Towards a formal framework for understanding guilds and food forests

Before I get started, this isn't a rant, although the first paragraph is going to sound like one.

Just about everyone in Permaculture talks about guilds. We all think we know what a guild is, but inexperienced practitioners keep coming back and asking the same questions that may or may not apply on two different pieces of land. I wanted a relatively simple means of defining a guild that would enable comparison between two similar guilds across different patches, perhaps on the same piece of land, or perhaps on two different continents, which would enable me and others to plan planting, picking, inputs, yields and so on. This is the kind of information that could go in a searchable database, or simply be used on the farm for planning purposes.

It's a tool: you don't have to use it, but it's intended to make things easier in some circumstances.

My pilot on this post suggests there's a problem with long words. This is not me attempting to explore the range of my polysyllabic vocabulary, but I'm not sure how to express some concepts without it.

My concern is that “just going and doing it” is the kind of thing most likely to lead to failure. My penchant for the kind of planning that would frustrate Eisenhower might be going too far the other way, but I like to have some idea what I'm doing.

My answer is to use existing concepts in ecology, and a system that I know works, as a framework for building patches in a forest garden, which would typically equate to a perennial permaculture guild. The existing system has its limitations for what I want to use it for, but it provides a useful foundation.

The intention here is to create a framework for discussing what works and what doesn't, as well as comparing yields, planning disturbance and picking requirements, and allowing us to settle ongoing questions about pest load and predator response when comparing native and non-native plants and so on.

I have used the United Kingdom National Vegetation Classification as this framework, mainly because I'm familiar with it and know it works. It is not adequate for this purpose, and I have added concepts as necessary. The main issue turns out to be the fact that a guild does not exist in climax isolation. My answer to this is to define the guild at the horizon habitat – the way the patch is intended to look once the canopy has matured - but describe the successional process the guild goes through to reach that point. This has the advantage, especially during research, of being able to refine this description if (and I would expect very often when) the plan does not meet the realities on the ground.

Any description will be a snapshot at a given time.

The NVC conventionally uses scientific names of plants for precision, and ecologists are expected to be familiar with expert terminology, but this makes the system inaccessible to many non-experts. With anything up to several hundred unique plant species to the hectare in a mature forest garden, the vernacular names of all uncommon species may be unknown even to the native speaker. I therefore propose the use of vernacular names with current scientific names in parentheses (eg. Apple (Malus domestica)). This meets all the demands for accessibility and precision.

I hope that this system will simplify the process of creating guilds and patches in permaculture-type gardens, and enable direct comparison between approaches, thus also simplifying communication between practitioners.

The community is a description of the dominant species at the intended horizon. This is in order to ensure descriptive continuity. A freshly-planted guild may be dominated by groundcovers while the tree and shrub species mature. The groundcover describes the subcommunity, which is likely to change during the course of succession.

In the NVC botanical subcommunities are defined for each woodland type, and I propose doing something similar. Not all subcommunities will suit any given woodland type, because different trees require different soils. In semi-natural habitats in Britain you won't generally find common reed (Phragmites australis) in a Scots pine (Pinus sylvestris) woodland, but you may find it associated with alder (Alnus glutinosa). That said, fruits related to the plums tend to require similar soils, are often grown on closely related rootstocks and will therefore be suited to similar groundcovers and require similar soil amendments. Grape (Vitis vinifera) may do well with apples (Malus domesica), citrus species and cherries (Prunus species) in areas without severe frosts, but be unsuitable at the latitude of Scotland, which would also preclude the growing of oranges (Citrus × sinensis).

The answer to this would seem to be to define subcommunities separately, according to soil needs and suitable climate zone, and mix and match accordingly. The habitat manager can then associate a given canopy community with a range of possible subcommunities, each of which may also be associated with some, but not all canopy communities.

To define both the community and subcommunity I propose using the existing DAFOR scale, already familiar to ecologists:
Dominant: A species covering at least 75% of the available canopy. It's unlikely you will use this much in practice unless you are converting an existing orchard.
Abundant: A species covering 51-75% of the available canopy. Again, it's unlikely this will be used commonly in practice.
Frequent: A species covering 26-50% of the available canopy.
Occasional: A species covering 11-25% of the available canopy.
Rare: A species covering 1-10% of the available canopy.

To define a community, identify the coverage of the main canopy trees at proposed maturity of the trees, until you have identified the majority of the canopy. A mixed patch of apple and cherry would then be defined as a apple-cherry community, perhaps with occasional sea buckthorn (Hippophae rhamnoides). The apples and cherries would be unlikely, in an open savannah-type woodland, to cover more than 50% of the canopy. This community would be called an Apple (Malus domestica)-sweet cherry (Prunus avium)-sea buckthorn (Hippophae rhamnoides) woodland.

The subcommunity will change with succession, but will also be managed. The advantage of this is that it allows both definition and description. To define the subcommunity, again list the main plants in order of coverage of the ground until you reach 50% of the cover. Then describe it, listing other common species that are occasional or rare as defined by the DAFOR scale. Most climbing plants cover little of the canopy and would be mentioned here.

The NVC, as a vegetation classification, does not describe the mycota of the habitat. Semi-natural ecosystems will include a number of fungi from edible ones through to species that may harm the flora. Deliberate inoculation of roots with edible arbuscular mycorrhizal fungi and of logs and woody mulch with decomposers not only provides an edible crop, but may discourage colonisation by these inedible or even harmful species. Accordingly, the description (or definition? please discuss) of the patch should include this feature. A few rare species are found on mature fruit trees (such as orchard tooth fungus (Sarcodontia crocea)), and this may be noted, especially when the manager is converting an existing orchard.

It's important to note that this is intended to be descriptive, not prescriptive. My own experience surveying using the NVC is that there are many habitats on the boundary between two or more classifications, and two surveyors may disagree on how to define subcommunities in particular. Even assessing relative dominance is subjective. The precise proportions of cover will vary over time: an early succession subcommunity will not be the same as that found once the canopy closes, and will vary according to local conditions and the needs of the habitat manager. An apple-cherry community may suit, even require, a denser canopy in a Mediterranean climate with more sunlight and less water than a similar community further north. An open canopy with perhaps 40-50% coverage (technically a savannah habitat) may produce greater yields, but a closed canopy will allow greater habitat complexity and the growth of unusual species requiring deep shade.

This allows the land manager and the ecologist as a scientist to consider factors such as:
* Requirements for soil remediation
* Requirements for pollination and the study of which pollinators may be present
* Function stacking
* The study of diseases and pests and the predators on them, and comparison with similar sites
* Differences in requirements according to climate zone and bioregion or ecozone
* Estimated and actual caloric and protein yields
* The planning of picking and processing
* Succession, including changes as a result of climate disruption
* … and so on.

This is as much a scientific tool as a practical one. It should help us settle questions about pests and pollination in relation to native and non-native species, for example.

It should be possible (even a simpler process) to describe hedgerows, meadows, aquatic habitats and annual polycultures in a similar fashion. This should then lead to the possibility of description of boundary habitats between them, and the advantages of a patchwork of sites within a broader garden, and possibly at a later date between contiguous gardens.



Here is a worked example, using a fairly basic apple-cherry guild. It's somewhat simplified - a patch in a research woodland such as the Martin Crawford's at Dartington may be much more complicated, but the same principles should apply.

