Andy Lester said:
On Oct 28, 2008, at 1:09 PM, Salve J Nilsen wrote:
Feel free to suggest a better title. (I won't, because I think there's
a motivational value in keeping it as it is.
[snip]
I have to ask what "motivational value" you see in having a Hall Of
Shame under any name. Please describe what scenario you see such that
this Hall of Shame will have any effect whatsoever on the Kwalitee of
work generated.
Well, I can start with stating my assumptions. I assume that most people
have some desire to improve themselves and their work, and I assume that
they are open to the notion that some of this can happen through critique.
I also assume that we're talking about topical critique, and not personal
critique (I know some people have problems keeping them apart, but I'm
willing to assume they at least can _imagine_ how this might work.) The
topics we're specifically talking about are Kwalitee-metrics. (I
purposefully ignore evaluating the metrics themselves. That's a discussion
for another time.)
Furthermore, I assume that authors who publish their code have some form
of pride of their product - at least enough so they feel comfortable
placing their name and reputation next to it.
I think these assumptions are pretty safe, but I do know there are
exceptions to them and that they _are_ after all _assumptions_.
So, how can a "hall of shame" motivate me, given these assumptions? Should
I find a module of mine in this list, I'd start with searching for what
this _means_. Are the points raised in the list (the Kwalitee metrics)
reasonable to me? Can I imagine the relevancy of their critique? If I find
that the critique is valid, then I feel have to consider how to fix the
issues raised by the critique (This is the point where I decide to do
something about it or not.)
If I end up fixing the bugs, then I have been "motivated." And since the
fix presumably has some value, I think it's not unwarranted to say that
the motivation leading to the fix also has some value.
All this is of course difficult to measure on anything but a personal
scale, but we _can_ assume that _less_ bugs will be fixed if we remove
feedback-mechanisms like the list.
Do you imagine that an author of a low-Kwalitee module is going to
stumble upon the list and see his or her module on it? The chances of
that are miniscule.
This is a side-track of the issue. The issue you raise here is about the
visibility and/or marketing of the list. Right now, the list is well
hidden, but with some well-placed words, this can be fixed.
Do you imagine that an author of a low-Kwalitee module, upon seeing his
or her module on the list, is going to go and modify the distribution?
The chances of that are miniscule * tiny.
Well. Even if the likelyhood of an author improving their own code is
tiny, it's better to at least give them the _opportunity_ to do so.
Shutting down the list entirely doesn't help with this.
On a side note; do you take this position in other arguments too? Would
you ever argue that a cancer drug should be discontinued just because
"only one in a million" uses it? Of course you wouldn't. You'd judge the
drugs on it's merits, not how much it's used. Please, let's talk about the
merits of the list instead, and how to improve it, make it more visible.
Most of all, what problem are you trying to solve? I suggest that
low-Kwalitee modules on the CPAN pose no problem whatsoever.
Maybe they don't today. Maybe we can improve the Kwalitee metrics in such
a way that low-Kwalitee modules become more of an issue than they are
today. If we shut down the list, the we lose this opportunity.
- Salve
--
#!/usr/bin/perl
sub AUTOLOAD{$AUTOLOAD=~/.*::(\d+)/;seek(DATA,$1,0);print# Salve Joshua Nilsen
getc DATA}$"="'};&{'";@_=unpack("C*",unpack("u*",':4@,$'.# <[EMAIL
PROTECTED]>
'2!--"5-(50P%$PL,!0X354UC-PP%/0\`'."\n"));eval "&{'@_'}"; __END__ is near! :)
