Seems like both a reduction in quality and quantity.
> Canadians continue to learn about politics and current events through Facebook and Instagram, but through a more biased and less factual lens than before and many Canadians do not even realize the shift has occurred.
Although this makes me wonder about the nature of news that tends to get shared on social media. Is it factual reporting or is it opinions we pretend are factual because they’re on the right domain?
All media are under some influence and has an agenda. "factual" information is another way to say that the narrative is controlled in accordance with societal values.
I think the take away is that Canada is loosing the ability to keep macro beliefs in check.
It I'd going it be interesting to see what the long term management of this is.
Interesting previous cases is how we shaped the media landscape after the second world War in order to (attempt to) eradicate nazist beliefs.
If there is an increase in the quality of news consumed it is not the result of the quantity of news consumed decreasing. It would have to be the result of the consumers seeking alternate sources, or some other factor.
> If there is an increase in the quality of news consumed it is not the result of the quantity of news consumed decreasing.
If you read 10x less news from Joe and Billy on Facebook but start reading from professional sources of news, they still might be bad but it will probably be a better article to read.
I'm not sure what you're trying to say here.
You are essentially saying that if you decrease the amount you read but increase the quality of what you are reading, the quality of what you are reading will likely go up. That is bordering on a tautology.
In the situation you describe you happen to be able to consume less news but increase the quality. That is not the same as the increase in quality being a result of a decrease in news. You seem to be confusing correlation and causation.
Seeing that more users are getting their news from TikTok, I doubt that the average quality has increased. Considering that TikTok is under control of the authoritarian Chinese government, the Canadian Facebook ban may have opened the door for a far more concerning situation.
> If you look at the reality instead of aggregated statistics and packaged narratives, you will find that gun deaths in US very much are not of the charts in places where religiosity is off the chart. If anything, these two things are negatively correlated: the more religious a place is, the fewer gun deaths (despite no scarcity of firearms among the religious).
Okay, and in those states, where do those gun deaths happen? The religious parts? The believers are doing the killings?
Look, I’ll make this straight, because there is a lot of willful ignorance here. Overwhelming majority of gun homicides happen in large cities, where the religiosity is low, and very few in suburbs and rural areas, where religiosity is high. Moreover, people committing those gun homicides are very rarely practicing religious people, and neither are most of their victims.
I don't think they are comparable. DuckDuckGo federate the results of other search engines (e.g. Bing) and add its own sauce on top of it. There is a limit to what it can do.
Google invested in the fundamental from the start and they pushed the limit of what search engines can do.
BM25 has been available for Lucene since version 4 thanks to a refactoring that allowed anyone to plug in a custom relevance score if I remember correctly . The big news is that starting version 6, BM25 is the default.
I started using ConTeXt (http://wiki.contextgarden.net/Main_Page) after many years of LaTeX. The command interfaces are more homogeneous than with LaTeX and it is quite easy to do complex layout with it.
ConTeXt is an awesome package. Its "only" drawback is that it's unsupported by publishers. I thought about writing my thesis in ConTeXt, but then I couldn't have easily reused the text from published papers.
I am currently using this toolkit and I must say that I really like it.
The main advantages of mallet over weka (the main java toolkit used in academic machine learning) for Natural Language Processing are:
- No need to map words and features to position in a feature vector yourself.
- Instances preprocessing can be defined in pipes that can be saved along the models. So no need to remember the pre-processing of data for each experiments.
- Contains algorithms for structured learning (CRF, HMM and general graphic models).
On the other hand, Mallet implements less algorithm (e.g. no Support Vector Machines to my knowledge).
In short, it is a nice toolkit to be aware of if you are planning to do Natural Language Processing.