Apple (Malus domestica)-sweet cherry (Prunus avium)-sea buckthorn (Hippophae rhamnoides) woodland with bean (Fabacaea)-artichoke (Cynara cardunculus)-nasturtium (Tropaeolum) subcommunity.

The canopy is dominated by occasional or frequent* Apple (Malus domestica) and sweet cherry (Prunus avium) with occasional sea buckthorn (Hippophae rhamnoides). Single specimens of other Prunus species. such as plums, or pears (Pyrus spp) or unusual species may be present.**

Stands of nasturtium (Tropaeolum spp), artichoke (Cynara cardunculus) and a range of possible legumes (typically lupin (Lupinus spp) and bush-type beans (Phaseolus vulgaris)) dominate the shrub layer. Jerusalem artichoke (Helianthus tuberosus)*** may be occasional, even frequent, outside the woody root zone, but will become less common as the canopy closes. Comfrey (Symphytum spp, most commonly Symphytum × uplandicum) may be occasional. A rhubarb (Rheum rhabarbarum) specimen may become occasional in later succession, especially in cooler latitudes. The herb layer contains yarrow (Achillea millefolium), dill (Anethum graveolens), chicory (Cichorium intybus), coriander (Coriandrum sativum), fennel (Foeniculum vulgare), bergamot (Monarda spp) and dandelion (Taraxacum agg), typically in drifts, and any may be rare to occasional. Allium species, such as chives (A.schoenoprasum), leeks (A. ampeloprasum) and garlic (A. sativum) are typically rare.****

A number of climbers may be present, such as climbing (Phaseolus vulgaris) or runner beans (Phaseolus coccineus), and hardy (Actinidia arguta) or common kiwi (other Actinidia spp), or grape (Vitis vinifera) in warmer climates.



Obviously this will change somewhat as more sun-loving species are shaded out by the canopy, but this system allows you to plan this process.

This subcommunity might be artificially encouraged to succeed to one dominated by currants and gooseberries, which might change the definition and would certainly change the description. A different subcommunity might be dominated by blackberries and raspberries, but have similar nitrogen fixers, so this would be a Rubus-Fabacaea subcommunity.

Note how there is stratification of the soil layer, with deep-rooted species such as comfrey and lupin bringing nutrient up from lower soil layers, a number of other food crops and several nitrogen fixers, along with a range of flowers throughout the pollinator seasons.


* When describing a particular patch you would say which
** If they are not single specimens it would probably become a different community, except in a larger patch.
*** A good example of why I advise using scientific names!
**** Note that this doesn't actually mean that there are only one or two of them, but merely that they don't take up a lot of growing space.


Discuss. Does it make sense? Does it work for you? What do we do about defining the mycota? What are the problems? How can we go about fixing them?

This post represents an incredible amount of thought and consideration. The only other place that I have found such a scientific approach is in Mollison's Permaculture, a Designer's Manual. We should all strive for such precision in our work.

Might I suggest that the reader skip ahead first to the last few paragraphs of Neil's post beginning with "Apple (Malus domestica)-sweet cherry (Prunus avium)-sea buckthorn..." to see what Neil is aiming for. Please do this before reading further in this, my own post, so that we are on the same page. This concept is so evolved that it nearly left me behind in the reading of it. I was delighted, however, when I came to the practical application of it and realized, concretely, how useful Neil's methodology might be!

To sum it up, I believe what Neil is working on is a uniform way for us to describe polycultures. For some of us more earthy types this might seem totally OCD but, in truth, if we are going to establish Permaculture as a viable alternative to "modern agriculture" we need a system that the scientific community will take seriously. This could very well be that system.

As a visual person, I would really love to see a collaboration between Neil and a photographer -- to make Neil's ideas more accessible. Having some beautiful photographs of forest garden systems, coupled with Neil's hyper-accurate descriptions, could prove to be a seminal moment in the evolution of Permaculture as a science.

I believe that Permaculture is at a tipping point. It is becoming very popular but in its popularity could be the seeds of its own demise -- if the science of it is not described with precision. Commonly I see a watering down of Mollison's original thinking. In this article I see a further evolution. Thank you, Neil, for letting us into your thought process. I will share it in the places where I know that other bright minds hang out.

The Regenerative Agriculture group is a pretty good place for this. https://www.facebook.com/groups/regenerativeagriculture/

I believe, also, that there are a few photographers working on collections of Food Forest photographs. I have noticed a few posts about this in the Permaculture Entrepreneurs group: https://www.facebook.com/groups/757945157664218/

Cheers,
-Karl

I love it! A great combination of the academic and the practical, just how I like it. I'm looking forward to hearing/helping the next stages of development.

Excellent work, Neil! In the course of developing as a student of permaculture, I have certainly found myself hunting for exactly this sort of information. It would be quite useful to develop a rich aggregation of guild outlines. Your explanation of their use as "descriptive not prescriptive" is, I believe, the key to such a tool. How helpful for us to be able to build on one-another's experiences by having clear descriptions of guilds and their progress to maturity!
I look forward to further development of this system.

The more I think about it, the more I think it needs work, but I always thought that - the key is in the word "towards".

The question of what to do about the mycota is bothering me. As Stamets points out, you can't really define a woodland without reference to its fungi, but there are mycorrhizal fungi which will be key to the community, and there will be other fungi which are incidental. I'm thinking along the lines of having key fungi as part of the definition and incidental ones as part of the description, but I'm not sure. Surveys based on fruiting bodies will often be inconclusive, and this is an obvious problem.

More seriously, and this does need to be fixed, is the question of definitions. I got too close to the thing and failed to look at the bigger picture, and I'm really annoyed with myself over it. My definitions are too fuzzy and, frankly, not good enough, certainly as a scientific tool. Strictly speaking, a polyculture is anything that's not a monoculture. Guilds and polycultures are not synonyms: a guild relates to particular types of relationships in a particular type of polyculture. A forest garden polyculture will consist of many patches, which may or may not be guilds, and this needs to be nailed down, because what I'm describing above will relate to one patch in a broader system with many other patches. You can probaby still describe any given patch (as distinct from a guild) in the same way: the NVC is not designed to tackle guilds in the first place.

I don't think this is a reason to abandon the idea, but this latter question in particular needs to be nailed down before we take it any further. I'm not up to that tonight, but feel encouraged to make constructive input.

Do we take the guild notion out of it, concentrate on patches, and accept that some patches may include guilds?

I'm going to let my intuition work on the problem, but constructive input should be considered welcome.

Neil, this is a great start to some very important work. I like that this methodology can scale; the same descriptive technique can be used on a series of small patches and the larger "macro-patch" they comprise.

However, as you point out, describing the composition of a patch, while very useful, does not necessarily describe a guild, which is concerned with the relationships between plants (and other organisms). Making this work harder is the fact that a guild itself is barely more than a theoretical construct. A person can design and plant a guild with a number of relationships in mind, but the reality is that far more interactions are being created than the planter realizes or has any ability to discern.

I imagine that part of the reason this formal framework hasn't arisen yet is that there are a staggering number of actual interactions in even a small patch, and many of those will not even be discernable to an observer. In the descriptive system you've set forward, communities and subcommunities are defined based on canopy or ground coverage. I'm struggling (as it seems you are as well) with how this can be augmented to include guild relationships, where there is no single metric, such as canopy coverage. Also, different observers are likely to to place differing emphasis on particular guild functions or relationships.

