[Dbix-class] 4/5 Why Matt's proposal is a farce

[Peter Rabbitson]

On 10/07/2016 08:40 PM, David Golden wrote:
>>
>> [...] I have grave reservations about the specific (now) 4-member team
>> outlined by Matt. The reservations are entirely technical and procedural
>> in nature [...] I will elaborate on these reservations if necessary
>
> As above, I think hearing your reservations and the rationale behind
> them would be valuable input.  It's clear from the comments from the
> community so far that stability (however defined) is important to many.
> Your thoughts on what would and wouldn't work will help shape the
> discussion of future governance.

Words are cheap. Let's look at the record instead: In the past several 
years mst has been directly responsible ( single-handedly or in part ) 
for the following:

- Shoving FATAL-ized warnings down CPAN's users throats. After years of 
incessant pushback finally semi-relented[1], but still continues to 
insert it into his CPAN projects to this day.

- Advocated for the insertion of a computed goto[2] into the core of 
Dancer2's routing dispatch[3], making it inherently unsafe, with 
currently little hope for a fix.

- Advocated for proliferation of Module::Build::Tiny, under the banner 
of "we need to keep trying new things", leading to massive levels of 
inconvenience/wasted time for both packagers and end-users.

- Was pivotal in ramming through the questionable ( to say the least ) 
changes in the Test::More namespace, under the banner of "we are going 
to try and find out"[4], knowing full well there is no rolling back from 
this action.

- Is currently salivating to yet again attempt to rewrite 
SQL::Abstract[5], despite clear signs that this is a bad plan[6] (mst 
himself seemed to agree with [6] around 2016-05-17)


In order to "balance out" his "enthusiast"[7] tendencies (if this is 
even possible given mst's personality) Matt proposes:


- ilmari: Decidedly another "ehthusiast". Great at proposing and 
implementing low-hanging-fruit fixups. Yet loses interest whenever the 
problem space required a more in-depth solution.

- castaway: Openly decries DBIC[8] for being unlike "... other
bits of CPAN, apart from maybe the ones in core, that attempt to be as
rigorous in their perfection"

- frew: True, the only person in the entire thread so far to echo my 
understanding of "stability"[9]. Has a very mature approach to software 
engineering, and while having "enthusiast" leanings as well is able to 
recognize when it is time to put his tools down. On the down-side: has 
(understandably) no patience for pesky squabbles, and has expressed 
unambiguously his involvement won't be "what it used to be"[10]

So in a sense we have the illusion of a committee, and the result will 
yet again be experiments at the expense of the user base, under a 
supposedly-well-meaning pretext similar to [4].


Many participants of these threads were very quick to declare "the BDFL 
model does not work". This despite the fact that virtually all coherent 
software in the CPAN ecosystem (including /usr/bin/perl itself) are a 
product of this very model.


Matt promises a better and more careful way going forward. At the same 
time none of his more recent high-profile actions have even the remote 
indication of some sort of maturity beyond his 2005-era mindset. For 
crying out loud: we are talking about the person who considers this Rube 
Goldberg machine[11] fit for every day use!!!


This is why I remain utterly skeptical. As they say (in Texas I think):
Fool me 5 times in a row...


[1] http://shadow.cat/blog/matt-s-trout/moo-2-strictures-2/
[2] 
https://metacpan.org/source/MAUKE/Return-MultiLevel-0.04/lib/Return/MultiLevel.pm#L95
[3] https://github.com/PerlDancer/Dancer2/issues/1125#issuecomment-179326519
[4] 
http://blogs.perl.org/users/chad_exodist_granum/2016/04/test2testbuilder-update-from-the-qah.html#comment-1702921
[5] https://irclog.perlgeek.de/perl6/2016-09-23#i_13264359
[6] 
https://github.com/dbsrgits/dbix-class/commit/07fadea8d#diff-b1e08875e88f9cfd4c48251700469d84R90
[7] https://youtu.be/A23Xoa5CE7g?t=2363
[8] http://www.nntp.perl.org/group/perl.modules/2016/10/msg96192.html
[9] http://lists.scsys.co.uk/pipermail/dbix-class/2016-October/012277.html
[10] http://lists.scsys.co.uk/pipermail/dbix-class/2016-October/012254.html
[11] https://youtu.be/j3r-lKlrrRg?t=1035