[Terrapreta] Charcoal Specification Development

[ch braun]

Dear Dave,

Thanks a lot for your explanations, it's much clearer now!  And I have not
much to add or to question, actually. I like the paralllelism between
charcoal and  cultivation trials. By the way:

>It didn't seem useful to parallel the PyrolysisMethod-PyrolysisEvent
relationship on the cultivation side. Maybe I'm wrong about that?
Would a generic 'method' of cultivation experiment make sense? It
would help to standardize testing, but I'm not sure if it would end up
being used in the same way unless it were pretty general, like 'small
pots' 'large pots', '25 sqft plot', etc.

To my point of view this could be useful to add a SoilAmendmentMethod...
especially if we suggest some "standard protocols" later on, that aim at
allowing different users to test some specific aspects.

Well  for the other open questions you ask, it's hard for me to judge what
is best, because I have now just started making some charcoal trials myself
and I think these questions need to be answered by people who have really
some serious experience in the field and can better determine the best level
of details which should be reached.

>I'm trying not to go overboard on complexity, but I don't want to set
it up in such a way that it doesn't represent what people might use it
for. Generally I end up going overboard anyway :D
Well, it's probably hard to find a better example of the typical trade-off
that exists between expressiveness and complexity than such kind of
specification!
Again, it would probably be quite helpful if other experienced testers could
give their opinion about that!

Sincerely yours,
Christelle


On 10/2/07, code suidae <codesuidae at gmail.com> wrote:
>
> On 10/2/07, ch braun <brauncch at gmail.com> wrote:
> > My main concern is actually about the "Material" part. What are exactly
> the
> > relationships between this class, "Soil" and "Charcoal" ? What should go
> > under "Attribute" and "MaterialKind" ? Could you give some examples so
> that
> > it becomes clearer ?
>
> A MaterialKind item describes a general kind of stuff. For example,
> Oak, Rice Hull, Charcoal, Compost, Fertilizer, Soil (unamended) and
> Soil (amended). A Material is a specific instance of a kind of
> material. For example, Oak from Dave's woodlot, Charcoal from Dave's
> TLUD stove #3, Loess soil from Dave's garden, or Loess soil from
> Dave's cultivation trial #2.
>
> Both Soil and Charcoal are Materials and so have the same attributes
> as Material as well as the additional attributes linking them to
> CultivationTrial and CharcoalTrial. A CharcoalTrial uses a Material as
> a feedstock.
>
> A Material would usually be 'providedBy' the same user running the
> trial, but not necessarily. If I wanted to test a PyrolysisMethod with
> a known feedstock I might get some rice hull from someone who had
> already defined it and done trials with it. Thus I could work with a
> know material so that I'd be able to compare my results to others with
> fewer unknowns.
>
> A CharcoalTrial produces a Charcoal item, which is a Material.
>
> A CultivationTrial is similar to a CharcoalTrial in that it takes
> inputs and produces an outputs. The inputs are all Materials, which
> can be just about anything, soil, charcoal, minerals, etc. When
> starting a trial you would create new items to describe the initial
> soil and all the stuff you plan to put into it (of course you would
> preferentially use existing items if they were accurate). At the end
> of a trial you have a soil, which is a Material. If you wanted to you
> could use this Material as the starting soil for another
> CultiviationTrial, with or with out amendments.
>
> Here are some questions I have.
>
> CharcoalTrial only allows for one kind of feedstock Material right
> now, but I imagine someone is going to want to use blended feedstock.
> Should we allow multiple feedstock materials?
>
> I'm not sure how closely we should track crop yields from a
> CultivationTrial. Is it enough to simply note how much there was, or
> should we include provisions for more detailed nutritional and
> physical analysis? Is it enough to put this information in the
> experiment log where others can find it, or are enough people going to
> report enough comparable details that adding fields of them would be
> worthwhile?
>
> For simplicity I only linked one Plant to the CultivationTrial. This
> does not allow trials where multiple crops grow in the same test plot
> (for example, a plot containing the Three Sisters, corn, pole beans
> and squash). That could be problematic, but I wanted to see what ya'll
> thought. Should I have a Plant entity describing kinds of plants
> 'Corn', 'Beans', 'Squash' and then a Crop entity that can link to one
> or more Plants? Most crops would simply be one Plant and maybe a note
> about the variety, but a few could be multi-plant crops.
>
> If you used the soil left at the end of one CultivationTrial as the
> input to another, a significant amount of time might elapse between
> the two trials. During this time soil characteristics might change.
> Currently only one set of analysis data can be associated with a
> material and it is assumed to be unchanging over time. Should a
> Material have a list of analysis events so that we can note who did
> the analysis, when and where?
>
> Dave K
> --
> "Our ignorance is not so vast as our failure to use what we know." -
> M. King Hubbert
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/terrapreta_bioenergylists.org/attachments/20071003/28c74787/attachment-0001.html