Hacker News Reader

Comment by onefortheroad | Hacker News Reader

I have worked with experienced software developers- and usually nothing new is tryied. You open your previous solution kit, you grab the most viable sollution and slap it on. Ductape. Done.

At best a little research is done at some experienced user forum, where the problem is declared unsolveable.

There is a reason why many unversity undergrads find such "genius" new approaches - its because they usually dont know that the problem is suppossed to be unfixable. Not professionally blinded, was the term, i presume.

Remember this is the same lab, who did send subs intot a world war without properly tested torpedos. If my assumption is right, half of them is traditional explosive cannon engineers, whos life wor would become obsolete on success.

View on HN

The military acquisition process (of which technology development is part of) is nothing like software development (even software acquisition projects barely resemble the way SV approaches software). It is frustrating to be a part of because of the endless bureaucracy, but during tech development they do indeed try new things. That is the entire point of Dahlgren's existence, is to research new stuff.

Remember that this is also the same lab that gave us the Norden bombsight, which many historians agree was a major factor in the allies winning WWII.

The bombsight is impressive however:

‘In practice it was not possible to achieve the expected accuracy in combat conditions, with the average CEP in 1943 of 370 metres (1,200 ft) being similar to Allied and German results. Both the Navy and Air Forces had to give up on the idea of pinpoint attacks during the war.’

https://en.m.wikipedia.org/wiki/Norden_bombsight

The ineffectiveness of the Norden is overstated, since that's the trendy thing to do right now.

Admittedly, the Norden did not live up to its hype. Almost nothing performs as well in real life as it does under perfect conditions- this is commonly seen when going from the "golden testers" in development that graduated from test pilot school to the "regular testers" during operational test. But once appropriate tactics were developed (this takes a while with all new technology) and they found a few bombardiers good enough to lead bombing formations, it was effective. Not at the level of pinpoint accuracy, but the performance improved quite a bit over the 1943 results. They never took the Norden off the B-17s because any time the Norden had bad results any previous technology would have been worse.

To the point of my prior comment, Naval Surface Warfare Center Dahlgren, the place that is working on the USN's railguns, is the same place where they were able to get Carl Norden's design into a production ready state where 90,000 of them could be made with the precision of a master Swiss watchmaker. Dahlgren isn't staffed by a bunch of software devs that call themselves "engineers" because they think it sounds better. They have real engineers working on solutions to their problems, and a history to back it up. While you can't assume they've looked at every possible idea, they have considered many novel ways to fix the main issues of railguns.

Reply on news.ycombinator.com