It is somewhat perplexing that the lexicon is supposed to be the biggest problem that Lojban currently has. Word creation is the one part of the language that needs no official blessing, no proposals, no anything. Anyone can literally just make up lujvo and use them. That's how it's supposed to happen, too. So why is this not done to get rid of this last thing that's holding us back from using Lojban for something real?
In one corner, there is a group of Lojbanists who despise lujvo that are more than the sum of their parts. I don't think this is a position that can persist. The gismu list does not cover every possible basic concept from which to construct every other possible compound concept. I even doubt that such a gismu list would be possible, even given a quasi-infinite gismu space. There are just too many ideas out there. So to me it's simply inevitable that there will be loads of lujvo that mean things that are not just obvious combinations of gismu. The rafsi will give a hint as to what kind of meaning the lujvo roughly has, and they can serve as mnemonic crutches for learners of Lojban as a second language.
Now, there are people that agree with me. In fact, the founders of Lojban had this very idea about lujvo. So supposing that making these kinds of lujvo was generally accepted, what, then, is the reason that still only so few words exist?
The problem I see is perfectionism; people get into long arguments each time a new word gets made. Sometimes the discussion ends without any word being made and without anything being gained at all. Some would rather have no word for "metaphor", "motherboard" or "morpheme" than one that isn't perfect to their standards. I do think that it makes sense to be careful about word creation and aspire after good words. Chances are that once in use, a word will keep being used, and so it better be a good one. But if taken too far, this attitude ends up producing zero new words, while a different approach might produce 10 "good" and 10 "bad" ones. I also think that the possibility of words that turn out to be dissatisfying being replaced by others is a very real one and should be realized. I'd rather have a bunch of sloppy words and be able to express myself than have no words and remain silent. One way gets the language moving, the other doesn't. I contend that movement in any direction is more useful than no movement at all, for the simple reason that movement entails putting things to the test. Only by putting words — or anything, really — to the test can we properly evaluate them.
The jvojva are great, they work well to the extent that they can, but they are not enough. I don't think it's necessary to use zi'evla or fu'ivla any time a jvajvo can't seem to be made. Rather, this just shows that the jvovja aren't everything.
Returning to the main point about lack of vocabulary, if there is lack and if the language provides an easy way to address this lack, why not just take advantage of that? It could be so easy to have words en masse. But we are scared of trying and then failing, so we'd rather not try at all. It seems a little bit unreasonable to me. Just a little bit.
Finally, I would like to ask people that have made it their hobby to bash other people's word creations one thing: Please, if you have to destroy ideas, at least suggest alternatives or your intervention proves only destructive. Meet every proposal you dislike with a counter-proposal. That way, at least the discussion is kept alive by new ideas.