I agree with Karl about the importance of this kind of work. I don't think that taking the guild notion out of it is the answer, but it is clear that integrating the notion of guilds into your framework will be a challenge.

I don't have much to add about the mycota issue, except to point out that there, too, observability may be problematic.

Awesome Neil!! I love where you are headed with this.

One resource to help along the way would be Ecological Site Descriptions. The feds have done a lot of the leg work for us here in the States. We just have to adapt and improve upon that work. ESD's can be queried here: https://esis.sc.egov.usda.gov/Welcome/pgReportLocation.aspx?type=ESD , anyone who would like help querying data, please PM me and I'd be glad to help.

Also, I think standardizing a descriptive methodology would be good. One existing method that comes to mind would be Ocular Reconnaissance Inventory Method. It's primarily a rangeland inventory method. I can't find any documents online, but have it in hard copy. I'll get it scanned into Google Drive and share a link, I think a brain like yours could certainly use it as a launching pad to develop something really awesome for this.

As I've been typing, I think that I will take your idea and apply it this summer. In Wyoming we have a lot of food forests at what were homesteads. I will try an do a couple descriptive analysis of one or two of these to hone in what exactly is making it work!

Thanks again Neil!

Neil Layton wrote:Strictly speaking, a polyculture is anything that's not a monoculture. Guilds and polycultures are not synonyms: a guild relates to particular types of relationships in a particular type of polyculture. A forest garden polyculture will consist of many patches, which may or may not be guilds, and this needs to be nailed down, because what I'm describing above will relate to one patch in a broader system with many other patches. You can probaby still describe any given patch (as distinct from a guild) in the same way: the NVC is not designed to tackle guilds in the first place.

I don't think this is a reason to abandon the idea, but this latter question in particular needs to be nailed down before we take it any further. I'm not up to that tonight, but feel encouraged to make constructive input.

Do we take the guild notion out of it, concentrate on patches, and accept that some patches may include guilds?

I'm going to let my intuition work on the problem, but constructive input should be considered welcome.

If we are speaking strictly, I don't think even a monoculture is a monoculture. Monocultures don't exist in any real way, it's just shorthand for "I'm only "really" planting this and all the other stuff is not even under consideration unless I have to spray it with herbicide" There are exceptions even to that: I know an organic farmer who grows single vegetables in rows but lets nature grow up around them. Is that monoculture or polyculture?

I think that guilds are just a designed polyculture. It's also true as with monocultures, "I'm only planting these things" being the new maxim.

But I think it's good to stick with the idea of guilds and noting that they might blend together into patches or be substantially different from patches (since they don't exactly follow the same rules).

I find that the hardest thing for me when actually working with guilds is having enough species (and enough diversity) to make a presence felt when working from scratch. I'm still very plant-poor and I use spontaneous plants (exept grass) as leverage for the time being.

Hope you get some good responses, although I have the feeling that you're really pushing the envelope here and maybe only people in the realm of Jacke and Toensmeier can provide valuable input. This is all very advanced, at least for me.
William

Here is the link to that PDF: https://goo.gl/jxFcQm

Description of Occular Reconnaissance Method specificall developed for a Wyoming state agency for grazing lease assessment. The fundamentals on the first few pages would be totally applicable to this situation.

Thinking more about this. What if a inventory of polycultures was set up. Strictly an inventory. Observers could inventory agricultural or wild polycultures and submit them to a "Region Administrator". That person would validate and QA/QC the information (just to keep data clean and standardized). The region administrators would be a network of individuals that would compile and organize the data. From the standardized data you could start running statistics and identify the best combinations of plants for any specific ecotype. But what about guilding?

Well, administrators (or observors) would record and cross reference existing databases (of which there are many) for plant use. Medicinal, nitrogen fixing, fruiting, fodder etc. (I'm specifically referring to the really good one in I think Gaia's Garden.

So then you could get together with a slick programmer person that could run logic out of the database and create a simple program that would have the inputs of your location's variables (soil, temperature, latitude etc) and then also have logic that inputs your goals (fodder dominated, wild dominated, fruiting dominated, or something to that effect) and then query against the crowdsourced data. The maturity part of the form below could be used for similar systems to show difference over time and through succession. So your output could be a "plant it this way for this late seral composition, but expect these early and mid seral compositions along the way". Giving you the option to maybe adjust the succesional composition from there and really tweak in your development strategy.

Here is an example of the most basic google forms that could inventory the data: https://goo.gl/mWeQ5Y

What started this thought is that we know a lot about individual plants. We know a lot about inventorying ecological systems. We don't know a lot about ecological SYSTEMS. So put two and two together is where I'm coming from. A lot of work has been done in the inventorying realm on natural systems. A lot of work has been done on conventional agriculture systems. We need to just do the same work on polyculture systems. The base methodology model is there, it just needs to be expanded and executed in the permaculture way.

That's as far as I got with the thought. Discussion?

Awesome Neil! I can't even begin to comment properly on this tonight, and will read it again tomorrow and hope to wrap my brain around it more cohesively and come up with something coherent and useful to add. Thank you.

What do we do about defining the mycota?

Perhaps two statements
(1) observed fruiting mycota associated with which vegetative and litter covers.
(2) Inoculants introduced with which vegetative and litter covers.

Additional descriptive listing
Invasives and weeds present or possible
(1) Probable interfering with desired succession and requiring control measures.
(2) undesirable but probably diminishing naturally with succession.

Thank you for this outline. I will try to apply it to the descriptions as I post the history of my project.

You guys are talking way over my head-which illustrates the need for a data base!
I'm new to the whole polyculture/guild planting (new to permaculture in general) and am soaking up as much info as I can find out there so I can make an informed choice on what resource (book/course etc) would best suit my "stage of development" in a particular area of study. I'm planning on setting up an off grid homestead over the next year and have a lot of material to process on ALL fronts. I mean learning about my region's climate and how it affects building infrastructure, energy resources, crops and systems from scratch.
As I have limited capital for setup and establishment before I have to be self sufficient and off grid there is a sense of urgency in my plans and I(and I expect others too)cannot be proficient in all fields all at once. Sometimes the awareness of the great lack of understanding and shear enormity of information to be absorbed and transmuted into reality is overwhelming.
Any resource that would encapsulate data/learning into a simple equation or list by imputing my unique parameters would prevent me making time wasting blunders. These mistakes I'd probably only discover maybe years down the line?
I think it's amazing that Neil has identified a potential solution that could, maybe with collaboration from others who also have extensive knowledge in this field, be very effective in taking permaculture design principles to the uneducated(in permaculture, me included) as well as gaining respect from the scientific community at large.
Fantastic article

Okay, before we go any further, let me nail down the mistake I made.

* So, the mistake: I treated "guild" and "polyculture" as synonyms, which is wrong, and I'm kicking myself for making such a glaring stupid mistake.

Let's assume we design at the level of the patch. That's what I had in mind with the above system of description.

* One problem is that in a forest garden you are not talking about one patch, but about many interacting patches. With that in mind, is it even possible to make the system work?

* IF we can make it work, do we want to use the system to describe polycultures in general (which should be fairly straightforward, because the system mine is based on was designed to do that) and use it to describe guilds incidentally, or do we explicitly separate the two?

* IF we can make it work, then yes, some sort of searchable database, perhaps on a wiki basis with some sort of quality control, might be a good idea, but first we have to fix the problems with the concept. We are getting ahead of ourselves.

As ever the issue I think about definitions .
Can a system work?
Firstly we need to define system secondly work .
Only when we have these not so simple definitions can we answer the question.
Polyculture is a fine word and guilds are fine words but it needs a size component in my eyes . How big should a "system " be so that
a) we can study it in relative isolation and not be too relient on edge effects
b) we can reproduce it easily
c) not become a monoculture because of size

For instance if we assume that a guild is a cube based on a dominant tree for example it will have a surface area of say 6 x2 and a volume of x3
Whilst say a grid of 9 trees would have a surface area of 30x2 and a volume of 9x3 so the edge effects for each tree is now 3.3 relative to 6 of the sole tree .

scaling up to a 100 trees surface area equals 240x2 volume 100x3 thus each system has 2.4 relative to 6 of the sole tree but now we also have a mono culture

As for work
Are we talking outputs ? how do we account for imputs ? and how about time .

I dont claim to have the answers to these questions I suspect their are no "correct " answers . What ever someone is trying to find out this questions need to be asked at the start before you do something

Wow, very impressive, thank you!

Now, I want to do you the unpleasant favor of subjecting your work to ruthless, impartial scrutiny.

Just to mirror what I understand:
1. Your stated goal is a system that's "descriptive" rather than prescriptive.
2. Descriptions of such systems could be useful for research, for example, on pest issues.

I'd like to hear an example of how this could be useful for research and what that research process would look like.

Some thoughts:

I agree that such a system would not be useful for prescriptive practice, since sites vary so much.

One of the Permaculture principles I find very helpful is "go from patterns to details." Many of the useful Permaculture tools, such as "guilds" and "forest gardens" follow this principle, which yields us practical, simple tools to use. It's hard for me to envision the utility of a system that starts with specifics.

For example, utilizing the diversity/resilience theory in ecology is a "patterns to details" tool that helps us act very practically to control pest and disease problems by increasing plant diversity and incluing habitat. It is also elegant and testable, in that diverse systems can be tested and compared to monoculture systems.

Next, the Permaculture systems and the traditional forest gardening systems (such as hedgerows and tropical home gardens) I most admire and wish to emulate tend to be very complex. It seems like it would be very difficult to describe any of these using such a system. It would be impossible to describe my own forest garden in this way. To describe Martin Crawford's garden, you would have to define arbitrary "bounds" as you suggested in your example. That means that there would always be significant outside factors having an unmeasured influence on research. For example, pest resilience research on one arbitrary "patch" of Crawford's garden could be strongly influenced by species present nearby but outside the test "patch."

I see more potential value with very simplified commercial Agroforest systems, but since these systems are so varied, and currently in a phase of experimentation (and hence diversification,) it would be tough to cobble together enough data points on any one type of described system to conduct meaningful research.

As a final thought, we use a similar system for describing "native" ecologies in the US, but that system is inherently flawed by the assumption that these were "naturally occurring" ecologies, and that Native Americans lived "transparently on the land" without effecting their ecologies. As such, I've never seen this system used to describe the "wild" naturally occuring "recombinant ecologies" that occur in my biome, which are the most interesting to me in designing polycultures. I have often suspected it would not be useful in such a context in my biome.

Mike Hoag wrote:Wow, very impressive, thank you!

2. Descriptions of such systems could be useful for research, for example, on pest issues.

This use, or research, presupposes the acceptance of the concept of the "pest." I'm coming more to the position of not accepting that concept, in terms of my own gardens.

One last thought as I re-read through the comments here.

"Patterns-to-details" systems can be tested, but create de-centralized, self-organizing structures to propagate themselves.

Anyone can take the "tree, nitrogen fixer, mulch-maker, pollinator, etc." model of a guild and apply it in nearly infinite ways. They can "apply self regulation and accept feedback," tweak their plantings, then come back to this forum, or to their local Permaculture meetup and report on their results.

It's a "free" system that doesn't cost our community anything to maintain. When we create such a tool, we've "caught and stored" our intellectual work into a Permanant, self-maintaining tool, a bit of "culture." Bill Mollison's classifications would call such a thing a "regenerative" or "procreative asset." It's Perennial.

A data-base of specific plantings sounds like it requires a host, an organization, meetings, continued research and testing.... all things that require ongoing inputs of capital and energy to keep going. Such systems are like "annual agriculture" in that way. In Mollison's terms, they are "degenerative assets" that either break down or require continuous inputs.

Just thinking out loud. Clearly, you're freakin' brilliant. I hope you'll keep us informed of your thinking and work!

Mike

Mike Hoag wrote:

A data-base of specific plantings sounds like it requires a host, an organization, meetings, continued research and testing.... all things that require ongoing inputs of capital and energy to keep going. Such systems are like "annual agriculture" in that way. In Mollison's terms, they are "degenerative assets" that either break down or require continuous inputs.

Just thinking out loud. Clearly, you're freakin' brilliant. I hope you'll keep us informed of your thinking and work!

Mike

I don't know about "brilliant". I'm having a crack at an identified problem. I should have made it clearer from the start that I wasn't presenting it as something finished, but was opening it for input and feedback. The word "towards" should be a clue, but I should have spelled it out.

One of the biggest problems is identifying a patch, and taking into account contiguous patches. A suburban back garden surrounded by patio has some biological links to the rest of the world. Patches in a forest garden interact in many ways.

Let me give a grossly oversimplified example. Let's say I plan a patch with six apple trees. We'll call this patch A. Let's say I plan another contiguous patch, patch B, which has six plum trees. I expect that one in seven saplings won't survive (I'd hope to do better, but I'm giving this for sake of example). I buy seven apple saplings and seven plum saplings, and plant them in my tree nursery. The apples all survive, but two of the plums die, so I respond pragmatically and stick the spare apple in among the surviving plums. I now have Frequent, even Abundant plums, and an Occasional or Rare apple in patch B, but that apple will be affected by, among other factors, pollinators and pests attracted by the apples in patch A.

It's this sort of lack of clarity that undermines the whole system. If someone can work out a way of addressing this kind of issue, then we can think about questions of databases. I'm going to let my intuition work on it, but I don't, at the moment, have a solution. It may be that we're going to have to rip it up and start again, or it may be I'm on to something useful, but at the moment it's not - yet - a usable tool, so I don't want people getting overexcited.

"Taking a stab at a known problem" is what defines the cooperative work of communities, such as scientific communities. It's not yet often done in our community, so it deserves thanks. The other part is having a community that is both supportive, but can provide impartial feedback.

I agree that it requires work to become valuable for applications, but I'll keep it in mind, too.

One of the biggest problems is identifying a patch, and taking into account contiguous patches. A suburban back garden surrounded by patio has some biological links to the rest of the world. Patches in a forest garden interact in many ways.

I'm just going to sort of run wild in the bush, and type as I leap over stuff here: I guess that your pattern and species description technique is meant to sort of hyper describe an individual system's parts and interactions so that anybody hoping to develop (or perhaps more importantly: sees the potential in their site to develop) a similar system can get the best possible snap shot(s) of this system as it developed. The problem that I foresee and which has been already articulated is that we can try to describe what we as individuals see and what we think is happening, but all of that is purely subjective to our level of understanding as well as what is actually observable. When we add on to this the incredible ecological complexity that exists (or doesn't) in the surrounding biome (I'm developing my food forest with direct access to hundred year old forest which connects directly to old growth and some of the largest protected untouched lands on the continent; as opposed to someone in a similar location but in a very settled space that has none of those deeper possible links to Nature), then we are really throwing a wrench in the gears of pasting one onto the other. I'm not thinking that the last statement is your intention with this. The intention seems to be to give people ideas of systems of things that seem to be working in concert including climatic/location ideas, and the additional description of whatever else is observable adds to the potential of another to understand it. And I think that whether you call it guilds or polycultures, or some new term that you or we make up, is somewhat irrelevant in this developing science. Until these systems have developed beyond their youthful growing pains and inevitable failures and setbacks (which happen all the time in wild/natural settings as they succeed and mature, and even these get replaced over longer periods of time), into stable mature forms we wont really have a claim to successful systems, just-as you well pointed out-snapshots of stage(s). When we consider the incredibly messed up place that Toensmeir and Jacke describe when the visited Robert Hart's forest garden, which had virtually no planning, and seemed to succeed anyway, we can maybe take a deep breath and relax into this a bit more, knowing that we can likely do better if we have some ideas about where to start and what might we be heading for. In the past I think that food forests were developed from plants that were basically in the process of being domesticated or at least managed from local wild useful plants. The patterns and associations that were associated with these plants would probably have been taken into account. The clearings in which these plantings were placed, or allowed to exist, were small 'islands' in a sea of their wild predators/prey/family/friends/foes, and their indigenous climate. This is VERY different from what we are doing with modern permaculture. What we seem to be trying to do with permaculture is try to take plants from similar habitats but from a diverse spectrum of disparate locations, and mix them together, and attempt to sort out the science of how it all sorts out in the wash. If I'm growing Siberian Pine with local alders, with a chinese hardy kiwi climbing them, with potatoes from the andes, and dandelions from France, and a blueberry... how can we expect to really understand it, or what Nature might throw at it, particularly at this young stage? We can describe what we have done and observe as many interactions as we see, and hope that we come out with something usable for someone else, I guess. But even Eisenhower or Darwin would be scratching his hair to a bloody mess knowing the impossibility of the task of even description, let alone understanding. Considering that we are talking about billions of organisms in a cup full of soil, and these vary from one part of the yard to another, let alone across the spectrum of similar climates globally, I think that if such as system is developed it should be opened with a description of just how imperfect it likely is, but it is a snapshot of something that is happening, and that is useful, no matter how it's sliced up or mashed together. Sorry if all this seems somewhat contradictory. Yes, I support the database! It's an awesome idea. I hope that my thoughts have some value to it's implementation, if not simply food for thought about it.

I think what most people are looking for, and Paul has voiced this, is a simple card that tells you what to do with your set of conditions.

Have sandy soil: plant this.
Have shade: plant this.

So a database of guilds could be a replication of that.
Have shade and sand and a sun facing wall: use this guild.

People set their perimeters and a guild pops up on the computer screen -- tailored for their site.
The guild would be one that covers most or all of the functions we know about -- (from Jacke) "food, fuel, fiber, fodder, fertilizer, "farmaceuticals," and fun"

As for a more advanced notion of defining what a guild does, how it functions, what benefits what, etc, -- I agree with the previous post, nature acts in ways we can't understand so our best bet is to just try and harmonize with the conditions on the site (wind, contour, more habitat, richer soil, etc) and let the pieces fall where they may.

I think Hemenway talks about the fact that we shouldn't be so quick to attribute any success or failure on the site to our actions.

Usefulness in guild design might be a result of thousands of people working with simple tools (like the cards above) and sharing their experiences of similar sites rather than one group of pioneering permaculture researchers engineering an perfect guild.

William

Neil
You put another way and far more elerquently what I was saying . What is a patch ?
That's why I was thinking about numbers . We are in a stage I think of like genetics before Mendel , we know something happens we know it's complex and we don't know how it works . Mendels genius was realizing he needed to look a distinct components and see what happens first . So instead of looking at the whole wonderful complex web of life can we not just look at one bit at a time .
For example plant a poly culture of apples in a 3x3 grid with associated plants then next door on the same type of soil plant 8 nitrogen fixing trees around an apple tree with the same plants . Just examine or measure the central tree and how well does it do ( crop weight apples ? )
Next plant another nine apples with one plant different in the mix and see how well that plant does compared to your first plot and how well the central apple trees does .
Etc etc until you find out works .
As far as the cards idea I would have thought eventually a computer programme would be the answer . Give it the soil type and fruits you like and press the button:-)

David

First, I really want to thank Neil for starting this important topic, viva the neurodiversity of permies.com! What a tremendous contribution you are making to permaculture and permaculturalists. I love these times when someone really hits the nail on the head in terms of what they can contribute, really productively and amazingly, to the community.

Second let me say that I have not studied ecological classification systems at all and a good bit of what is being discussed above is way over my head. And I think that can be a valuable perspective too. So here are a few thoughts:

I think science at its most valuable keeps in mind practical, real problems experienced by many people in the real world and provides guidance as to what to do in the real situations they face.

I think a system with a not-too-steep learning curve might inspire many people all over the world to offer input, which would greatly increase the power and usefulness of the data due to its sheer volume. The dataset if very robust will inspire continual further input, a positive spiral of usefulness it would be great to encourage, or even essential.

For that and other reasons, I think the system needs a clear goal, and I'm going to be bold and propose a draft version of one, albeit paraphrasing what William has said above: The goal of this database and the data collection involved would be to help people dealing with the environments in which they live to design successful guilds, patterns and patches of polyculture along permaculture design principles.

I know Neil that this initially may seem to go a bit against your idea of it being descriptive, not prescriptive, so let me explain. I actually agree quite strongly that the data it contains should simply describe existing conditions observed in the real world and make no claims of applicability to anyone else's particular situation. At the same time, I'm equally sure that that should, in fact, be its purpose -- if no one can use the data in any real-world situation, we've done nothing. So I'm just saying that it should perhaps contain all the major, long-identified useful parameters -- things we look at and consider all the time about any particular ecological environment in order to decide what to do -- to cast light on possible relevance/irrelevance to the database user's own personal situation.

As to the definition of a guild, a patch or a polyculture: I am not sure that this is actually of prime importance to this marvellous project or its usefulness. At least not in the cut-and-dry sense that the "hard" sciences often strive for. I'm sure it would be useful to have extensive guidelines for what each of these phenomena might represent, with copious examples, but in the end it's just to allow us to search the database better and understand better the style and scope of what's being described. I think by giving the definitions clean and clear edges, we might often end up cutting off some very interesting data about real situations that might otherwise enlighten our search if not previously defined out of existence. So I'm all for keeping definitions fuzzy, overlapping, and full of examples and photographs.

And I'm all for giving people a bit of subjective leeway in providing their views, hopefully supported by photographs so others can also opine, as to "this is a guild of these four species surrounded by larger patches of whatnot and so-and-so." I think we need to allow for productive input now, and perhaps sort out details later and according to context.

As an example, I would say that almost anyone reading this could identify a meadow if they saw one, a clearing in the woods. And we could pretty accurately define the edges. A child, given a satellite photo, could draw over the edge easily. But if one had to scientifically define what a meadow is, one could get washed away in a storm in a teacup. So guidelines, yes, there must be some guidelines I suppose for when a meadow is big enough to simply be a field surrounded by trees. But I wouldn't put too much effort into those definitions on the front end as I think their relevance and specific characteristics would shape up greatly on the back end, after inspiring many thousands to contribute observed field data to the project.

