[lug] First Programming Language

[George S.]

On 1/4/2018 2:02 PM, Steve Litt wrote:
> On Thu, 04 Jan 2018 11:35:25 -0700
> "Michael J. Hammel" <mjhammel at graphics-muse.org> wrote:
>
>> On Thu, 2018-01-04 at 10:27 -0700, Jed S. Baer wrote:
>>> Oh, come now. Laying poor programming knowledge at the feet of
>>> Python,
>>> vs. any other language?
>> I'd actually lay it at the feet of universities who teach Python
>> because it's easy to learn, without backing it with the foundations of
>> programming practice.
> Puh-leeeeze. Python is popular: There's a chance of the student getting
> a job based on knowing Python. I'd imagine most university and Junior
> College instructors will add material about good programming practice,
> but even if they don't, the newbie will learn that as a junior
> programmer on his/her first gig.
>
> With its indent-dependent syntax and lack of matching braces, Python
> gets rid of one of the biggest stumbling blocks to newbies just trying
> to learn how to run a loop and terminate on a certain condition. Its
> lax typing means that they're not constantly and forever going to be
> stopped in their tracks by a type violation in an otherwise sound
> algorithm.

Leaving them to find mysterious and bizarre bugs in their code years 
after it was written. PHP is a great example. The parameter order for 
array_key_exists() is reversed from property_exists(), and if you screw 
it up, the compiler won't catch it.

Weakly typed languages are a QA disaster. They dramatically up the 
requirement for coverage testing. I don't have any numbers, but my 
decades of professional programming experience have taught me that for 
any non-trivial application, the life-cycle cost is much more about 
ongoing maintenance than original time to develop. Weakly typed 
languages just multiply that cost by another factor.



-- 
George S.
*MH Software, Inc.*
Voice: 303 438 9585
http://www.mhsoftware.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lug.boulder.co.us/pipermail/lug/attachments/20180105/982cb2c2/attachment.html>