It was proposed in general in 1788 by Claude Boniface Collignon, and proposed to the National Convention in 1793 by Jean-Charles de Borda, one of the multiple prominent French technocratic military engineers of the era. It was only passed in late 1794, and then shelved in mid-1795, with well less than a year of use.

Complaints at the time included the massive cost of replacing clocks - far more costly than simple rulers and weights - especially while a war of national survival was going on. But another was the massively widespread and above all frequent use of timekeeping. Even units of distance and weight don’t come up in everyday life as often as time of day, and everyone was used to the old ways. The French Revolutionary calendar lasted longer, but eventually it was also shelved by Napoleon in 1806, and only what we know as the metric system for other quantities survived.

A few clocks were produced with decimal time, and many included the traditional system of hours too.

Overall, even when everyone knows the complexities and what may be regarded as irregular ‘flaws’, far more involved and arcane systems that are much more complicated to learn can have massive cultural inertia when getting everyone to change system would go against extremely frequent habit, even if legally unenforced (for example, English orthography!). That said, the decimal system itself is a cultural choice, probably an artefact of our ten fingers, and numbers like 60 and 24 are highly divisible (having many factors), which is helpful - so it’s not so clearly that much worse.

There was another recommendation to decimalise time in the 1897 by a commission led by Henri Poincaré (of the famous conjecture and regarded by many as the founder of modern topology, certainly algebraic topology, and cousin of the later technocratic prime minister), though it never became law, but it too was left by the wayside.

On a related note, another longer-lasting but now also essentially defunct quantity that the French Revolution decimalised was angle: the grade, later grad or gon, was 1/100 of a right angle. Scientific calculators still offered it, pretty much unused, until recently, and (as an anecdotal aside) it has a bit of a humorous status among STEM students and professors from that era. This encountered a bigger problem: although there is no very fundamental unit of time (at least none that would also be practical for ordinary use, so the tiny Planck time aside…) and decimalisation makes sense for convenient compatibility with our general number system, angle actually does have a ‘natural’ unit which makes many fundamental formulas particularly neat: the radian, the angle subtended by an arc the same length as the radius, or 1/2π of a full 360 degree revolution. If we want to be more scientific, then, there seems little value to the ‘middle’ role of the grad, and the existence of three systems is itself an inconvenience. Nevertheless, it was used in at least some quarters for far longer than decimal time, used less often than time but arguably a tiny bit more convenient for engineers and architects, etc. than degrees and radians.