These are general principles the apply to all textual data enter for THDL work.
Use ONLY Unicode fonts whether Roman script with or without diacritics, Chinese characters, Tibetan script, or another script. Unicode is the world standard for how to encode the world's scripts. See:
For Tibetan, use the THDL Extended Wylie Scheme for rendering the Tibetan in Roman Script with proper spelling (blo bzang grags pa), and use the THDL Simplified Phonetic Scheme for rendering in easy to pronounce form (Losang Drakpa). For those of you who know Tibetan, it will take a bit of time to figure out the latter - do it. For those of you who don't know Tibetan, I just ask you to be consistent and use one or the other.
If you know Tibetan, and you are producing lists of Tibetan terms for any reason, you must produce the phonetic version.
In text-dense materials - essays, paragraph summaries, etc. - Tibetan names should be given phonetically, and on the first occurrence the Wylie version should be given parenethically, followed by a comma with the birth dates and death dates. Thus [Tashi Lhundrub] (bkra shes lhun grub, 1234-1289).
For reference in text materials to place names, personal names, text titles, etc. refer the name to the corresponding WIKI page if there is one. Just get the page's global name, and use the markup [John Doe | globalname]. You should also verify the word occurs in one of the relevant glossaries with the proper spelling - place names, personal names, text titles, or terms