I would love it if this project were simple enough to inspire thousands of people, at least people who know very weill how to identify local species and environmental conditions, to simply go out into their natural environments and describe what they find, adding the info to the database. And add pictures of same. Now that would be a huge contribution to permaculture and to humanity.

In regards to the definition of spaces by percentage of canopy cover: I think that might be a useful starting point only if hell-bent on only providing descriptions of things we could describe as "guilds." I have a different idea of what I think might be more useful though. Again, with its fuzzy definitions, I immediately think of the classic permaculture layer system. So a useful observation "event" would include the characteristics of the canopy, understory, shrub layer, herbaceous layer, groundcover, rhizosphere (more detail*) and the vertical or vining layer. In effect it is the above system x7, i.e., I think it might be helpful to have a description of each of these layers separately, as is contemplated in the departure point of the UK system which seems to be only based on the canopy layer. I think it is important to have a system that makes no assumptions about canopy cover and simply dedicates itself to describing an interesting environment that someone has found and wishes to describe, for the purpose of inspiring as many people as possible to contribute quality inputs from their local environments all around the world.

In addition, some vital inputs I can think of would be: latitude, elevation, Koppen or other climate classification system, average annual rainfall, season of observation, and climate antecedents (a mild winter, a 16-month drought, a recent hard freeze, average weather for the season, two months of the usual summer drought after a wet and cold spring, etc.), and numbers, coverage in square meters and spacing of species observed.

*Information re the "rhizosphere" or soil might include everything one is able to test for: soil type (sandy loam, washed-out clay, etc.), mycorrhizae inoculated or naturally present, pH, mineral structure, etc. I think this part should allow for many details to be added or for few, according to what the observer has the ability to report accurately.

And as far as accuracy and completeness in general, I think a good system should allow for reports/observations that offer a variety of degrees of completeness and accuracy. So that a person knowing only a good amount about local botanical species and with little or no equipment could offer a good report, and someone who also knew how to test for soil charactaristics and the bacterial/fungal environment to a huge degree of accuracy, and with access to a great lab, could also register an "observation event" in the database, albeit a much higher quality one.

That's all I can think of for now, really excited to have an idea developing of a system that could allow people to plant/plan better guilds, wherever in the world they are!

Edited to add a few more ideas.

Nice post Dave.

I'd just add that one initial goal of such a project might be to do an inventory of past and current plant databases. I know of Apios (apiosinstitute.org) which seems to be attempting to give people a plant database. Another database is
practicalplants.org, and yet another is permacultureplantdata.com

They all seem to have some of both adaptive strategies and guild-design.

Is there something that these databases are *not* doing that this discussion doesn't cover. Basically, is there an open niche? Or are people trying to fill a void that simply doesn't exist or wouldn't exist if those databases were more ___________ (fill in the blank).

Other plant databases:
http://www.perennialsolutions.org/perennial-plant-databases-organic-agriculture-gardens-permaculture-hardy-hybrids-varieties-information.html

Attempts to make "palettes" for various climates/elevations.
http://www.perennialsolutions.org/useful-perennial-plant-profiles-hardy-organic-gardening-plants-permaculture-urban-resource-garden.html

William

Hi William et al,

Ah, yes, the ego-rush of us ourselves inventing the wheel. I do agree it is best avoided. Thanks for knowing about and providing the links to all those very useful resources!

OTOH if we have a good idea and some momentum behind it, the effect is really great if we can use that too, despite any wheel-reinventing effects.

In any case, I took a brief tour of the resources you listed.

I didn't see what I wanted to see (almost) anywhere. And the process of looking for what I wanted clarified for me exactly what it is I am looking for so let me articulate that before I forget!

I would like a really clear way for loads of people with knowledge of their local species and a camera (and with at least rudimentary abilities to describe local setting, soil and climate) to go out and describe the species they find thriving together in nature, or in food forests and other planned polyculture settings. In a structured way that would aid searchability by people looking to establish guilds or other polycultures in similar climate etc circumstances. And with a time lapse feature too for people to describe the evolution of the site and the specific guilds/polycultures it contains over time.

Neil, William and others, does this jibe with what you're looking for or what you think would be useful?

But back to your list of resources, William. As you said, it would probably be a good idea to have a look at what's out there already before embarking on a project like this. I looked over the links quickly and I think they might serve as good background information, and in one case give us some ideas for structure. And, no one seems to be really doing what we're talking about here.

The Apios resource seems to be extremely limited at the moment. For instance, only three polycultures seem to be listed, and all three have only two members, so that would be pretty far from our permie ideal of 4, 5, 6, etc. consciously planted species with more certainly naturally present and there to describe for whoever has the inclination. In any case, their aims don't seem to be as rigorous or scientifically helpful as what Neil is suggesting.

permacultureplantdata.com seems to be a paid membership site that seems to be directed mostly at permaculture design consultants, people who make a living from permaculture. (Which is great, no knocking that, of course!) I notice that in order to create data in their database, you need to register as a researcher and pony up $40/year. Again, not knocking it. That probably makes for high-quality data as only serious types will pony up any dough at all. But for instance, it might put off the lovely and knowledgeable hippies at my local "nature school" who are always wandering around the local mountains observing species -- it would be so nice to just give them a standardized and user-friendly way to report what they observe, with results useful both for scientists and (most importantly) the permie on the ground looking to set up successful and productive long-term relationships between plants. Anyway, worth exploring, I believe I've read a reference before to their high-quality data, but in the end, it might not work out as their end game seems to be to serve the permaculture consultatnt rather that the average permie on the street, which is where the vast majority of energy, questions, and real projects come from.

practicalplants.org, as a wiki, seems like it might in principle have been a better fit. But alas, it looks as if the project was started and abandoned in 2012 and since then the forums have only been updated by Russian spammers. It might be interesting to keep in mind the structures/parameters they have set up for plant search though, even though they don't seem to work.

The first link to perennialsolutions.org is to a list of plant databases, which may sound promising, and many may be good resources, but only two seemed directly relevant to our quest of setting up better guilds and guild-like things. One is the Apios resource discussed above, and the other is a "bosque comestible" (en español) project which seems like it was intended to be directly relevant, but alas, it seems to have disappeared from the web. The second link is to a page titled "Useful Plant Profiles." That info may also be quite good, but it does not seem directly relevant to successfully designing your own guilds, i.e., sybiotic relationships between plants and other growie elements.

This was just a cursory review, so of course would require a more serious look if a project gets off the ground here. But so far, this mini-tour of the links kind of confirms my conclusion from previous research into "how the heck do I plan successful guilds and what should I plant where?" issue: There just ain't much specific, usable advice out there. Companion planting, maybe, i.e. "plant this next to that," but then you need to spend all day looking up compatibilities and incompatibilities of pairs of plants to finally work up a few guilds that seem feasible. I nearly drove myself nuts last spring doing that. Actually, now that I mention it, I'm vaguely remembering a really helpful companion planting site at least for annual vegetables, which is the only thing I can plant in my garden anyway, that basically instructed you to divide up your annual veg into 4 or 5 sets that all work well together internally (maybe not full-fledged guilds but heading in that direction), and then practice annual crop rotation. Maybe tomorrow when I'm birghter and fresher I can find it and add that to the list of background info.

