Share your knowledge: Contribute to the "Crop Wiki" project!

General Discussions
21

I would be more likely to contribute to this project if it had a database format. I can think of many projects that have done similar work and I wonder what this contributes to specifically for adaptive gardening?

To have a database where I could easily search up all the crops that have an crossing rate of plus 50% (or “medium” or however you want to operationalize it) would be very useful to me though.

Or a database of existing landrace projects for a particular species. With contact information.

Or a search for all the crops I can broadsow easily etc.

With a clear text document, this kind of information becomes more difficult to search up other than having to analyze all the text / read it through.

1 Like
22

Good idea. You have an example we can look at? There’s a group that’s been writing crop specific emails that have gone out through the summer, focused on adaption methods, but we haven’t taken the next step of turning them into web guides or putting them online.

23

I like how Plants For A Future (PFAF) has collected +8000 plants in a database format that you can then search with specific parameters. I use the advanced search function to explore possible new candidates for growing projects. https://pfaf.org

In an ideal world, GTS would collaborate with PFAF to build on their resources and add the knowledge that we’re looking and that is also lacking. PFAF has collaborated with Eric Toensmeier to include all his data from the reference work ‘Carbon Farming Solution’ so I think it is possible they would be open for other collaborations.

Another example, in Danish, with much less data and coverage of plants, but a very intuitive interface: https://dyrk.nu On this site you can order crops after sowing time, see the optimal germination temperature etc with pedagogical icons.

24

I think pfaf is unreliable. Unlike Wikipedia or the collaborations here in this forum, it doesn’t feel like I can correct errors.

I’m sure it has some good listings.

But I pulled up Cucurbita ficifolia just now as a spot check. That listing has several significant errors. I’ve corrected some similar errors already on the corresponding Wikipedia article, but I don’t feel like I can do that with PFAF.

Do you have a different experience working with that site? I honestly do not reference it anymore because I’ve found too many listings that have incorrect information.

2 Likes
25

I agree! I have tons of Info to Contribute on both Domesticated Crops, Wild Edibles, Phylogenic Trees & info on which crops can & can’t cross.

The formatting is what makes it difficult. I really enjoy this format I’ve already made, could GTS take notes from this Skeletal Structure?

image
image549×503 30.2 KB
image
image508×505 96.6 KB
image
image383×506 123 KB

image
image524×507 118 KB
image
image521×500 59.5 KB
image
image264×505 92.2 KB

image
image378×497 80.2 KB
image
image338×499 101 KB
image
image423×493 83.1 KB
image
image378×504 103 KB
image
image304×501 65.1 KB
image
image378×504 85.9 KB

image
image1052×499 126 KB

@julia.dakin The formatting is simply the reason why I can’t really contribute. If it were more like a Wikipedia x Pfaf style data base, I’d probably fill out the entire thing as I already do for my own personal notes pretty much everyday (So much so, I think I’m the reason google docs started counting Google Doc Files as Storage :sweat_smile:).

I use PFAF all the time! Fantastic Resource! It’s my goto for when I need to see if something is edible (Of course it isn’t my only source).

That would be EPIC!

ehh… I think both have the strengths & weaknesses. PFAF has a lot of Scientific names to update, Wikipedia has so many wild edibles to add on. I use both regularly & Combine that with Research papers, Foraging books from experts like Sam Thayer, John Kallas, Green Deane, and my Own Personal Experience.

There’s also a lot of excellent Wild Flower websites to help ID plants, which I found very useful too!

Interesting, I haven’t seen any big errors on the Cucurbita ficifolia page on PFAF. I just saw some more edible traits missing, like Tender Greens/Shoots, Flowers, Immature fruit are all edible if not bitter. I think they should’ve mentioned Don’t eat any part that taste bitter, because those are the toxins for each Cucurbita species & many other Cucurbitaceae plants.

Also the USDA Hardiness zone is Completely useless for annuals, as annuals don’t have to survive winter, they just have to make seeds before winter.

PFAF use to have a Comment section but it seems no one can make new comments any more. I’ve never tried to edit a Wikipedia page, so I have no clue on how that works.

PFAF sources it’s info form sources listed here. I very much appreciate that because if there’s anything I doubt about the info, I can look it up from where PFAF got it.

PFAF has been mostly correct, Im curious about which plants entries had incorrect edibility info? Perhaps I just haven’t seen them or just ignored the incorrect info & forgot about it.

