[article] Language Design Deal Breakers

[Timon Gehr]

On 05/26/2013 03:19 AM, bearophile wrote:
> Jonathan M Davis:
>
>> and I think that null can be extremely useful.
>
> ""null"" values are extremely useful, no one disagrees on this.
> Good functional languages use even more ""null"" values compared to most
> (or all) D programs!
>
> ------------------------
>
> Walter:
>
>> It was Hoare's engaging presentation on it that turned it into a cause
>> celebre.
>>
>> Null pointers aren't even remotely the source of most programming
>> bugs. If they were, then languages that disallow them would be
>> super-productive in comparison. But they aren't. They're just an
>> incremental step, and elevating it into a "deal breaker" is frankly
>> ridiculous.
>
> Even if it's not a big problem, in the end this problem is now "solved",
> because all new languages (Scala,

scala> val s : String = null;
s: String = null
scala> s(0)
java.lang.NullPointerException

> Rust, all the Java-like languages
> appearing on the JavaVM), plus most or all functional languages, don't
> have pointers nullable on default.
>
> Bye,
> bearophile