This is a nice piece on the Thesaurus Linguae Latinae (the TLL), which is aiming to be a truly comprehensive research project, mapping the semantics of the entire Latin language from its earliest evidences in the 7th cent BCE until around the 7th century CE, when it began to lose its grip on a populace as anyone's first language (and slowly shifted into Italian, Spanish, and French, and also into a secondary language of the elite). It has been going continuously, more or less, since 1890; there were a few small interruptions during the World Wars, but even then, a few people carried on. It began as a project of the Imperial German government, but is now internationally funded. (There are a number of good guides to decrypting the TLL, here is one I found quickly on-line, that seems ok.)
This is one of the great works of a previous era of scholarship, that may still be of use for scholars now and in the future. Certainly a lot of the work is painful grubbing through texts, that you would imagine a computer could be programmed to do as well.
Consider: the entry for civitas, or "city," begins mid-way down column 1229 and ends near the top of column 1240. For someone like me, working right now on Augustine's de civitate Dei, a thing like this is incredibly valuable, though its comprehensiveness may be a bit too much of a gift, as ancient thinkers would not have had access to all the references gathered here. (Even all the Christian references would have been too many, in all likelihood, for Augustine to have had.)
Ancient history and literature seems a perfect field of research in which to undertake these projects. (And the Oxford Guide to the Historical Reception of Augustine is another good example of this.) After all, the data is largely accessible, and the problem centrally seems to be one of mass and scale; who can master all this information? So a long-term project, that could systematically harvest the fruits of a multitude of human intelligences in a systematic way, would seem beneficial in a way that any assemblage of individual projects would not be.
This raises for me the question of what scholarship is, and what it can be. Much archival work may seem boring nowadays, but in fact most scholarship through history, and even through the nineteenth century, was composed of this kind of work. Why do we think, today, that real "scholarly contributions" consist centrally of works expressing our individual interpretive judgments on texts, thinkers, or issues? What is the relationship between subjective effort and scholarly product? And on what timetable should that be measured?