2 Likes
26

If someone from PFAF asked me to review their C. ficifolia page to point out potential issues, this is what I notice at first:

  • It is missing many common names, including any common names from the region it is native to
  • The range is listed as “East Asia, obscure” which is blatantly incorrect
  • The relatively low edibility rating for a squash is suspect to me
  • Lists ficifolia as a perennial instead of an annual, following outdated or copy/paste information
  • Indicates that ficifolia is day-length sensitive as a species whereas experience here and among other firsthand growers is that it is not day-length sensitive, or that only some varieties are day-length sensitive
  • Propagation instructions only describe starting it in a greenhouse

If I was asked to give my overall impression as feedback, I would say that it comes across as a page that has not been edited by a human who is looking at the total information. I would say that in general it relies on references that are too far removed from people with firsthand experience growing or using the species.

My question is, if I wanted to help improve PFAF, is there a way for me as a volunteer technical writer with expertise on this topic to help make additions and corrections?

1 Like
27

I have noticed a lack of usual common names, and the edibility ratings are very strange, almost like they are only based on personal opinion.

I fear this type of thing is only going to continue with the overuse of AI, I have seen so much (sometimes potentially dangerous) incorrect information lately.

28

Yea, most of these are incorrect, I mostly just ignore the false info & scroll to the Edibility. Perennial Ficifolia is outdated info for sure, altho I wonder if technically every squash is actually truly short lived perennial, it’s just disease & the stress of making fruit kills the plant. Somewhat similar to how Indeterminate Tomatoes are truly perennial if frost doesn’t kill them.

Yea the edibility rating is kind of subjective. I’ve always took it with a grain of salt.

Well that makes sense, I’ve never read the Propagation info. I mostly use PFAF for the Edibility info, never was interested in the propagtion, day-light sensative, how to grow, ect info cuz I know PFAF is not good for that.

I don’t know. It’s something you’d probably have to email them about. If you’re gonna Volunteer, I’d probably volunteer too.

Yea, common names are kind of stupid, so many plants share the same common name like Black Nightshade or gourd for example. Scientific name are much more consistent, altho they change when new phylogenic trees are updated.

The Edibility rating is based on Personal Opinion. For the Edibiltiy part, there is a (K) to note that the author is describing the plants edibility to separate from what the other documented sources say.

Yea… AI is only as good as the sources it comes form. Even Forager Sam Thayer said most foraging books copy & paste all their info (Sad to say, it’s been going on well before AI).

Having intuition, Not ignoring your gut feeling, & knowledge will help sort though the AI BS.

1 Like
29

I feel like we’ve come full circle on people needing intuition and common sense, and Darwinism will follow.

1 Like
30

I will go back a bit to where I left this discussion. I mentioned PFAF as an example of how to use a database and the accompanying advanced search function such a database provides. I mentioned the possibility of collaborating with PFAF as an example of how to write these crop guides in a way that would build on existing information (that includes correcting false or out-dated info). It could be something else than PFAF. The question was: Do we have examples of crop guides being written in database formats?

@markwkidd, your question about how you could add information to PFAF, touches on something that site lacks - an open interface to the world, like Wikipedia. A way to collaborate and improve the site.

If I could dream a bit of a site for crop guides, these would be the ingredients

  • Database organized
  • Advanced search function
  • Heavily referenced (sets criteria for evaluating truth value or at the very least offers pointers to readers for where to learn more)
  • Collaborative functions, ie. users can add and edit, perhaps only through prior membership status or changes having to be validated by editors (that kind of stuff is easier to decide through experience; again, wiki sites are good case studies)
  • Free and open source ethic
  • Initial purpose to have use value for the adaptive agriculture community, but built with an openness to be potentially of use to any farmer or gardener.
1 Like
31

It’s not specifically for edible plants but
https://www.wildflower.org/plants/
is my favorite and most used plant database and would probably be a good example to work off of for certain features/layout.

32

This is helping me refine my thinking. I genuinely would spend some time improving PFAF if that’s something they want. I’ve already do that on species articles I’m interested in for Wikipedia.

In terms of where I have the most interest and energy for new work, I would find much more value in a smaller database where I knew there was a designated, active volunteer curating contributions to each listing versus a large database that was less interactive or allowed anyone to make edits without review.

Not that both kinds of resources can’t exist. But I am much more interested in a resource created by people who can reasonably be expected to have grown the crop. Or maybe another crop in the same genus.

