Sorry for the late reply, I visit this group only once in a while.

Why translate the Niddesa? - I. B. Horner may have translated the Parivaara, the last book of the Vinayapi.taka, because she had already made a second translation of the other  canonical Vinaya books, and this book would complete the set. She also was in a position to see that her translation would be printed.

Most of the canonical Paali texts have been translated into English by now, the cutting edge of present translation efforts is second, better translations of much read canonical Paali texts, translations into other European languages, and translations of non-canonical Paali texts.

Untranslated are: the two Niddesa-s, and a portion of the Abhidhammapi.taka.
The Niddesa is full of repetitions, because it was handed down orally.
It is a very early commentary, so the technique of explaining the canonical text is in a fairly raw state. It contains many quotatins from the Suttas. But it is certainly worth translating. There seem to be also many lengthy passages inserted by the author(s).

The Cullaniddesa published 1918 by the PTS in Latin script was edited by W. Stede, who also completed the society' Paali-English dictionary (PED).
He separated the text into "Part I" (page 1-73), a skeleton with the canonical verses, each verse followed by the list of words explained in the Niddesa and a number. Under this number the actual explanation is placed in Part II (page 75-287). There the explained words are arranged in alphabetical order from a to h, and numbered from 1-700. This cuts out repetitions.
This section also serves as a word index, because the words of the Vatthugaatha (added later to the canonical text and not treated in the Niddesa) are included in Part II in their appropriated alphabetical place, but without serial number.

While I understand the temptation to present the text in this format, I think it is not appropriate for a classical text that was composed orally and handed down on palm leaf for nearly two centuries ina different format, which we find in printed form in the oriental editions. - I do not know whether the PTS has been able to produce a second critical edition in Latin script, which sticks to the format the text has had for nearly 2000 years. Nowadays it is possible to download the text from the internet based on the 6th council edition in Burmese script, transform it into Roman script and print it out from your computer. But this is cumbersome. The book won't fit neatly on your shelf with the other canonical texts in Roman script.

Back to the English translation of the Mahaniddesa:
As a tentative approad one could start with a critcal reading of the existing translations of the basic text: there is the new translation of A.t.thakavagga by K. R. Norman.
For the translation itself one could experiment with the approach used by W. Stede in his edition of Cullaniddesa.
One could then divide the Paali text of the Mahaaniddesa in Part I (canon) and Part II (early commentary), and arrange the words commented upon in alphabetical order.
Since it is a translation, the sequence would have to follow the Latin  alphabeth, not the Paali alphabeth.
But it might be helpful to give the words a serial number and add an index of the corresponding Paali words to the translation.