View source for Proposal for reform

We need more gismu since with our current vocabulary we cannot translate a lot of technical vocabulary from other languages.

Cultural neutrality must be maintained by avoiding borrowing terms that do not belong to a particular culture. For example, fu’ivla like cmacrntegra are not supposed to exist and should be replaced by a lujvo; details of a framework that can make this possible is explained below. fu’ivla should only be used for culture-specific concepts (such as spaghetti and kimono), uncommon species (which should be based on the scientific name, for a widespread species, or the name in the language most closely associated with it, for a species that is largely confined to one culture), and a few things that defy all simple description. Moreover, the cultural gismu should be removed because their existence seems to suggest that certain cultures are more important than others. In 500 years, the dominant cultures of the world may be very different than the most important cultures of today, meaning the cultures referred to by cultural gismu are not static concepts. The best way to avoid all of these controversies is to do away with cultural gismu and use fu’ivla to refer to cultures. Similarly, the alphabet selectors lo’a, jo’o, je’o, ru’o, and ge’o have no place in Lojban.

We also should not set the expectation that Lojban words do not change meaning. Inevitably, as time goes on, some things will change. For example, horses may evolve into a very diverse group of species with little resemblance to each other, none of which can breed with today’s horses, so the meaning and use of xirma will definitely change. Also, one day a yellow citrus fruit other than the lemon may become more significant, and in that case people would reconsider what they mean by pelnimre. In short, natural languages do change over time, and if Lojban insists on not changing its vocabulary, then it will become less and less usable.

Chapter 12 of the CLL is extremely awkward in practice, resulting in lujvo that are so complex that it is, to an actually pretty good approximation, impossible to determine the place structure of a lujvo that one encounters for the first time. Instead, we should think about having a systematic way of assigning place structures to brivla based on concept alone. This also means that even the official gismu should be reconsidered so that everything is more consistent; for example, the place structures of gismu like latna, remna, and lanme, and the words zbabu, canre, kunra, and rokci should be redefined so that their place structures are consistent with each other. Also, remove unnecessary sumti places. New vocabulary, which a speaker of a natural language undoubtedly continues to encounter regularly, becomes impossible to learn if the place structures are complicated.

It is definitely possible to define more gismu according to the morphological rules (moreover, I do not think the rule prohibiting similar gismu that differ only in a consonant that are similar makes sense; if all Lojban phonemes are supposed to be very distinct (which seems to be the assumption since many cmavo sound similar by those standards), then presumably they can all be distinguished. If we remove this rule, we can define more gismu, although I am not yet sure if that is necessary). In particular, we should have gismu corresponding to every common prefix, suffix, and word root that are commonly used in scientific fields to logically create words for useful concepts (e.g., pre-, post-, ultra-, super-, hyper-, hypo-, infra-, epi-, -genesis, -escence, etc.) Many languages that follow linguistic purism to a significant degree, such as Chinese and Icelandic, have translations for these word parts so that technical vocabulary can easily be translated. Lojbanists can look up to such languages as inspiration for creating new words. Moreover, they should not try to make their lujvo definitions rely too closely on gismu places like Chapter 12 suggests; instead, they should think about including only the necessary gismu places so that the terms can be learned easily.

Physical quantities (mass, charge, impedance, etc.) and their units. Lojban should have a systematic way of handling them.

The mekso system is way too awkward (first of all, the word mekso itself is egregiously based on English. Change the spelling.). Look up to LaTeX to see the vast number of ways mathematical symbols can be arranged. Often, in actual scientific work, the notation can actually be ambiguous to save space and be easier to process (for example, superscripts can be used for purposes other than exponentiation). Lojban needs a new system that can reflect the kinds of things in mathematical expressions that LaTeX handles really well.
There are still systems that are very difficult to come up with from scratch in Lojban, such as chemical nomenclature. It is certainly possible to have a gismu for every element up to ununoctium. Furthermore, we should come up with a system of naming compounds that works by mapping bijectively between possible IUPAC names and Lojban names.

Lastly, I have an optional proposal. It is more awkward to borrow terms from a language with a larger phonemic inventory to a language with a smaller phonemic inventory than vice versa. Maybe Lojban should have more phonemes in order to more easily accommodate a wider range of languages, and also to ensure that anyone who hears a Lojbanized scientific name the first time can deduce the original spelling exactly (although the latter goal can also be achieved by carefully setting rules on Lojbanizing scientific names).

Here is my proposal for a course of action to the BPFK:
* Clean up the original gismu list
* Expand the gismu list to include gismu that are useful for the purposes mentioned above
* Completely rewrite Chapter 12, along with other changes that I have suggested or that Lojbanists have agreed on.
* Devise systems for handling symbolic systems used in scientific publications in other languages.