<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Dear Frank,<br>
<i><br>
"Agree - if they are relevant."<br>
</i>Of course, the questions need to be relevant. Let's discuss then
the relevance of these questions. I know first that the question
raised by Fabio Riva, Francesco Lombardi and their colleagues, and
that they have, fortunately for us, started to answer, is very
relevant.<br>
<br>
Dear Ron,<br>
<br>
What you are doing is using an appeal to authority, also called
"argument from prestige":<br>
<a href="http://philosophy.lander.edu/logic/authority.html">http://philosophy.lander.edu/logic/authority.html</a><br>
<br>
"This person has a great CV, so probably he/she is right". Appeal to
authority is something I don't like, because I think everyone's
opinion matter, as long as they bring a reasonable argumentation.
Everyone's voice counts the same, regardless of the nice CV. Of
course, we will probably trust more the opinion of a doctor in
psychology than the one of a carpenter on psychology questions, and
vice-versa. But this is not the situation here.<br>
<br>
And you might not want to go down this road, because if you start to
have a look at Crispin's CV, or Harold Annegarn's, or Philip
Lloyd's, with his degrees in chemical engineering and nuclear
physics, you might become very interested in the work they did on
protocols. They did a lot of work, in their field of expertise.<br>
<br>
"<span class="">there are a few in this dialog who <u class="">would</u>
completely throw five years of hard work by many out the window.,
with no intention to ever return."<br>
What you talk about is sunk cost:<br>
</span><a href="https://en.wikipedia.org/wiki/Sunk_cost">https://en.wikipedia.org/wiki/Sunk_cost</a><br>
<br>
This is very important, because I got this argument already from WBT
supporters, and I am sure it was one of the main arguments in the
discussions at ETHOS. Sunk cost is a very common bias in
decision-making. It is a shame if 5 years of hard work are lost, but
they are already lost. If it is a dead end, it is a dead end, no
matter how many you bang your head on the wall.<br>
What we try to prevent now is other years of wasted efforts. Stop
the bleeding.<br>
Excuse my French, but I'll use a not very elegant but powerful image
to illustrate that: if you have been piling shit for 5 years, you
might be frustrated to get rid of all that shit, because even if it
is shit, it took a long time to make a pile that high. But even if
you keep piling layers and layers of shit on top, it will never turn
into a pile of gold.<br>
By the way, do you know how much money was spent improving the WBT?<br>
<br>
The sunk cost concept is to start doing the right thing <u>now</u>.
It is never too late to do the right thing.<br>
<br>
Along those lines, I would like to ask you other questions:<br>
<ul>
<li>How many years of efforts were lost, because of the
uncertainty of results obtained from the WBT testing?</li>
<li>How much money and efforts did stove companies, project
leaders wasted, because their stove went from one testing center
to another, with conflicting results?</li>
<li>How hindered were they in their mission, which is to solve
poverty and save lives? Talking about social impact.</li>
</ul>
I let you ponder on that.<br>
<br>
So in fact I'll try to be true, maybe yes, maybe I want to "kill"
the WBT. If it is not fixable, and as I say I doubt it is, and if it
does more harm then good, then yes, it should be "killed".<br>
But now I am just asking to put it aside.<br>
<br>
<p class="">"It seems the WBT is not able to inform us on
performance.<span class="Apple-tab-span" style="white-space:pre"> </span><b
class="">[RWL2: Please explain further."</b></p>
<p class=""><span class="">The 10 studies and Crispin explain it
well: we are not sure the WBT test tells us how the stove
performs.</span></p>
<br>
<i><b class="">"a) affordability (the second item in Dr. Chiang’s
list)? Do you really wish this in an ISO test?"</b></i><br>
Not really no, affordability is relative, it should be left to
project leaders, companies, and the end customers to decide what is
an affordable stove. I don't believe there are standards for
affordability. I am not sure I understood your first remark, but
what I understood is that you said: "Dr. Chiang is committed to
affordable stoves, so her work on the WBT, and the WBT as it is now,
reflect that". She certainly is, but so are business and project
leaders like Vahid, Camilla, Mouhsine or Sujatha, and they certainly
each have spent much more time working on these issues.<br>
<br>
<b class="">"I see <u class="">NO</u> other way to inform on
performance."<br>
</b><span class="">HTP and CSI, and a few other protocols.</span><b
class=""><br>
<br>
</b><b class=""><b class="">"There is an ongoing (even today) WG3
discussion on field testing</b>"<br>
</b><span class="">Field testing with the right protocols, sure,
because it is contextual and provides a much more accurate
picture.<br>
<br>
"</span><span class=""><b class="">Are you fundamentally opposed
to any test placing stoves into 9 [more or less] tiers?</b>"<br>
No. It is practical to have tiers, a bit like energy ratings for
refrigerators or light bulbs.<br>
I am opposed to that if it is scientifically invalid, like it
seems to be.<br>
</span><b class=""></b><br>
<b class="">"And where are you on including charcoal output in the
reporting of performance?"</b><br>
<p class=""><span class=""></span>I have no strong opinion on that.
