This is background for gpt.
Here I explain th details of the static site generation schema for st33v.com.

Brifely, the site is built from two 'faircamp's:
 one for the main site and 
 song of the day (sotd), which lives in st33v.com/sotd (i.e thats its base url)

detail:
st33v@cr4y:~/dox/st33v.com$ tree -La 2
.
├── faircamp
│   ├── campsite.png
│   ├── catalog.eno
│   ├── deploy.sh
│   ├── drMorbius
│   ├── eli
│   ├── .faircamp_build
│   ├── .faircamp_cache
│   ├── robots.txt
│   └── st33vTM
├── forge
│   ├── automationUseCase.txt
│   ├── in
│   ├── out
│   ├── script
│   └── template
├── .git
│   └─[redacted for clarity]
└── sotd
    ├── 2016-01-29-pluto
    ├── 2026-01-29-devonian-dunkleosteus
    ├── 2026-01-30-grouse
    ├── 2026-01-30-llmtm
    ├── catalog.eno
    ├── .faircamp_build
    ├── .faircamp_cache
    └── sotd_cover.png

The two static site are held in the two .faircamp_build directories

rsync copies their contents to st33.com and st33v.com/sotd, respectively.

BUT the base site know nothing of sotd, so the --delete directive deletes all of sotd.
This is not what we want.

Question: How can we protect sotd from the ravages of rsync?

There is also a second question around the robots & sitemap generator. Are we allowed to have a robots.txt & sitemap.xml in sot as well?

Or is there a more elegant way to include the entire /sotd path in th first script?