roweb
roweb copied to clipboard
ropensci
Website todos in May
I'm afk from the 10th to 17th but will be back afterwards. Here is a rough plan of things I'd like to get done on the website in May.
Major
- Finish a first iteration of the three themes (things you can do with our packages). Starting with the mapping and spatial tools, then the data tools, finally a page on the reproducibility tools.
- Finish a tutorial page for all the remaining packages
- Fix the sidebar on the tutorials page. This is a design issue and I'll see if I can work on making it read headings from the markdown and autogenerate. I'll also make it float all over the page.
- Explore possibilities of OpenCPU widgets so people can try out code inline.
Minor
- Finish out events page. I have a plan to do this, including a form where people can request a larger event (since we are not doing smaller workshops anymore)
Approximate milestone
June 1.
sounds good.
I can help with finishing getting tutorials up
Automating linking sidebar to headings in markdown will probably require standardizing headings used, which we haven't really done yet. like maybe you want to get all H3's or H2's
I played with the opencpu thing at the hackathon, and have a basic template. Ramnath also suggested we want to limit to just a few interactive code widgets on a page (e.g., making each part of a tutorial interactive would make page slow I think, or maybe get complicated, don't remember what the reasoning was), or perhaps open a new window with the interactive bits.
I'm hoping to figure out a way to generate these tags without standardizing templates. A little Ruby to read the H2 and H3 tags and nest them accordingly. This will take a good bit of work, since I am quite far behind the rest of you on Ruby (especially @cboettig) but I'm hoping to use some downtime in May to learn new things. If it works, it frees us up to use non-standard headings (although I would still prefer that we use some standard headings like Installation, License, Bugs, etc).
Great on the opencpu templates. We should put those through some solid (internal) testing before we deploy to everyone else.
@karthik Like you say, it should be easy enough to grab all h2/h3 tags and write links for them (see the nokogiri gem for parsing HTML)
When the docs project is done we should have a gem which should make it easy to write your plugins (as well as your posts) in R rather than pure ruby if you prefer.
Alternatively, pandoc will generate a TOC in html with these links, so it might then be just a matter of CSS to place the resulting <div id="TOC"> in the desired nav element... (but then we're not using pandoc as the parser right now I think).
Nice, thanks Carl. I'll looking into nokigiri shortly. Right about pandoc.
I'll take a crack at this and get back to you guys with more questions and thoughts.
bump. Tutorials are done (thanks @sckott) but maybe we should revisit the utility of having openCPU integration on some pages. Should we bring Jeroen into this conversation?