hfuunfug / unfug.org

Site source for unfug.org
5 stars 5 forks source link

structure of talks page #45

Open fkemmer opened 8 years ago

fkemmer commented 8 years ago

I'd love to see upcoming talks in reversed order (i.e. the very next one on top). Moreover, historic talks should be divided into terms to find them easier.

Also, as long as there's "no description", they shouldn't be links.

Lastly, adding the speaker right next to the title could be beneficial.

matthiasbeyer commented 8 years ago

Upcoming in reversed order

  • [x] This should be rather simple to implement. Will do.

Historic divided into terms

  • [x] Rather complicated. Not necessary for the first publish of the page, is it?

No description => No link

  • [x] Even more complicated. That's because the links are generated from the pages, so if there is no page there is also no link. But not linking to a page which exist seems the wrong attempt to me.

Adding speaker next to title

  • [x] You mean in the list of talks, right? That's also a bit complicated, because you would need to read the content of the item. Maybe nanoc provides this data. If not, this would need more effort to implement.
matthiasbeyer commented 8 years ago

@fkemmer I implemented 3/4 things you wanted to see implemented.

fkemmer commented 8 years ago

Historic divided into terms

Rather complicated. Not necessary for the first publish of the page, is it?

Is it? There are very specific term boundaries (which, to my best knowledge, haven't changed), which should allow something like "if last talk in February and next talk in March: print a separator in between"

However, I'm unaware of the internals of the software, so this might in fact be rather complicated

matthiasbeyer commented 8 years ago

However, I'm unaware of the internals of the software, so this might in fact be rather complicated

It is, a bit. I will try to hack this together, but this adds complexity to the page ... which shouldn't be that much of a problem:

$ time bundle exec nanoc
[nanoc]: Site compiled in 1.57s.

real    0m2.127s
user    0m1.988s
sys     0m0.130s

So I think it's worth it.


Edit: Above was a fresh compile, without the cached items around.