In a previous blog post I've introduced how the science journals have installed content aggregators to combine existing information of decentralized blogs into a large scale hub. This kind of news aggregation is working with timebased selection. That means, in the newsfeed only links from within the last week are available. For example, the best links about neural networks.
Is a software like Planet plus a website which displays the rss feed the best technical answer to this problem? There is something available which is more powerful and allows to curate the news with manual adjustments. I've found in the internet a small but beautiful project called Unnews:
It is not a real Wiki but some testing playground to try out the idea on a practical example. All the news in this wiki are fake but the website shows how the workflow in general is. What is shown on the main page is a timeline of events. It is equal to a chronological content stream. A new news is created by adding a new wiki page. The article gets a paragraph and the URL to the source. Additional the article is categorized by the subject (for example sport) and by the date (2019-01-02). All the news article are put into the combined RSS stream.
So we can say, that Unnews is some kind of advanced Planet news aggregator. The news are hand selected and the overall wiki system is edited by many people at the same time. The typical Wiki-internal discussion and banning guidelines are used to block spam and nonsense information. Everybody who is aware with Wikipedia know, that the admins are always right and quality first is the goal.
What it's unclear right now, if the idea of Unnews can be transfered into academic domains with the purpose of content aggregation. The hypothesis is that it's possible. Because the result is an RSS stream which contains of chronological news which are referencing to content outside of the wiki. In theory this allows to build a virtual community around existing decentralized blogs.
No comments:
Post a Comment