By default, tags can only consist of the following characters: [a-z 0-9 + # - .]. However, the correct transliteration of many Buddhist terms drawn from other languages requires the use of characters with diacritics. As such, it would be beneficial for this site to have unicode tags enabled. Can we get that?

3 Answers 3


As @bytebuster points out in his answer, if we don't approach this correctly, this could lead to an unmanageable explosion of synonyms, once you account for differing transliteration/transcription systems. So, here is my proposed strategy for handling unicode tags.

  1. Figure out what word(s) we should use for the name of a tag about a given concept. See also: What language should our tags be in?
  2. Determine whether the tag name counts as "English" or not.
    1. English words like "enlightenment" count as English, obviously.
    2. Loanwords like "karma" that have been in English for long enough count as English. (This is a grey area, but I think we will be able to intuit the right approach in most cases.)
    3. Other things, like "pratityasamutpada", count as non-English.
  3. If the tag is English, we're done.
  4. If the tag is non-English:
    1. Identify the transcription that is most commonly used in English-language scholarly works.
    2. If that transcription lacks diacritics, set that as the tag, and we're done.
    3. If that transcription contains diacritics, set that as the tag, and use the most common diacritic-free equivalent as a synonym pointing to it, so people can type it.
  5. If (and only if) we start seeing other transcriptions being used as tags, also add those as synonyms.

This seems laborious, I realize, but really shouldn't require very much work on the part of our moderators once they get used to a process like this.

The major point here is that, as @bytebuster points out, there is no value in anticipating all the possible transcriptions that could possibly be used. We should only add additional transcriptions beyond the one or two prescribed by the above procedure in the event that we are getting lots of questions using a different transcription.


Here is a list of tags that need unicode in order to be spelled correctly:

  • - should be "pratītyasamutpāda" (macrons on ī and ā)
  • - should be "saṃsāra" (underdot on , macron on ā), though this term is well-integrated into English as "samsara" (no diacritics), and could probably do without the diacritics.
  • should be "bön".

(Please feel free to update this post with additional examples.)

  • 3
    One issue to check before renaming might be searchability/usablilty. We would want "saṃsāra" to show up when searching for "samsara".
    – Earthliŋ
    Commented Jun 18, 2014 at 0:33

SE engine already has Unicode in tags. Some sites, like Russian.SE widely use them and even keep Unicode versions as main ones.

Yet another consideration. samsara is a transliteration (to Latin script), whist saṃsāra is a (phonetic) transcription.

People occasionally use various transcribed and transliterated forms of foreign terms, so it seems to be inevitable keeping all versions: samsara, sansara, saṃsāra, sangsara, and samsaara.
Also, nirvāṇa, nibbana, nippana, nibbaana, and nippaana. I have also seen niivarana, but it does not seem to be correct.

Literally, every term would have kind of five synonyms (plus numerous misspellings).

The bad thing about phonetic Unicode symbols is that they can't be typed on a keyboard, unlike Russian tags for Russian.SE site.

My suggestion is to keep only one transcribed form and one transliterated form.

  • I would suggest using anglicized versions like samsara, nirvana; I'd also suggest using English in cases where the Pali/Sanskrit cognate is not widely used/known. Commented Jun 18, 2014 at 13:02
  • This gets even more complex is the matter of different groups using different transliterations, e.g., nibbana vs. nirvana (mentioned in the post), which makes it difficult to standardize when both terms exist in different languages. Tag aliases can help somewhat (this is how we have been handling it on MartialArts.SE).
    – Hrafn
    Commented Jun 18, 2014 at 17:19
  • @DavidH.Clements Is "nibbana" vs. "nirvana" a transliteration issue? I thought that "nibbana" was Pali and "nirvana" was Sanskrit.
    – senshin
    Commented Jun 20, 2014 at 6:49
  • Yup, that's it exactly, I wasn't clear ^_^
    – Hrafn
    Commented Jun 20, 2014 at 22:02
  • @senshin This is called consonant assimilation that occurred during the evolution of Pali. Commented Jun 23, 2014 at 2:48

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .