RSS is easily replaceable by just scrapping a website, technologically speaking it doesn't make sense to have everyone making a request to website every minute. We could have a central service responsible for that, but I guess it will centralized and some people don't like that
A custom analysis by the user - maybe tech illiterate -, from every user, for every, supposedly complex, website, instead of some xml-izing wrapping over 'SELECT * FROM news ORDER BY pubdate DESC LIMIT 100'?!
> a request to website every minute
RSS takes less bandwidth than doing the same operation through direct visit. And of course, one is not supposed to do it that frequently; though again, some browsing agents may load pages more frequently than every minute.
I can only condone scraping if the official API is so convoluted that scraping the page is literally the easier option to program against.
If gwene goes under, it'll suck but I'll have a gwene like on my own server. If RSS goes under, it'll go under here and there, and not wholesale. Good technology is resilient like that.