Re: Readability (was: Fixing what's broke)

[Dimitre Novatchev <dnovatchev@gmail.com>]

Hi Norman,

> I expect you to spend zero time counting brackets -- that's what computers are for.  Even vi can do bracket matching, and I imagine that most folk are using editors significantly more functional than that.

I am not a computer and still like (and need) to read code.

So in the code you provided:

>>> ;; The XSLT identity transformation
>>> (lx:namespace ((#f "http://www.w3.org/1999/XSL/Transform";))
>>> (stylesheet version: 1.0
>>>   (template match: "node()|@*"
>>>     (copy
>>>       (apply-templates select: "@*|node()")))))

if I position the cursor after the third ending right bracket, I am
completely lost what is in scope/visible at that point.

This proposal gets a firm

  -10

from me, and I am a typical developer, who must read and understand
code in his work every day.

I find that one of the good features of XSLT as compared to languages
using { and } as block delimiters, is that in XSLT one knows at a
glance where an <xsl:template> or an <xsl:function> or an
<xsl:variable> or an <xsl:choose> ends.

The proposal to change this to an unreadable ")))))))))))...))" will
lead to spending more and completely unwarranted, additional time in
doing my work. Such a feature is not productive and has negative
impact on the work of developers.


-- 
Cheers,
Dimitre Novatchev
---------------------------------------
Truly great madness cannot be achieved without significant intelligence.
---------------------------------------
To invent, you need a good imagination and a pile of junk
-------------------------------------
Never fight an inanimate object
-------------------------------------
You've achieved success in your field when you don't know whether what
you're doing is work or play



On Sun, Dec 12, 2010 at 9:54 AM, Norman Gray <norman@astro.gla.ac.uk> wrote:
>
> Dmitre, hello.
>
> On 2010 Dec 12, at 16:21, Dimitre Novatchev wrote:
>
>>> ;; The XSLT identity transformation
>>> (lx:namespace ((#f "http://www.w3.org/1999/XSL/Transform";))
>>> (stylesheet version: 1.0
>>>   (template match: "node()|@*"
>>>     (copy
>>>       (apply-templates select: "@*|node()")))))
>>>
>>> This would be hideous for document-centric markup, but is excellent for those cases -- XSLT is a fine example -- where there's more markup than text.  It becomes valuable simply because you can see more of the content on the screen at once.
>
> [...]
>
>> The only thing that is seen at once is the large number of closing
>> brackets and that it is impossible to say if this is syntactically
>> correct at all (how much time should I spend counting  left and right
>> brakets?
>
> I expect you to spend zero time counting brackets -- that's what computers are for.  Even vi can do bracket matching, and I imagine that most folk are using editors significantly more functional than that.
>
> I'm not suggesting this as a NextXML syntax -- simply that the discussion in this subthread seemed to have drifted towards compact syntaxes, and it seemed time for another outing for Lx.  Also, one of the motivations for the perennial suggestions of compact syntaxes (and I think the only good one) is that they're easier to read.  This is one reason why people enthuse about Relax, and no-one ever gets excited by XSchema.
>
> [XML's wonderful, and all that; but readable?  Urgh -- only with a debugging hat on.]
>
> I agree that </> isn't worth bothering with.
>
> All the best,
>
> Norman
>
>
> --
> Norman Gray  :  http://nxg.me.uk
>
>
>
>