Re: [xml-dev] Requirements of the editor of a working group that isdevel

[Shlomi Fish <shlomif@shlomifish.org>]

Hi Roger,

see below for my response.

On Tue, 13 Nov 2012 13:32:30 +0000
"Costello, Roger L." <costello@mitre.org> wrote:

> Hi Folks,
> 
> Suppose that you are part of a group of people developing a new XML markup
> language.
> 
> One person is designated as the editor.
> 
> I recommend that that person (a) be a SuperProgrammer, and (b) is required to
> single-handedly implement a reference implementation of the markup language.
> 
> [Definition:] SuperProgrammers look just like ordinary programmers but are 2,
> 10 or 100 times more productive [1]. 

Well, before you go ahead of yourself, where do you cite this "2, 10 or 100
times more productive" (aside from the c2.com wiki)? According to:

http://use.perl.org/use.perl.org/_Ovid/journal/39801.html

<QUOTE>
One bit I appreciated, from another presentation of his, is the debunking of
the "best programmers are X times more productive than the worst" myth. He
cites "28 times" as a commonly used figure, but I usually hear "10". However,
he lists the studies for this information and they usually find the best
programmers are only 5 times more productive than the worst and this is
consistent with other fields.
</QUOTE>

> 
> I have met a few SuperProgrammers. They are truly amazing.
> 
> I cite two examples of working groups that met this requirement:
> 
> 1. Tim Bray was the editor of the XML specification. Shortly after XML became
> a full recommendation Tim and Lauren Wood went to Australia and got married.
> On their honeymoon Lauren messed up her right knee and had to have it
> operated on. As Lauren was convalescing in the hospital Tim wanted to be by
> her side. But no point just sitting there so he brought his laptop and
> implemented the world's first XML Parser and called it  LARK (any guess on
> how he came up with that name?) To this day LARK is one of the best XML
> Parsers.
> 
> 2. James Clark was the editor of the XSLT 1.0 specification. Each W3C
> technology goes through a number of working drafts before becoming a full
> recommendation. Interestingly, every time a new XSLT working draft was
> announced as complete, James would announce *the next day* an implementation
> of the new working draft. Ditto for when XSLT became a full recommendation.
> The name of his XSLT processor is XT. To this day XT is one of the best XSLT
> 1.0 Processors.
> 
> 3. (I am guessing here) Time required to implement LARK: 2 weeks. Total time
> required to implement XT: 1 month.
> 
> 4. Cost of LARK: $0. Cost of XT: $0.
> 
> Tim Bray and James Clark were able to do what they did because:
> 
> a. Tim knew XML intimately (he wrote the XML specification) and James knew
> XSLT intimately (he wrote the XSLT specification).
> 
> b. Tim and James are SuperProgrammers.
> 
> Adhere to this requirement and you will have a successful development of the
> markup language. Don't adhere to this requirement and you will likely fail or
> it will be extremely expensive or both.
> 

How do you know that not adhering to the requirement of having a
"SuperProgrammer" (whatever that means) will mean you will likely fail? And as
I note here: http://shlomif.livejournal.com/4408.html there is more than one
aspect based on which a good programmer can be judged. For example,
http://en.wikipedia.org/wiki/Daniel_J._Bernstein may write very good code, but
he does not want to play by the rules that most other people like to, and
he prevented qmail from being effectively maintained for ten years (when he
made it public domain, at which point, it became largely irrelevant due to
competing mail servers heavily superseding it), see:

http://www.shlomifish.org/open-source/anti/qmail/

Even if we can consider DJB to be a "SuperProgrammer", I still don't want him
on my team.

> Thoughts?
> 

I think the whole premise of this post is silly. Newly created XML markup
languages succeed to one extent or another, based on several factors. I'm not
sure how much will the editor be a "SuperProgrammer" contribute to it. You
might as well argue that you should only hire "SuperProgrammers" for every task
at hand, but life is more complicated than that.

> /Roger
> 
> [1] http://c2.com/cgi/wiki?SuperProgrammer
> 
> 


Regards,

	Shlomi Fish

-- 
-----------------------------------------------------------------
Shlomi Fish       http://www.shlomifish.org/
List of Portability Libraries - http://shlom.in/port-libs

Larry Wall *does* know all of Perl. However, he pretends to be wrong
or misinformed, so people will underestimate him.

Please reply to list if it's a mailing list post - http://shlom.in/reply .