Post #2744168
2026-05-01 07:20 UTC
@thomasfuchs @zrail so I can imagine all kinds of difficulties with that proposition and the likehood (or otherwise) of follow through.
What I do have direct experience with is estimating concrete construction projects. Think sidewalks, paving, airport runways. But not on bridges or overpasses.
Locally industry standard for sidewalk is a 2 year holdback for warranty work. After two years you correct any deficiencies to collect the last 2%. Municipalities will require bonding to be eligible to bid. Your bonding agent is your most important “supplier”.
For bridges, owned by the province, warranty periods are at least 25 years. The bonding is so expensive and comes with so much overhead it isn’t worth bidding on. Could we have? Sure. Would we have been a solid contender? Sure. But in the meantime we could have secured work pouring kilometres of sidewalk in new subdivisions with significantly less hassle. Just one reason why those projects are so astoundingly expensive.
Replies (1)
-
@mce@mefi.social 2026-05-01 07:22
@thomasfuchs@hachyderm.io @zrail@hachyderm.io For an open source project you’d need an entity that could “bond” the work. Whether that’s financially or by reputation I’m not sure what the model would look like. It might be possible if the problem was sufficiently scoped - ie just a kernel on static hardware that could still be affordable manufactured in 25 years.