I'm inclined to think the following hypothesis is true: both _under_-investment and _over_-investment have negative consequences for otherwise promising technol

I'm inclined to think the following hypothesis is true: both _under_-investment and _over_-investment have negative consequences for otherwise promising technologies. To be fair, the above statement may be too generalized to be usefully descriptive. There are at least two additional dimensions that probably should be included in any descriptive theory: 1. What is the nature of the technology in question? What do we know about similar technologies in terms of their evolution? 2. Over what time frame? This needs to mesh with the "nature" question (1, above). --- Here are some example cases or scenarios: ## Unproven directions / little institutional support? How does an idea go from a spark, maybe a paper, to having some institutional support? Perhaps someone in academia can speak to this. Overinvestment, it would seem, might create too many splinter groups, leading to a dispersion of necessary talent and connections. ## Promising opportunity for further research / good institutional support Long-term, sustained investment in pure science works well as a foundation, since it allows universities and students to build knowledge over many years, sustain careers, and foster multi-generational knowledge transfer. Overinvestment seems negative to the extent funding streams come from a fixed pie. ## Mature enough for technology transfer to small companies (often funded by more risk-tolerant investors, such as Angels, sometimes VC's) Funding of consortia that bridge academic and commercial interests. Such as the famous MCC of Austin, TX. Overinvestment would tax the resources of educational institutions, which struggle with the value proposition and tuition and fees increase _much_ more rapidly than inflation over the last ~20 years. ## Proven technology, but unclear market fit / sales to early adopters? This seems to be an area where investment works well. Small companies need to customize the technology and sell it. Overinvestment might lead to too many squirrels chasing too few acorns, though. Markets tend to do well when there are enough competitors to drive prices down and innovate. They serve humanity more broadly when market failure are mitigated and minimized. ## Proven market, though perhaps still niche / unscaled / viability for the mainstream? Similar to the case immediately above. Additionally: overinvestment runs the risk of creating inflated expectations and distorting labor costs. ## Proven business / opportunity to improve efficiency of known problems (e.g. practices -- perhaps 'boring' or 'messy' to some -- that nevertheless can dramatically improve quality and/or reduce expenses A sustained ecosystem of companies that build expertise among their employees, who share best practices indirectly (i.e. adoption of source control, automated testing, continuous integration) helps drive continued improvements in an otherwise 'mature' industry. --- Even if the above questions could be made more specific and testable, it would require _significant_ structural and behavioral changes to move technology investment towards a "happy middle".