Remove discontinued series from "similar series" list.

Jemini

Well-known member
Joined
Jan 27, 2019
Messages
2,037
Points
153
I can get why a low view-count series may appear in the "similar series" list, it helps the little guys to be discovered. However, if they have stopped updating their series and it's not because they have completed the story, I think there ought to be a way to have it removed from the "similar series" list. I'm all for directing traffic to some little guy who's trying hard, but a small time series from an author who's not even trying anymore is just taking up space for someone who would be more deserving of the attention.

Also, if the "similar series" list is always populated with these lower quality works where even the author is not taking it seriously, it makes people stop paying attention to that list. These sorts of things are supposed to be recommendations to the reader that "if you enjoyed reading this, you might enjoy this other series as well." Most websites with this feature will track what the users are watching, and recommend other series enjoyed by the same readers. In the case of SH, you have the "reading" list. This would make setting up such an algorythm easier here than it would be on other sites since it could simply track other series that appear on the same reading lists. (Not sure if the servers are strong enough to run all those numbers, but it strikes me as a viable solution.)
 

Jemini

Well-known member
Joined
Jan 27, 2019
Messages
2,037
Points
153
On paper, I agree with this.

But I guess I just have some queries - and these aren't necessarily directed at you, just open questions for discussion. Because. You know. Internet discourse, of course I have to specify it's just open discussion questions.
  1. Out of curiosity, when dealing with authors who do not mark their stories as "hiatus", either because they "intend to update" it but haven't, or authors who have gone completely MIA, how would the a l g o r i t h m decide who's "not even trying anymore"? Would it be X months since last update - in which case authors could game the system by periodically uploading chapters that are just announcements? Which I suppose would still filter out the abandoned stories anyway. Or would it be X months since author last logged in, accounting for authors who are totally MIA? Or a combination? Or a system where users can flag stories to be marked "hiatus" and if the author doesn't respond in x time, then the story switches to hiatus?
  2. In the case of cross-referencing user-lists, this doesn't seem very useful at all, at least for the purposes of, in your own words, "some little guy who's trying hard". If anything it seems backwards. Stories with more people reading it i.e. specifically people who have it put on a reading list, would be more likely to pop up in "similar stories". If there is someone who has a story that has 3 people reading it, then it's less likely to get flagged to show up for recommendation because there's not enough of a userbase to cross reference. Not to mention the question of how to handle the situation of people who have reading lists specifically for stories they don't read, to have them marked to avoid showing up in updates and whatnot. So stories people specifically blacklisted because they hate them also end up getting recommended? Besides which, I personally hate the idea of getting recommended stories based even partially on "x user read this story you're reading, but they were also reading this other story and since you have this one single thing in common that you both read maybe every other garbage pile they partake in will take your fancy", because, well. I just don't think they give accurate recommendations.
I'm pretty sure the "similar series" list is based off a cross-reference of shared genres and tags, which I think is the best way to do it especially because they're literally similar because they share genres. The problem isn't how the stories get selected. For me, the problem is that the same handful of stories are constantly being recommended as a similar series. Which, again, works against this idea of "exposing people to a variety of stories they might not have looked into". If there was a simple way to just mark on the user side "I'm not interested in this story" to avoid having it pop up over and over and over well past my interest in it, that's what I would prefer. On twitter (and twitter is a hellsite but let's put that aside for now), they recommend other users to follow based on things like "x people you follow are also following this person", but in the "who to follow" window, if you're not interested in following a certain someone that gets recommended, you can hit a little "x" and that person gets removed from the list and replaced with another recommendation. People you hit "x" on show up again sometimes, but are much less likely to.

I'm not saying to use this system because following users on twitter isn't the same thing as being recommended stories on SH, but some kind of system to mark a story as "I'm just not interested" to have it not pop up over and over would be nice. These aren't stories I hate enough to put on a blacklist reading list, I just don't have even a passing interest in them as it were.

Well, I completely admit the 2nd paragraph suggests something that goes completely against the stated intention of the 1st paragraph. This is mostly because the 1st paragraph was intended mostly toward helping the authors, and the 2nd toward making this site more reader friendly and attracting and keeping more readers. I should probably have specified as much.

Another option I just thought up that would likely be far more simple to implement would be to somehow allow the author to remove a series from their own "similar series" list, and have it replaced automatically with a different one that meets the same judgement criteria. This would not necessarily be a fair system, but it allows it to be as fair as the author decides they want to be. So, sorta a case of relying on the good will of the author. I do get the feeling though that most authors would know the struggle of other authors and would be more likely to show some good will.
 
Top