search

dreger / 1000-nights

📖 My attempt to document and complete Ray Bradbury's 1,000 Nights.
235 stars 19 forks source link

Nice idea, but lack of structure makes collaboration difficult #3

Closed pgchamberlin closed 10 years ago

pgchamberlin commented 10 years ago

Hi,

I really like this idea, Bradbury's idea is wonderful, but I think maybe you could structure this project in a better way to encourage collaboration - it's difficult to know how to make suggestions / pull requests at the moment.

Maybe a folder structure would help? Possibly something which separates each of the days' study out into separate files like:

Obviously this would lead to a lot of manual duplication - you'd need to manually maintain the index and study materials - but it's better than nothing.

The benefit to having different files for each day of the project is that you can then include the texts themselves directly in the project (where copyright allows) without cluttering up the project. In fact, it might be interesting to try to do this project entirely with public domain texts - using sites such as Project Gutenberg for example.

Anyway, nice idea :-)

techwraith commented 10 years ago

This would make creating an RSS or Atom feed from this repo much easier as well. Giving this data some sort of structure would allow us to do things programmatically with this stuff :)

pgchamberlin commented 10 years ago

That's a good point - it could be a starting point for an API of some sort / the basis for reader apps.

dreger commented 10 years ago

I agree -- something should be changed to allow for more structured collaboration.

For the time being, I'm going to ask that new contributions come in the form of new issues, which I'll go through and add to the resource list / schedule.

Long term, however, I'd love to have this data be exportable by way of JSON or API. Let's use this thread as a point of discussion for both these ideas.

techwraith commented 10 years ago

Well to get started, it would be pretty simple to parse the markdown files into some easy to work with JSON. From there we can just keep those JSON files in the repo and people could use them in whatever services they'd like to build.

I don't think prescribing an API for people is the right way to go for this. I think it's best to keep this as a simple data repo.

dreger commented 10 years ago

@Techwraith I like it; that sounds like a more feasible short-term plan.

As per the structure of the repo, here are my ideas:

Have there be two levels of directories (/short-stories/author-name/) and then have individual .md files for each work that goes with that author. If permissible, we can put the original text in those files, but, at the very least, we can put the link to the online resource.

Aside from being more organized, this is nice because it lets the "link" on the schedule.md page point to a file in the repository rather than an online resource. Although the local file may just contain the link to the text, it makes it easier to update should we ever need to change a link / text out.

techwraith commented 10 years ago

Sounds like a good plan to me, and it's really easy to get at programmatically.

dreger commented 10 years ago

I've updated the repo structure to help with organization. Things should be much better from here on out. Check out the READMEs in each of the main directories (short stories / poems / essays) to see how I've set up author / piece organization. Thanks for your help, happy forking!