hello /g/, i come from a Holla Forums background but i am pretty versed in the gentoos so let's talk face to face
as /lgbt/ as this may sound, i am here to tell you what pisses me off and to propose a solution.
take a look at pic related. its what i could find in 2 minutes. but the issue is MUCH broader.
basically most major media outlets are fighting a fight against free commenting. whereever you see a slightly controversial article, the comments are deleted by the thousands.
i mean i would understand if they """flagged""" """offensive""" comments and you'd have to click on it to make it show, as a way to make the general experience """friendly""" (whatever that may mean). but they are blatantly deleting and censoring the comments.
i understand some comments might be spam or off topic. but if the sites internal voting system gives the comment 30 "stars" or whatever, it's not fucking spam, gtfo telling me what i can or cannot read
---
anyways, this """idea""" has been lingering in the back of my head for several years (which means its not a fad but still urgently needed), its blatantly simple and most likely exists in some for or another but i came here to gather some opinons on how to best design some system like this.
basically we would want a system that is
1) separate from nytimes/gawker/reddit/facebook/etc.
2) but allows people to comment on particular content on nytimes/gawker/reddit/facebook/etc.
3) has one of the common ""selection"" systems to filter out spam/shitposts/bait, for EXAMPLE reddit-type voting system (its an EXAMPLE you goddamn faggots, i know this particular system is an echo chamber for popular faggot opinions)
this would tilt the momentum back into our hands, because at the moment, the globalist media owners and goverment censors are winning.
IM FUCKING FED UP WITH OPENING A NEWS ARTICLE WHOSE CONTENT MAKES ME RAGE, ONLY TO FIND THE ADMINS DISABLED COMMENTS FOR THIS PARTICULAR ARTICLE.
i mean we need something like disqus but hosted separately, not embedded in the website itself.
challenges to solve:
1)links may take slightly different forms (added parameters, etc.) but still point to the same content. do we manually keep per-site algorithms that figure out what goes together? do we just ignore the problem altogether and consider all links unique? or do we develop some algorithms that detect links that might point to the same content, say, they both are from the same domains and both have the ID 54828737 somewhere inside, and then suggest to the user to join those comments?
2) do we run this system decentralized? it would greatly improve availability, and maybe we could implement a protocol that could dynamically accept new versions and extensions, say, for now, we just record the date and the content. then somebody wants to add an author name field. this field will get sent with the information and old clients will just discard the new fields?
i know im drifting into details, and i would prefer to build something simple and working at first, so i stop here.
i asked at the fourchins but all the replies very cancer AF, so i came here.