2 Likes
33

You guys are the best, but my brain and time are unfortunately saturated, but I really want this project to happen, I’m nominating you guys. My request is that we could create an email text that pulls from various boxes in a database.
Dummy example

=CONCATENATE(
“Dear “,A2,”,”,CHAR(10),CHAR(10),
“Thank you for your interest in growing “,B2,”. We’re excited to share these “,C2,” seeds with you along with some growing guidelines.”,CHAR(10),CHAR(10),
“Growing Instructions for “,B2,”:”,CHAR(10),CHAR(10),
"1. Planting Time: ",D2,CHAR(10),
"• Soil temperature should reach at least ",E2,“°F”,CHAR(10),
“We’d love to hear about your growing experience. Please share photos and updates of your plants with us at “,S2,”.”,CHAR(10),CHAR(10),
“Happy growing!”,CHAR(10),CHAR(10),
T2
)

And that is because I expect the people who use databases are not the people who most need the information :slight_smile: In our local projects people say they want emails, and I’d LOVE to be able to write a planting or seed saving email for 100 different species using a formula.

2 Likes
34

Yeah, programming is cool as hell. The kind of automation I like. We probably need someone with much better programming skills than we have here in the thread (you already outrun me with your example, Julia!). I have found lots of overlap between gardeners and data-savvy people, often in the form of IT people leaving the computers to start growing a garden. Maybe we could find someone (2-3 people) in the periphery of this community willing to begin hacking away at this project? I would even fund the project with a monthly donation.

35

This is what I’m working on right now, which is actually very cool, for the Seed Library Network. It generates seed packet labels based on information you put in in the ‘Information page’. This is my practice sheet so ignore details.

But you can see the potential if I had a hundred species to draw from a database, for any seed project person/seed librarian who doesn’t know about that species, they can still accurate create labels for it. Auto Seed Wrapper. And can create custom size labels, make them 2 sided, add photos etc. So yes we need a programmer! But also with Claude and Gemini it feels like knowing programming might be becoming obsolete? It’s more a matter of designing, then getting AI to write the code and formulas. But I really am speaking beyond my depth here.

36

If I had the option between using a database edited rigorously by experienced people in their field and referenced with sources versus an AI-generated database, I would choose the former in a heart beat. And with my understand of how LLMs work (they cannot evaluate the truth value of anything, but they can give the impression of doing so), I don’t see how this will change in the foreseeable future.

1 Like
37

A lot of folks will refer to a spreadsheet as a database, which it technically isn’t, but it can take the place of one in simple situations such as what it sounds like you are describing. It’s just simple completing a form of sorts, multiple times, using information from different rows of a spreadsheet.

For example,

Dear {fist name} , {last name}
Here is the information you requested about {species} You should you plant your seeds when soil temperatures reach {sow-temp} and so on. The brackets I used are just an example, syntax varies between programs and applications.

An email marketing program like Constant Contact or Mail Chimp may be capable of doing exactly what you want with a minimum of hacking, or maybe none at all. The spreadsheet could be created in something like Excel or Google sheets and uploaded to the service or created directly in the service. And you don’t really need any programing or syntax, just name the columns and rows of the spreadsheet so the program knows where to get the info.

These are just simple old fashioned computer programs, not AI. They don’t make judgements; they just do what they’re told.

38

I’m learning now about new capacities for spreadsheets and emails, so I agree you are right that a spreadsheet would work for what I’m talking about, however I don’t think it would for what Malte and Mark K are… maybe they can explain why/why not.

Also @malterod not talking about AI generated content, just the code /programming for building the platform I think you’re talking about, may just need designers and not programmers these days.

39

I am posting to share this web page that compiles all of the guide emails that were sent to people that ordered the Cucurbita ficifolia / chilacayote squash seed packs for 2024. There were several of these guides for different crops that were sent out in emailed installments over the northern hemisphere growing season.

Because the messages were only sent to those growers who ordered packets, I assume that some people who are active in this discussion may not have had an opportunity to see an example of the material yet.

As a gardener and a writer, I’ve enjoyed contributing material to this guide. I think this is a practical format that can complement more quantitative, database-type information. I like the way that these GTS email guides prompt me to reflect on the different phases of growing (and eating!) a crop. Each installment works feels like a friendly reminder of what to be planning and what to observe during the different seasons of the year.

Link: Compiled 2024 chilacayote grop guide emails

Edited to improve clarity.

2 Likes
40

This is an excellent profile! Lots of great info. Thanks for sharing Mark. Now im really hoping to eventually get some seeds for this species

1 Like