Typesafety versus Worse is Better - was Re: Fwd: is there any word processing software for the pdp11?

[Johnny Billquist]

On 2014-12-04 04:01, Toby Thain wrote:
> On 03/12/14 9:45 PM, Johnny Billquist wrote:
>> This argument is so old, and so tiring.
>>
>> Can't we just bury it?
>>
>> Yes, some bugs will be caught by stronger types and more type checking.
>> The same stronger type system will cause people to swear and be upset
>> and try to work around it from time to time.
>
> I don't see this happen. It certainly doesn't happen to me.

Any comments like that just make me take the person less seriously. 
Either you are forgetting it, ignoring it, or you are writing really 
trivial programs.

> However, dynamic langs waste hours and days of my time debugging.

That is a potential downside, yes. How much varies per person. Just as 
the "have to work around the type system" time varies per person.

>> Some people are better at not making silly errors than others, but all
>> people sometimes do create errors.
>>
>> People who are more prone to create errors will obviously gain more from
>> type checking.
>
> We *all* make errors.

I think I said exactly that. Just look at what you quoted.

>  > However, I've found that this same group of people are
>> also very creative in writing logical bugs in programs, which no type
>> system ever will catch.
>>
>> Essentially - stupid programmers will create bugs in any language.
>>
>
> *All* programmers. "A sufficiently smart programmer" does not exist.

You are starting to sound like a bigot. You are not reading what I 
wrote, but instead are just trying to preach.

>> Unfortunately, as education moves more and more into just teaching stuff
>> like Java and Scheme, people understand computers in a more and more
>> abstract way, and start doing more and more weird broken things that
>> people 40 years ago would never dream of doing since they would be
>> thinking of how this translates into machine code much more.
>
> Abstraction is your friend. It is there to help you make fewer errors.
> Most bugs can be traced to working at the wrong level of abstraction.
> (And also a lot of wasted effort.)

Depends, and varies per person.

>> So in the end, with people getting less proper education, a stricter
>> language is our only hope. But for people who are good at programming,
>> the obstacles of stricter languages becomes annoying and reduces
>> productivity, for much less gain.
>
> Education isn't the answer. Better tools have always been the answer.
> This business is culturally unable to admit that some tools are just bad.

I believe I just ruled out education above, and noted that better tools 
are the only hope. Why are you repeating what I write, while trying to 
argue against me?

We only differ in the reason why things looks the way they do today. And 
reasons do not actually change what is needed.

	Johnny

-- 
Johnny Billquist                  || "I'm on a bus
                                   ||  on a psychedelic trip
email: bqt at softjar.se             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol