[Terrapreta] Where to get the right charcoal?

[ch braun]

Dear Dave,

Thanks a lot for your (very constructive!) comments.

>I think that distinct entities (feedstock, carbonization method,
experiment, maybe soil) should be defined separately and >with a distinctive
name.
So basically, my idea, as reflected by the xml structure, was indeed to
split a whole <BiocharExperiment> into 4 separate and independent parts:
1. <User>:  all useful infos about the person who is doing this experiment
2. <CharcoalProperties>: all what characterize the charcoal used to amend
the soil (includes "feedstock" and "carbonization method", and what you
mention as "all the relevant static details of the feedstock")
3. <ExperimentalProtocol>: all other parameters which, besides the charcoal
used, describes the experiment (includes "experiment" and "soil")
4. <FinalAnalysis>: the data which should be measured to serve as results

You are right, each different instance of such a category should then have a
unique name so that we have eventually a collection of different users
(U1...Um), charcoals (C1...Cn), experimental protocols (E1...Ep) and results
(R1...Rq).
Then, each new charcoal experiment would be described by a tuple
(Ui,Cj,Ek,Rl) and it would be easy to compare them.

>First, what does a FeedStock entity represent? Is it a generic description
of a possible feedstock (oak, from a forest in >california) or does it
represent a specific feedstock (Dave's oak feedstock from omaha nebraska). I
think you're going for >the latter.
Exactly, for instance "Ci" would describe in a generic manner one "charcoal
class", i.e. one possibility, one "recipe" to make charcoal (oak, from a
forest in california)  while the actual instanciations of this class (Dave's
oak feedstock from omaha nebraska) would be described by (Ui,Ci) where Ui is
Dave...
To be completely precise, I guess unicity of a charcoal experiment would
actually, in my design, require an additional field in case  Ui makes
several times the same experiment Ek with charcoal Cj and records the set of
data Rl:
(Ui,Cj,Ek,Rl,1) (e.g. done in June 2007)
(Ui,Cj,Ek,Rl,2) (in September 2007)
...

> Similarly, for Carbonization the static details of the carbonization
method should be described. Something like (..):
I actually totally agree with you, that was also my intention, so compared
with my proposal your changes for the parameters are:
* you added:
- title
- target temp
- description

* you removed:
- mode (whether it is bought or selfmade)
- minimal temp
- peak temp
- heating rate
- duration

For my point of view these were all "static" parameters? By target temp you
mean average temp ? otherwise is that not the same as peak temp?
"Description" is definitely good, should be added at several places in the
specs actually to allow giving further details.

Now the big question is the "title" i.e., to which granularity should the
"classes" be defined? You also addressed this question:
>When analyzing the data it would be best to only have one char experiment
defining that batch of char, with multiple >cultivation experiments linked
to it. Thus the need for distinct names/identifiers for char experiments.

Apparently, you would rather go 1 level "further" than I do ( i.e. you would
split a "Ci" in feedstock Fij, carbonization Cik...).
I thought it would be for the moment better to stick to the 4-parts design ?
What do you think? And should the "result data" be a subpart of "experiment"
or left alone (as I suggested, part 4) ?

Well I stop it for now before going to further details and wait for your
comments... I have tried to make the overall design I have in mind clearer
and I actually have the feeling that for many aspects we have exactly the
same intention, but it's a bit tricky to express it clearly!

Sincerely yours,
Christelle


On 9/25/07, code suidae <codesuidae at gmail.com> wrote:
>
> I'm probably getting ahead of myself. I'm thinking a bit in terms of
> system architecture and not just moving data. I think that distinct entities
> (feedstock, carbonization method, experiment, maybe soil) should be defined
> separately and with a distinctive name.
>
> First, what does a FeedStock entity represent? Is it a generic description
> of a possible feedstock (oak, from a forest in california) or does it
> represent a specific feedstock (Dave's oak feedstock from omaha nebraska). I
> think you're going for the latter.
>
> As far as the Feedstock part, I was thinking something like this:
>
> .<Feedstock>
> .   <Title>Dave's Oak</Title>
> .   <Material>
> .      <Name>Oak</Name>
> .      <Source>Forest</Source>
> .   </Material>
> .   <Origin>
> .      <Continent/>
> .      <Country/>
> .      <Region/>
> .      <Location/>
> .      <Latitude/>
> .      <Longitude/>
> .   </Origin>
> .   <Composition>
> .      <AshContent unit="% dry mass" />
> .      <(etc...) />
> .   </Composition>
> .   <Details>Oak wood chips, about 1x1x0.25 inches...</Details>
> .</Feedstock>
>
> The idea is to describe all the relevant static details of the feedstock
> completely without reference to the specific experiment. Dynamic details
> (moisture content at the time of use, etc) would go into the experiment.
>
> Similarly, for Carbonization the static details of the carbonization
> method should be described. Something like:
>
> .<Carbonization>
> .   <Title>Dave's Barrel Technique #1</Title>
> .   <Method>Top-lit Updraft</Method>
> .   <TargetTemp unit="C">550</TargetTemp>
> .   <TargetPressure />
> .   <Description>Put stuff in a 55 gallon barrel, light it and try to keep
> it from all turning to ash.</Description>
> .</Carbonization>
>
> A char experiment would define a particular kind of charcoal. A
> cultivation experiment would reference a char experiment to define the
> charcoal used, which may or may not be unique to that cultivation
> experiment. Each char experiment would include notes about that event using
> that method, for example, maybe it was windy that day, or for some reason
> the temp never reached the target. Those facts would be specific to the
> carbonization event associated with the experiment, not the technique itself
> (though it may result in a new technique).
>
> Information about the nature of the feedstock and carbonization method
> used and the feedstock quantity shouldn't be part of CharcoalProperties,
> that should describe the physical properties of the char, not the details of
> what it was made from. Details about what it was made from would be defined
> in a Feedstock element that might be used by multiple experiments. Overall
> I'm thinking something like this:
>
> .Charcoal Experiment
> .   Name
> .   Producer Info (user/batch/etc)
> .   Feedstock Used
> .   Carbonization Method Used
> .   Carbonization Event Info and notes
> .   Charcoal Composition and Properties
> .
> .Cultivation Experiment
> .   Scope (plots,pots, etc)
> .   Soil condition
> .      Pre-amendment
> .      Post-amendment
> .   Amendments Info
> .      Amendment type (char, compost, fertilizer, etc)
> .      Application Details
> .   TestPlant
> .   Procedures/Log (rain,watering, pest control, fertilization, etc)
> .   FinalAnalysis
> .      Soil
> .      TargetPlant
> .      ControlPlant
> .      etc
>
> The idea is that we are describing two separate experiments, charcoal
> production and cultivation with soil amendments. For portability (data
> sharing) purposes it makes sense to include the details of the relevant
> charcoal production experiment with the cultivation experiment data, but it
> should be possible to represent a charcoal experiment independent from a
> cultivation experiment. It's possible that multiple people will be using
> char produced by a common producer, possibly even from the same batch. When
> analyzing the data it would be best to only have one char experiment
> defining that batch of char, with multiple cultivation experiments linked to
> it. Thus the need for distinct names/identifiers for char experiments.
>
> Again, this sort of detail is probably way beyond the scope of the file
> format, but it's necessary to know how things are supposed to fit together
> so that we don't end up with a mess of data that is difficult to sort
> through.
>
> Just some ideas.
> Dave K
>
> On 9/25/07, ch braun < brauncch at gmail.com> wrote:
> >
> > Hello Dave,
> >
> > Actually I had also thought of a solution such as this exemple for the
> > part you mention:
> > <BiocharExperiment>
> > <CharcoalProperties>
> > <Feedstock>
> > <Type>
> > <Category>Forest</Category>
> > <Species>Oak</Species>
> > <Details>50 years old</Details>
> > </Type>
> > <Origin>
> > (...)
> >
> > I think that should be clearer....what's your opinion ?
> >
> > Sincerely yours,
> > Christelle
> >
> >
> > On 9/25/07, code suidae < codesuidae at gmail.com> wrote:
> > >
> > > Bah! Ignore this, I clicked 'Send' instead of 'Save', my intention is
> > > to complete this message later today.
> > >
> > > DK
> > >
> > > On 9/25/07, code suidae <codesuidae at gmail.com> wrote:
> > > >
> > > > I'm a little confused as to the intention and organization of the
> > > > /BiocharExperiment/CharcoalProperties/FeedStock/Type/* elements. It seems to
> > > > me it should look more like this:
> > > >
> > > >
> > > >
> > > > On 9/24/07, ch braun <brauncch at gmail.com> wrote:
> > > > >
> > > > > Yes, this is the current version (both files are exactly the same,
> > > > > it is an xml file, but I just added the .txt version  in case of browser
> > > > > problems...).
> > > > > I tried to merge the comments which were already made here when
> > > > > the specs were discussed for a while, but there are still open questions (
> > > > > e.g. parameters needed in "CharcoalPhysicalProperties").
> > > > >
> > > >
> > > > --
> > > > "Our ignorance is not so vast as our failure to use what we know." -
> > > > M. King Hubbert
> > > >
> > >
> > >
> > >
> > > --
> > > "Our ignorance is not so vast as our failure to use what we know." -
> > > M. King Hubbert
> > >
> > > _______________________________________________
> > > Terrapreta mailing list
> > > Terrapreta at bioenergylists.org
> > > http://tech.groups.yahoo.com/group/biochar/
> > >
> > > http://terrapreta.bioenergylists.org
> > > http://info.bioenergylists.org
> > >
> >
> >
>
>
> --
> "Our ignorance is not so vast as our failure to use what we know." - M.
> King Hubbert
>
> _______________________________________________
> Terrapreta mailing list
> Terrapreta at bioenergylists.org
> http://tech.groups.yahoo.com/group/biochar/
> http://terrapreta.bioenergylists.org
> http://info.bioenergylists.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/terrapreta_bioenergylists.org/attachments/20070926/cc604957/attachment.html