Last thing for today: I am sensitive to Neil's desire to define things, even if I think leaving some fuzz around the edges in this case will be helpful. So I've been thinking of what actually a guild is and what we could say about it, definition-wise. The intelligent thing to do would be to read the literature, I'm sure Bill or Geoff or Sepp or Masanobu or one of those all-stars has written something pretty definitive on the subject, but as I said it's late and I'm not that intelligent at this time of night. So I will speculate. A few characteristics I might throw out:
--Minimum of 3 visible major species planted or ocurring together, and usually more
--Seeming or proven symbiotic relationships between the species
--Use or occurrence of layering
--A maximum size of approximately: (before it's so big & complex it becomes some other kind of polyculture)
--Maximum of 1 tree in the canopy layer, and/or in the understory (before it's larger than the unit of a guild)
--Stability of symbiosis over time: Not sure what to say, if anything. I.e. one plant doesn't muscle the others out

Feel free to ignore, correct, add to or subtract from the above as I'm not running on all my cylinders at the moment

OK, that's all for tonight, time to hit the hay!

Just to say I'm not ignoring the conversation. I'm reading, taking a series of good points on board (there has been some excellent input today!) and trying to make better sense of it all and fix at least the more glaring problems.

I've sort of got a handle on it in my head, but as a usable tool it would be as unwieldy as a 50-kilo hammer.

I think the guild-patch distinction is fixable. It gets harder each time I look at it, but it's fixable.
I think a sufficiently clear definition of a distinct patch can be found (no reinvention of the wheel here: we can just lift the idea from elsewhere)
I think there are multiple options for handling the mycota (as distinct from the rhizosphere).
The First Law of Ecology threatens to burn the whole thing to the ground.

Keep brainstorming, please. I'm thinking, but don't yet have something that looks properly coherent, or at least usable.

Here is an example of crowdsourced information in nature which is used by scientists. USA National Phenology
I have used there forms which one can print and take to the site to record and then report on line.

Have you tried the analog method or build a guild method?

There is a the database we have that assists people in developing resource guilds.
From our site:
The Natural Capital™ Plant Database is a repository of temperate climate plant information for ecological design. Our partner designers have combined the best sources of plant research and documentation in order to provide the highest integrity for a wide variety of users. Whether you are a first-time gardener or an experienced permaculture designer, we hope you find the information you need to take your knowledge of plants and ecological systems to the next level. We use citations from multiple sources and provide detail on plant characteristics, tolerances and behaviors, ecological functions, human uses, concerns, and plant associates.

I think with works from Martin Crawford, Dave Jacke, and our book, Integrated Forest Gardening, there is sufficient definition and models for understanding the forest garden model. Plant guilds are well defined groups of plants with integrated mechanisms for facilitation (I hope that is not too many syllables); whereas Food Forests is a vague reference to woody agriculture.
These are grouping of polyculture plant guilds.


This is polyculture guilds for terraces.


The proximity, structure, and the ecological services define the plant guild's properties. I think it is a graphical representation you are looking for.
Keep the scientific manes in a Key and witness the shape and placement of the plants that create resource sharing and partitioning.

Perhaps I should retitle the thread "Towards an Open Source Formal Framework...".

I'm not interested in something that only affluent professionals can access. This needs to be usable by someone on a tight budget. There are other discussions on the site about elitism and exclusion. It's a big enough issue already.

I am not sure what you mean by that. Affluent Access. The database? Books?
The database is free unless you want other services. Authors have to eat too.

The notion of some sort of database is secondary to what I'm wanting to do. I'm interested more in defining, describing and comparing. I don't think a database is a bad idea, but it comes later.

Here's the problem: if we use your trademarked system, or even base ours on yours, we end up in a situation where you can charge to use information I would much rather was free. Authors have to eat, and as an aspiring author, never mind someone who thinks everyone should eat, that's important to me. I don't want to even base something on yours given that I don't want to find a message from your lawyers in my mailbox, or hear about such a message in someone else's, locking down the system unless money is paid.

I'm not saying you would do that: I'm saying you could do that, and that's enough for me to want to have nothing to do with it so, for the record, I'm not even going to look at your site.

In an attempt to keep things on track....

Neil Layton wrote:The notion of some sort of database is secondary to what I'm wanting to do. I'm interested more in defining, describing and comparing. I don't think a database is a bad idea, but it comes later.

I think this is what you were saying in the opening post when you stated that you wanted something descriptive not prescriptive. A database is inherently prescriptive. You type in your info and it spits out some useful things.

So, I think it might be best to redirect the discussion to what Neil seems to want.

One thing that strikes me about the initial post is the absence of animals. I understand that you're mainly aiming this to deal with plant species, but animals show up even if you don't plan on it. There are ways of planting that may encourage or discourage animals, depending on your goals. An example of that would be Mark Shepard's use of animals, fruit, and nuts.

Even the interaction with infrastructure could be important both for animals and plants. Don't know how or if that fits.

The other thing that strikes me as important is the idea that functions and yields differ from site to site, depending on one's proclivities. There seems to be a disconnect between prescribing a sort of simple template of what would be "the best" in a given area and "describing" how one system is inherently better based on the needs of a specific landholder. In the second case you are basically working by example and each would-be landholder might choose the example that suits them (sometimes despite the fact that the future site is actually not analogous). As for the simple template, that's where you get into the whole database thing.

William

William James wrote:

One thing that strikes me about the initial post is the absence of animals. I understand that you're mainly aiming this to deal with plant species, but animals show up even if you don't plan on it. There are ways of planting that may encourage or discourage animals, depending on your goals. An example of that would be Mark Shepard's use of animals, fruit, and nuts.

Again, this is something that would be descriptive, and dependent very much on ecoregion. That's just basic biogeography.

That said, these novel ecosystems will attract a novel fauna. I think that's something to be studied, but is off down the road somewhere. I'm inclined to kick that into the long grass, to mix metaphors. if you want domesticated livestock as part of the ecosystem, that would be another matter. My own inclination is to avoid livestock entirely, for reasons that are off topic, but it may be something we need to discuss. We're talking about flora, and talking about talking about (sic) mycota, but maybe we should involve fauna as well.

Then there is the question of livestock barriers, which feeds into the question of interactions, which goes back to that First Law of Ecology problem I alluded to.

William James wrote:Even the interaction with infrastructure could be important both for animals and plants. Don't know how or if that fits.

Again, an interactions question. I'm on it, but it's the single biggest problem I've found so far, so feedback on this would be especially useful.

William James wrote:The other thing that strikes me as important is the idea that functions and yields differ from site to site, depending on one's proclivities. There seems to be a disconnect between prescribing a sort of simple template of what would be "the best" in a given area and "describing" how one system is inherently better based on the needs of a specific landholder. In the second case you are basically working by example and each would-be landholder might choose the example that suits them (sometimes despite the fact that the future site is actually not analogous). As for the simple template, that's where you get into the whole database thing.

William

Perhaps a better way of looking at it might not being that a system is "inherently better" but inherently optimised given the needs of the land manager? You make a good point. This is partly about being descriptive not prescriptive.

As for infrastructure, one crude methodology would just be to describe some typical features. I'm thinking a house-barn-igloo-office typically has a "hot" side and a "cool" side, depending on exposure to sun, it also has wind deflecting properties, so some sides will be more wind-prone than others. The plant guilds that go around infrastructure could be (in some part) guided by the inherent features of all large and permanent objects on any site.

There's also edge effect where "nature" meets the infrastructure and there whether it's hot or cool or whatever affects the surrounding guilds.

