This is the first post in a while, but also an important one, as it’s the first time I’m writing under the brand new incarnation of the website, after migrating from wowchemy to quarto.

I started looking at quarto over the summer, when I was preparing my talk at Vibass, when I went to beautiful Valencia, last summer. I had reached a very high level of (probably even higher nerdy…) satisfaction with my xaringan slides, which I still like very much. But I as I started to read about it, I thought quarto had some very nice features and, to some extent, it looked like a bit easier to manage as a whole environment. So I gave it a go and actually quite enjoyed hacking the basic presentation template to make it my own format.

Then, I started playing around with building websites using quarto — and after a very limited struggle to get my head around some of its features, I think it is fair to say that the management of the whole website is simpler using a single engine, rather than fiddling with wowchemy/hugo and then blogdown to render the whole thing.

Porting my old wowchemy/hugo/blogdown website has not been a massive amount of work… The worst part was to deal with blog posts, simply because I have quite a lot of them. But a little bit of Unix bash has helped a lot making sure that some common substitutions in the text of the various .md or .Rmd files was fairly automated (for instance, in wowchemy, I used to add tags to my posts, talks, etc, while quarto does it with categories. So I needed to do some clever use of sed to change all instances of tags: ... to categories: ...).

The other interesting thing I found is that, while wowchemy has been setup explicitly to do with academic websites (and so has a set of nice features hard-coded in how the various pages and widgets are rendered), in quarto I didn’t necessarily had that, automatically. But basically all of it could be dealt with by combining R code (which I could manipulate fairly easily) to include{{< … >}} commands in quarto to add suitable code that would render in HTML the way I wanted.

For example, I can add the “social media” sharing buttons (like at the bottom of this page) by simply writing some R code that uses the icons and adds a a href tag with an automated reference to the relevant website page (using here::here()).

Another thing that, I think, has ended up being nicer than the wowchemy built in procedure is the rendering of the publications list. I now handle everything through pure R and DT::datatable(). I had to fiddle slightly with the underlying data (I did it by creating a yaml file with the information for all the papers/publications, which is closely linked to an overall .bib master bibliography file) and then I can create a highly queryable table (which looks like a “normal” list).

All in all, I like the look of things and I thought I was ready to make this the new “official” look of the website. I may fiddle with this once more as I go along and some of the pages may still need some more updating/polishing, but I think this is close to nerdy satisfaction…