On 2021-04-20 4:50 p.m., Richard Eisenberg wrote:
We are as a loose encampment in an open field with a few night sentries. Safe Haskell is a slightly-crumbling earthen rampart along two sides of the encampment. As such, it's really just an obstacle, and does little (but not nothing) to protect us. I was thinking to clear away the obstacle. But of course the better solution is to reinforce the rampart, build two more sides of it, and create a proper defensive position. This will be hard, and I do not propose to take charge of such an act now. But I recognize that the existing structure naturally forms part of this greater whole.
If your main goal is a defensive position you shouldn't be in an open field to begin with. You should encamp on a high ground, with plenty of nearby lumber to build ramparts, and a secure source of fresh water to withstand a siege. That kind of position limits your maneuverability, admittedly, but it's much easier to protect. Sorry, I couldn't resist your analogy. I'm glad that Safe Haskell continues to be maintained, but if you start with the goal of security Haskell is really not the right place to start from. No general-purpose language (open field) is. You want to design from scratch, starting with a secure core language (high ground). You can use Haskell as an inspiration; Marlowe and probably some other blockchain languages do.