Added a star to favorite comics and a favorites menu

[Akshat752]

[jvarn]

@Akshat752 It's a good idea, but I think the implementation needs some more thought. The context of the navigation and transcripts boxes is the currently displayed comic. Favourites have a different context, so should live somewhere else; but I'm not sure where. My second concern is that (I assume) favourites use browser local storage, which means they are essentially transient. I don't see a way around that, but it is something to think about. Thanks for your continuing interest and contribution. Edit: since Adams was American, I'm not opposed to using the U.S. spelling "favorite" :-)

[jvarn]

@Akshat752 Since starting this project, I've been pondering the possibility of adding titles for all the "Untitled" comics. On one hand, I'm not sure how I feel about giving titles to comics that Adams didn't. But on the other hand, it would make individual comics more quickly identifiable (such as when they were added as favourites). I imagine a generative AI tool could help, but not without a human doing some QC on the results. Thoughts?

[Akshat752]

Hey @jvarn sorry for the late reply. I could move them to the displayed comic context but I'm not sure that would make sense since they aren't tied to a specific strip and are instead meant for the entire site.

Regarding the favorites being transient they will only be removed if a user clears their site data. Otherwise, the local storage will remain their indefinitely. I'm not sure where else they could be stored without creating an accounts system with a database. I could try to implement a way to export the data to a file (or probably just an array) that the user can copy and import into the site in case they accidently clear the site data, but I'm not sure if most people would use it.

[Akshat752]

@Akshat752 Since starting this project, I've been pondering the possibility of adding titles for all the "Untitled" comics. On one hand, I'm not sure how I feel about giving titles to comics that Adams didn't. But on the other hand, it would make individual comics more quickly identifiable (such as when they were added as favourites). I imagine a generative AI tool could help, but not without a human doing some QC on the results. Thoughts?

I'm not sure that generated titles would actually be that useful. I think that would be a bit redundant and would just clog up the search. Maybe it would make more sense to generate keywords since the current search looks for keywords. But in my opinion I still think it would just be a lot of extra data that's not that useful.

However I have also been thinking of how to improve the searching on the site. I think we could start by letting people search for keywords only within a specfic date range. So if they want to search for a comic with the word "shorts" before 2000, they could add something like "before:2000" along with their keyword.

I also think that it would be very useful to try to also store the characters in each comic. For example along with the transcript we could store something like "dilbert,wally,boss" if those characters were in a comic. So if they are looking for a specific strip and they don't remember dialog or the dialog that they do remember occur in a large amount of comics, they could simplify the search to only list comics with specific characters.

Regarding getting character data I want to look into training a model to recognize characters. Obviously I would have to manually review comics to make sure nothing is improperly tagged but I think it would be a worthwhile endeavor.