant to encourage long term sustainability of knowledge assemblages, such as Wikipedia, are provided. Open issues and possible directions for future research are also discussed.

6
Living Room / Re: How Digg Gets Everything Backwards.. And How to Fix It
« on: September 09, 2006, 07:20 PM »
Since we are talking about diggs idea, I would like to post a link of its less known precursor: http://intermix.org Intermix stands for INTERnet Metropolitan Information eXchange. It is a free software authored by Roger Eaton around 1999 or even earlier. It implements collaborative evaluation and ranking of news, the same idea that later was picked up and explored by digg. Intermix has less users, but has some advantages over digg. I like that Intermix allows to rank stories by two criteria: interest and approval. To my knowledge none of the recent collaborative filtering systems has this. Sometimes I discover a website that is definitely interesting, but I don't agree with its POV. Since most sites interpret bookmark as an agreement (digg) or don't differentiate between the two (SU), I normally will not publicly bookmark such a site. Intermix is flexible enough to accomodate such cases.

7
Living Room / Re: How Digg Gets Everything Backwards.. And How to Fix It
« on: September 09, 2006, 06:22 PM »
While dig is about taking a consensus and presenting what everyone must/will like based on that consensus, Stumbleupon takes the whims of the users and uses it to attempt to serve, based on the individuals preferences

I agree with you. In fact, SU has the equivalent of digg's front page (check http://buzz.stumbleupon.com/) averaging the preferences of all users, but it is not the main point in SU unlike digg. SU buzz page is better implemented than digg's buzz/front page. For example you can't vote right from the SU buzz page, you will have to go and see the website first. DIgg makes it too easy for you to vote without even reading the story. Digg interface is thus conducive to information cascades, while SU design will break at least some of them. As a result SU has less dependent sample of user preferences to aggregate, than one which digg collects. Hence, SU can have a better aggregated estimate even with a smaler sample size. Of course, as you mentioned, the main strong point of SU is that it recognizes different clusters of users with shared preferences, instead of pulling all the users into a single cluster as digg does.

8
Living Room / Re: How Digg Gets Everything Backwards.. And How to Fix It
« on: September 09, 2006, 01:09 AM »
IMHO, this is not Digg wonderful idea. This is an idea of collective decision making (includes collaborative filtering). It was originated long before digg, and long before the Internet, and allows different implementations. Digg's implementation is not the first and also quite poor, too vulnerable to abuse and information cascades.  http://www.shmula.com/197/digg-as-a-game has a good analysis and two good suggestions for digg. On the other hand, the suggestions are pretty much strightforward and were implemented at my site right from the beginning and long before digg registered its domain name. But maybe  the popularity of digg, at least partly, is related to its present design that is so prone to abuse and decision errors. It seems that the goals of attractiveness and quality might be in conflict in this case. I wonder, if digg was better implemented and protected from abuse from the beginning, would it be so popular?

9
Here is my take on this. I see the Google advertisement program as organically complementing the Google pagerank system. Basically, when searching Google your attention is split approx 80/20 between older content that received high human evaluation and a new content selected automatically for relevance and not yet human-evaluated. In this case, Google gives you a choice between older and more reliable, and newer and less reliable. You can either choose to exploit or to explore. I think this is quite reasonable because human evaluation through placing links and getting high pagerank takes time, advertising allows to buy this time through buying people's attention. In a sense, investors bet on the value of their site to force its faster evaluation by people.

10
I agree that digg is too imitative. In addition, Digg allows only to promote a website, but not to demote it. Once some website of little value is promoted by a small group of people of critical size, it is bound to go up just because it has that critical mass. This is similar to stock market where playing long is encouraged, but playing short is discouraged. Together with high level of imitation in both cases, this design is known to produce bubbles. The other design for web aggregator would be to allow users to both promote and demote. I used it to make yellow pages that adapt to user preferences, reddit.com uses it for website ranking. In addition, the total counts of promotions and demotions in this design are not shown to evaluators, and this makes evaluation less imitative and less subject to information cascades than in digg.

11
I think, success of collective intelligence and social websites in particular is best described in terms of evolutionary computation. There are some minimum conditions necessary for evolution to take place. These are diversity, change, and selection. Internet websites satisfy diversity requirement quite easily due to the global nature of the Internet. In addition, these websites are built in a way to encourage humans to perform change and selection operators. In digg's example, the activity is structured around suggesting a new message (change) and promoting an existing message (selection). This provides for the simplest evolutionary process that will slowly drive the population of messages toward increasing fitness. Most social websites I had seen so far, implement at least a minimal evolutionary model.

If we look at a crowd from the same perspective, we notice that it lacks every necessary condition. Members of crowds normally lack diversity, they often come from the same background and have quite similar opinions. It is often homogeneity that brought them together in the first place. It is quite unsafe to try to challenge the opinion of a crowd so crowd doesn't provide some change mechanism. There is not much opportunity for selection either due to lack of diversity. Crowds are subject to positive feedback and information cascades, see James Surowiecki "The wisdom of crowds" (his analysis why crowds fail to be intelligent).

12
IMHO, the arguments about wisdom of crowds mainly result from a wrong word choice.

Crowds are large and very loosely organized groups of people. I think, it is not anonymity per se, but rather perceived individual ineffectiveness and irresponsibility makes a crowd. In a large and disorganized body of people, it is easy to think that individual decisions has no impact on the group behavior. As a result, people don't feel any responsibility and actually stop thinking as individuals. In this situation the crowd is like a superfluid liquid free to move in any direction without any reason and resistance.

This is not what we see in many social websites. I think it is quite wrong to use the same term for participants of a social website. In every case, where a group of people is intelligent, we find structural organization in their activity. People at those websites are organized by social software that structures their interactions. Of course, it is not a traditional organization motivated by managers through employment obligations and compensation. Instead, this is a participatory organization, where social software coordinates people's activities in a more subtle way. This can be done just by altering the effort associated with different kinds of activity.

IMHO, social software has some advantages over human executives. It is more scalable, more responsive, fair, and transparent.  It can be protected from corruption at a much lower cost than human management hierarchy.

Pages: [1]