Tami and Crispin present convincing arguments. I agree with you it
should be reported, for sure, but in a scientifically correct way.</p>
<p class="">"<b class=""><span class="Apple-tab-span" style="white-space: pre;"></span>b)
context? This is the purview of WG3 - field testing."</b></p>
<p class=""><span class="">Sure, but it is not the purview of the
WBT, that's one of the problems with it.</span><b class=""><br>
</b></p>
"<b class="">shipping? I doubt you can find any ISO test procedure
that includes shipping as an issue."<br>
</b><span class="">I am not talking about shipping, I am talking
about developing a stove by using the WBT for months, and one day
having this stove being used for real, by real local cooks, in a
completely different geographical and cultural context. Surprise,
surprise.</span><b class=""> </b><span class="">This is where you
get the first inter</span><span class="">esting</span><b class="">
</b><span class="">results.</span><span class=""> Contextual
testing, even when performed in a different location, limits the
surprise.<br>
</span><br>
<div class=""><b class="">Don’t you agree that all the above are
best left to the companies and individuals selling stoves, not
those involved in lab (or field) testing?<br>
</b><span class="">Ideally, but in reality the companies and
individuals, to whom I will add the large international
development and humanitarian organizations, are very short on
time and staff, and under a lot of pressure. And they are not
scientists (they've been told that well enough). And they don't
have the equipment. So they delegate the testing to
organizations they trust.</span><b class=""> </b><span class="">The
organizations </span><span class="">they</span><b class=""> </b><span
class="">trust must provide them testing results that can give a
fair (it is never perfect) idea of how their stove will perform
for their context. The WBT does not allow that.<br>
</span></div>
<br>
<div class="">
<div class="" style="orphans: 2; widows: 2;"><span class=""
style="color: rgb(97, 97, 97); font-family: 'Gotham SSm A',
'Gotham SSm B', sans-serif; font-size: 15px; background-color:
rgb(255, 255, 255);"><i class=""><span class=""><b class="">"3)
usability, and </b></span></i></span></div>
<div class="" style="orphans: 2; widows: 2;"><span class=""
style="color: rgb(97, 97, 97); font-family: 'Gotham SSm A',
'Gotham SSm B', sans-serif; font-size: 15px; background-color:
rgb(255, 255, 255);"><i class=""><span class=""><b class="">4)
access to a broad range of technology </b></span></i></span></div>
<div class="" style="orphans: 2; widows: 2;"><i class=""><span
class=""><b class=""><span class="" style="color: rgb(97,
97, 97); font-family: 'Gotham SSm A', 'Gotham SSm B',
sans-serif; font-size: 15px; background-color: rgb(255,
255, 255);"><span class="Apple-tab-span" style="white-space: pre;"> </span>and
fuel opportunities.</span><font class="" face="Gotham
SSm A, Gotham SSm B, sans-serif" color="#616161"><span
class="" style="font-size: 15px;">”</span></font></b></span></i></div>
<div class="" style="orphans: 2; widows: 2;"><i class=""><span
class=""><font class="" face="Gotham SSm A, Gotham SSm B,
sans-serif" color="#616161"><span class=""
style="font-size: 15px;"><span class="Apple-tab-span" style="white-space:pre"></span></span></font></span></i></div>
</div>
<div class=""><br class="">
</div>
<div class=""><span class="Apple-tab-span" style="white-space:pre"> </span><b
class="">I contend, mostly from several stays at Aprovecho, that
lab testing using the WBT, can give great insight on these as
well."<br>
</b><span class="">4 maybe, 3 certainly not.</span><b class=""> </b><span
class="">This is probably the worst one.</span><span class="">
As users often say, while laughing, when we ask them to just
boil water: "no one does that!". They think we are these kind of
eccentric westerners.<br>
Contextual testing is the only thing that can give insight on
usability. Usability is one of the most if not the most
important aspect when making a stove for a customer.<br>
</span></div>
<br>
<b class="">RWL4: Clearly the present protocol is limited to
boiling/heating water. Are you urging that every lab adds roti
testing? Don’t you think that stoves used in part for roti-making
sometimes also heat water? <br>
</b><span class="">I don't ask for a protocol for each dish, I ask
for a contextual protocol flexible enough so you get valid results
whatever the dish is.<br>
The HTP and CSI do that.<br>
</span><b class=""></b><br>
Frank very well mentioned this idea of making progress, and I would
also really like to highlight this important point.<br>
The world is turning, the world is progressing. Other fields of
activity evolve, their practices change, what they do one given year
differ from what they did the previous year. This is called
"progress". <br>
It shouldn't be any different for the stove sector.<br>
<br>
It is perfectly understandable that a testing protocol dating from,
what, 1987?, becomes obsolete and stops being used.<br>
<br>
Testing centers will remain testing centers, testers will remain
testers, they will still be needed, they will just evolve.<br>
As far as I know, no one is married to the WBT.<br>
<br>
Note that I don't have interests in the HTP and CSI. These protocols
can be used, or other ones that are better, it doesn't really matter
to me. The only important thing is that we are able to test our
stoves and understand how they perform, so we can improve them, and
serve our customers.<br>
<br>
Best,<br>
<br>
Xavier<br>
</body>
</html>