OO advocate tells me it's possible to be precise with no formalism. This is a trap. It's the reason we got so lost in the hype of 80's/90's OO. Demand a formalism. Otherwise we're stuck in the reference implementation hell, living the defects, instead of designing them away.
-
-
99.9% of all FP code also lack formal specifications (or to say, is as formal as OOP code).
2 replies 0 retweets 2 likes -
Replying to @seanmcdirmid18 @shajra
I'm assuming
@shajra was referring to formalisms for new language mechanisms, not formal specifications for arbitrary code?1 reply 0 retweets 3 likes -
It isn’t clear, there are of course multiple dimensions of formalisms. In that case, Java has a formal specification of course (if you believe Gilad did stuff at Sun), Mads is still the C# spec lead. It isn’t something that is very relevant to most programmers, however.
3 replies 0 retweets 0 likes -
(Most program defects have little to do with the formal specification of the underlying language)
1 reply 0 retweets 0 likes -
Replying to @seanmcdirmid18 @shajra
if the designers of Java had been forced to write down the canonical forms theorem for their language before it was considered ready to release then maybe someone would have thought about how much complexity "null" adds to everything. instead it was "oh null is easy to implement"
2 replies 1 retweet 4 likes -
Null is incredibly convenient to programmers and you piss them off when you try and take it away. “Hey, just use these options and explicit unwraps for everything” never went down well. Don’t just assume these choices were made for petty rather than pragmatic reasons.
3 replies 0 retweets 1 like
If you believe using Option is more clumsy than using null, you've really missed the point of the data structure, and doing it very wrong. You'll never be able to show the code where using null is easier. Never.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.