William

How did I miss this thread until now? Neil I hope this is still something you're thinking about, as it seems like an incredibly important area for Permaculture to advance.

If this is still something people are thinking about I hope this helps; if not, it's helpful for me to jot my thoughts down  

Neil it seems the main problem you are concerned with is the first law: everything is connected to everything. Hear me out as this will go a little far afield, but I'll loop it back I promise! I propose that your concern is extremely well warranted, and that the problem is much much older than Permaculture or Ecology (as approaches or methods of discovery/inquiry). I think what you're running up against is what the Ancient Greeks called the problem of the One and the Many. There is a thing that exists, but that this is in a class of things; it is defined by the grouping and the grouping is defined by them. An animal is of a species, and the species is composed of animals. The closer you look at this problem the more difficult it becomes.

For example, the closer we look at the human, the more complicated it becomes. From a scientific standpoint, what is a human? Is it the thing that's made up of human cells, with human DNA? Not really, as we know that 90% of a human's cells aren't human cells. How about a civil approach: is a human a political being, a citizen who voices opinions and takes action that impacts the humans around them? Perhaps, but there are increasing amounts of humans without coutnries or even regions to which they are citizens; there are increasing numbers of humans totally isolated from their fellows, living in illusory worlds of fantasy and digital reality.

If you try and define something by what it is, you run up against what it interacts with. If you try and define something by what it interacts with, you run up against what it is.

I propose that the problem is the solution: perhaps the very valid concern you present of Patch A's interactions fundamentally affecting the success and even identity of Patch A is the key to how this approach can work.

The answer lies in moving beyond the One and the Many in our thinking. I believe the best way to do this, (best in terms of how accurately it can describe what is happening, how fruitful the approach can be, and how verifiable and concrete its conclusions can prove to be), is to approach things as a continuum. The World is composed of nations, made of regions, made of communities, made of people, made of ecosystems, made of microbes, made of cells, made of organelles, made of atoms made of even tinier bits. A Planet is composed of Biomes composed of ecosystems composed of communities composed of species... well, you get the idea.

At each level, there is an identity that can be defined by three things: population, history, and edge interactions. We understand this best at the level of an ecosystem I think. When we talk of a forest, a polyculture, a field or a garden, we Permaculturists typically use ecosystem terminology (with a healthy dose of Permaculture perspective, of course   ) We speak of soil type, elevation, slope, landscape features like rivers, ponds, boulders and other geographical furniture, present species, rainfall, climate, previous influences on the land, species interaction, wind water and sun. To put it in the categories I'm proposing, we speak of populations- species, soil, landscape features; we speak of history- previous influence, climate, elevation, slope, species interaction; we speak of edge interaction- elevation, slope, climate,  wind, water, sun and the geographic features with which they interact.

Population is made up of the constituent material parts of a thing: the microbes in our body, the species in an ecosystem, the ecosystems in a Biome. At each level there is something that is more than the sum of its parts, but population is the inventory of those parts. Each of those parts have parts, but typically we can stop at one level down in our consideration (since those parts are more than the sum of their parts, as a fish is more than a bunch of microbes, a skeletons, scales, gluey stuff etc).

Hisory is a two-sided thing, I think. On the one hand, it is the place as it exists in a given moment of time, comprised of all the events that have led up to that moment. In this way history is the descriptor of each and every particular thing, as history as I mean it here is just another word for how they came to be where they are. On the other hand, however, history is something we use in a prescriptive way. If Jane has run 15 marathons in the past, many over rougher terrain than the upcoming race, and Jim has never ran for more than an hour in his life, I will say with a high degree of confidence that Jane is going to finish well before Jim does, if he finishes at all. If a species has for thousands of years grown and adapted to arid, desert conditions, and another species has for thousands of years thrived in a coastal drylands that receives little rainfall but lots of fog, I would be confident that the coastal plant is going to fare much better somewhere with regular fog, even if both require the same amount of annual rainfall to grow. As you can see in these two examples, history and how it applies has components that must be considered on multiple levels of the continuum: those specific to an individual organism, those specific to the genetics of its species, those specific to the ecosystem, like soil communities, microbe symbiotes, etc.

Edge interactions, or the flows of energy into, throughout, between and out of a population, are probably the most complicated, and the area we are as a civilization the least cognizant of. By my lights, most of Permaculture has been the attempt to develop understanding of this. Where the air meets the ground, the sunshine meets an opaque surface, the river meets the riverbed; where the root meets the soil, one organism meets another, one 'patch' in the forest meets the others. I think that most of the time something fails that seems like it would have gone wrong, it is because someone failed to take into account edge interactions, or took something for a matter of population that was actually a matter of edge interaction. The building was made of steel beams and thick concrete, but they hadn't considered the impact of such a strong earthquake. The trees plants were well suited to the region, but in the dry season can catch fire and were planted directly across the prevailing wind. When Bill Mollison was asked about the book Carrots Love Tomatoes, his answer was "in that book, you will find plant associations between plants with totally different growing seasons. It's useless". Perhaps it's not entirely useless, but this was his point: the author was talking about identities, not minding their history and edge interactions. You get the same thing from Geoff Lawton, who never gives a talk about Permaculture without at least 1 use of the words system, element and design. 'Design' is using 'elements (populations with history) to influence and alter a 'system' (ongoing pattern of edge interactions).

I was thinking about this earlier with allelopathy: when I first heard the term, it was to hear that some plants just kill off other plants to reduce competition, the big example of course being the black walnut. The more I have researched it, however, the clearer it has become that allelopathy, far from being a single, inherent aspect of a plants identity, is the word for a huge range of effects and plant interactions. Some plants called allelopathic interfere with other plants ability to uptake iron; some with their ability to photosynthesize; some only impact others when they die; all of them affect certain plant species, but not others. Here is a good overview of that. Really, allelopathy is (or should be) the word describing a type of interaction. One that is specifically between plant organisms, and that is harmful to one or the other. I'm confident that over time we will develop a much more robust vocabulary for interactions between things: thing A being 'good' or 'bad' for thing B is just insufficient, as are 'parasitism, mutualism' and the rest.

So with time and this kind of approach, we could start to understand the edge interactions of an apple-cherry guild much as we're coming to understand the edge interactions of rainfall. Perhaps the more we learn of these patch entities, the clearer our approach will be, just like swales as tree growing systems became obvious the more water harvesting features were built across landscapes. I think so, at least.

allelopathy

Allopathic Medicine For example uses the premise That if you kill the organism associated with a disease you will cure the disease. It doesn't work exactly that way. The organism is just a symptom of the body's disease which still exists and will manifest itself in another symptom
Allopathic Agriculture Base on the same premise farmers were taught that if they killed the weed, insect, bacteria, virus there problem would be solved. It doesn't work exactly that way. If the balanced continuum of life is not present, the so called pests will persist, even adapt to the poison in order to do there work.
Permaculture Design for abundance by optimising the continuum of life.

Note: Allelopathy (allelopathic) is a totally different word (with a different meaning) than Allopathy!

Neil Layton wrote:...some sort of searchable database, perhaps on a wiki basis with some sort of quality control, might be a good idea...

I'm also just discovering this amazing thread.
Has there been any progress on the idea of a wiki since the OP (original post)?
We could possibly use the Appropedia.org/Permaculture_wiki as opposed to re-inventing the wheel.

Just found this thread.  I really like the idea.  Any movement yet?