Can extinct species be added later?
Yeah, but the standards for that particular server seem to require several citations for the placement. Genes, while only salvageable from recently-living organisms, are an independent line of evidence that gives more certainty than the guesswork involved in placing fossils.
If they're keeping species out of the tree due to incertae sedis or uncertain placement, I'm guessing that's why.
According to the paper, they were using a combination of molecular and morphological methods. If they were only using genetics, I'd understand, but since they're using morphology in their classification it shouldn't have been a problem to include fossil taxa.
You're lucky to find a few bones of a given fossil specimen, though. Extrapolating that to the description of an entire species leaves uncertainties that this server would rather not include in the tree.
It's an open format, though. Anyone can make their own tree however they like it.
I wasn't suggesting that anything known from "a few bones" should be included in the database. However, that is not the case with most of the taxa I would want to include.
Most of the taxa I mentioned above are complete enough to be used.
Tyrannosaurus,
Triceratops,
Allosaurus,
Diplodocus,
Stegosaurus,
Dimetrodon,
Archaeopteryx, and
Homo erectus are all known from multiple near-complete specimens.
Mammoths are even harder to explain, as is the Shasta Ground Sloth (
Nothrotheriops), also missing. These are not only well known from complete remains, including soft tissue, but genetic data has been reported for both of them (frozen mammoths from Siberia, which recently allowed the complete genome to be sequenced, and ground sloth dung from a cave in California). There is an "
Elephas primigenius" on the tree, which would appear to be the woolly mammoth, but sunk into the same genus as the Asian Elephant (which is tough to explain as the two lineages have been separate for 5 million years), but no Shasta Ground Sloth. Neanderthals are missing as well, despite the complete genome being available.
Australopithecus,
Tiktaalik, and
Ambulocetus are not as complete, but the important parts are there and their phylogenetic position is pretty clear from what's known. Moreover, I would argue that those three are so important to our understanding of evolution that it is a serious oversight not to include them.