From: "James Britt (rubydev)" <james@...>
Date: 2002-01-02T10:50:43+09:00
Subject: [ruby-talk:30010] Re: Ruby/Python: Software Engineering

>
> All:
>
> Thank you for your messages. If I may summarize the responses:
>
>    The style/formatting of one's code plays no/little part in its quality.
> This follows from style/guidelines/formatting having no/little part of
> code's readability/understandability.
>      Msgs: 29889, 29896, 29913, 29914, 29971,

The syllogism is confusing.  Style and formatting clearly affect readability.
Readability has nothing to do with quality code, unless you value style over substance.

>
>    There are/can be no useful coding style/guidelines because everyone is
> unique.

Now I'm baffled.  No one said that individual uniqueness --> are/can be no useful coding
style/guidelines.

(It is unlikely, though, that there is the One True Style.)


> Therefore, there is no/little way to improve productivity through
> such things.

A flawed syllogism becasue the premise is false.

>The collorary is that languages (keeping to the current topic)
> ought to allow for as close a mapping between the developer's thinking and
> the code as possible and that this increases productivity.
>      Msgs: 29914, 29924, 29936, 29939, 29940, 29943, 29971,

>
>    It is not possible to completely enforce "good" style so it is not/less
> important to work towards this.
>      Msgs: 29939, 29943,
>
>    Software Engineering does not exist. <grin!>
>      Msgs: 29943,

Now, I know *I* said something along those lines, but I hardly think it reflects
a consensus.  There exists something that people call "software engineering"; some
believe the term "engineering" is a misnomer. That's not to say software development
does not contain features found in engineering disciplines.

>
>    Methodologies and Tools exist independently; completely. Therefore, you
> can increase overall productivity by increasing it in either domain because
> they will not (negatively) affect each other. Also, Tools can not be
> designed with Methodologies in mind: One tool fits into a methodology as
> well as the next.
>      Msgs: 29935, 29939, 29940, 29967,
>
>    Ruby does actually have some constructs whose design goal is more
> towards readability/understandability than functionality (eg. CONSTANTS,
> "simple syntax", variable scope shortcuts,etc.). [It does interest me that
> this was only brought up once.]
>      Msgs: 29977,
>
> I seem to be the odd man out here; I feel the opposite is the case in each
> of these (except the last).

I'm tempted to say that you are going out of your way to skew each comment
to appear inordinately at odds with your own dogma, but that would be
uncharitable of me.

Perhaps there is a language (i.e., English) disconnect.


James