Talk:Cruciferous vegetables

From Wikipedia, the free encyclopedia

Contents

[edit] More visual representation

I would love for someone to think of (and implement :) a more visual representation of the concept that inspired me to build this table in the first place: a map showing foods with taxonomic boundaries superimposed. the large taxonomic regions would thus pop out. problems: Eupedia 20:46, 27 February 2006 (UTC)

[edit] problem: choice of area metric

such a map requires a decision on which scalar quantity would map to surface area on the food map. kilograms/year consumed per year by humans? humans and domestic animals? economic value? area cultivated? etc. Eupedia 20:46, 27 February 2006 (UTC)

Why not multiple maps?--Curtis Clark 05:25, 28 February 2006 (UTC)
Go for it! BTW, thanks for all the awesome cleanup. Eupedia 06:55, 14 April 2006 (UTC)

[edit] problem: space-filling geometry

this is math problem: how do you project a "weighted dendrogram" (i made up this term by analogy to graph theory, but there might be a better term already existing -- anyone?) onto a contiguous plane? that is, assume by analogy that each food is like a country, each with it's size determined by the problematic metric discussed above. then, all of these food-countries would need to be packed together into a map of the food-world, with no white space (all land belonging to one country or another). easy enough so far, since the borders can be made to have any shape, but here is the clincher, and why it is a math problem: does there exist, for any weighted dendrogram, an area-mapped plane packing that preserves the connectivity of the dendrogram? Eupedia 20:46, 27 February 2006 (UTC)

It seems to me that the nature of a map implies a network of connectivity rather than a tree. But then I'm a biologist, not a topologist. I wonder instead about using a cladogram, and weighting the area of the branches by the measures you mentioned. This would be somewhat analogous to the sensory homunculus of neurobiology.--Curtis Clark 05:34, 28 February 2006 (UTC)
LOVED the sensory homunculus. i think a cladogram is a subset of dendrogram (tho i could be wrong), but your point is well taken: cladograms from genetic barcoding are an example of such a weighting, tho they do it with length, while i was envisioning area. hmmm. Eupedia 06:59, 14 April 2006 (UTC)
sweet! i just figured it out. there is another way to showing a branched hierarchy that i have seen in a piece of software called diskvision. you can download it for free and try it out. it shows the space taken up by various files/folders on your hard drive by area. it's awesome. the link shows a picture of how it looks. i know the guy because i suggested a mod to his program (to show number of files), which he implemented. so "all" that's needed now is a dataset. where can i get a list that shows class, order, family, genus, species, and variety for all important foods? Eupedia 07:14, 14 April 2006 (UTC)

[edit] Alphabetize the table by common name

I think the table should be alphabetized, and my first thought was by scientific name, but then I wondered if common name might not be more appropriate. Thoughts?--Curtis Clark 03:35, 25 February 2006 (UTC)

definitely not, and here's why: the reason i made this table was to show the intensely close grouping of many foods into few taxa. this is shown by sorting taxonomically, but would be lost sorting on common name. now i could see an argument for moving the common name to the right-most row, if someone wanted to do that... Eupedia 20:29, 27 February 2006 (UTC)
But that is part of my point: the important food species seem to be a few isolated dots and clumps in a vast sea of non-edible or non-interesting species. Within Brassicaceae, for example, most of them seem to be in one single genus, Brassica. So, by listing only the edible species, the table gives a somewhat distorted view of the situation. Here is a question one may ask: is the number of food species in a family a more or less fixed fraction of the total number of species in that family? If not, is Brassicaceae "food-rich" or "food poor", in relative terms? Jorge Stolfi 05:05, 28 February 2006 (UTC)
Interesting viewpoint, but it falls apart statistically in that the decision, even in a phylogenetic system, of what one calls a family is at base arbitrary, so it's not possible to make a convincing null model to test against. If one chose a family equivalent to the accepted subfamily Brassiceae, it would be food-rich. On the other hand, the broad interpretation of lumping Brassicaceae, Cleomaceae, and Capparaceae in a single family would render it food-poor.--Curtis Clark 05:40, 28 February 2006 (UTC)

[edit] Merge into Brassicaceae

Apparently the only reason to group these vegetables is that they all belong to the Brassicaceae family. If that is true, then the table should be moved to a section of Brassicaceae. Health information should be given in the articles of individual species, and therefore discussion of that aspect at the family level both superfluous and probably impractical. Jorge Stolfi 00:31, 26 February 2006 (UTC)

[edit] better, Merge into Brassica

On second thoughts, only a few of the "Cruciferous vegetables" lie outside the Brassica genus. So it would make more sense to drop those few exceptions and move the table to the Brassica article. Jorge Stolfi 01:02, 26 February 2006 (UTC)

Somewhat oppose:
  1. Only eight species of Brassica are represented in the table, versus six non-Brassica species. This is in no sense a Brassica article; in fact one of its values is to point out the cruciferous vegetables that are not Brassica.
But Brassica accounts for 26 entries; and the other 7 are quite different plants, foodwise speaking, so I fail to see the point of discussing them in the same article. If you include watercress, why not not include lettuce and asparagus, too? Foodwise, they are more "Brassica-like" than watercress! Said another way: as a food or agricultural science article, that set of species does not make sense; as an article about the family, there are no nutritional or agricultural attributes that are shared by all species, so the article will never be more than just a list of 35 "random" edible plants. Jorge Stolfi 09:29, 26 February 2006 (UTC)
I think it's ironic that you inflate the Brassica total by including all the cultivars, which are important only as food, and then deprecate the importance of food.
Certain plant families are heavily used as food:
  • Poaceae: not only all the grains, but also bamboo shoots, baby corn, lemongrass, and sugarcane.
  • Fabaceae: The seeds of many species are eaten (beans, peas, peanuts). Are peanuts more like filberts than they are like peas?
  • Rosaceae: Are strawberries, raspberries, plums, and apples all that similar except for being rosaceous?
  • Asteraceae: You mentioned lettuce. Include sunflower "seeds" and artichoke hearts, and you already have a varied group.
In each case, all the foods belong to a single plant family. In no sense is that random. And this is an encyclopedia.--Curtis Clark 15:22, 26 February 2006 (UTC)
Sorry, I am obviously unable to express myself clearly. No one is denying that the other genera of the Brassicaceae include many plants that are important "foodwise". The point is that it doesn't make sense to discuss "all the edible plants of family X" in one article, because (a) there is practically no food- or agriculture-related attribute that holds for all those plants, so one would end up discussing each plant or each genus separately (which is best done on their respective pages) and (b) the really important food/cutlivation groupings usually cut across many families. So, for example, it would make a lot of sense to have an article on "grain crop species" or "forage crop species" or "edible species adapted to dry conditions" or "edible species with high vitamin C content" — irrespective of family boundaries. It would make sense also to discuss "genetics of the Brassicaceae", "metabolic processes of the Brassicaceae", "cellular anatomy of the Brassicaceae", "breeding and hybridization strategies for the Brassicaceae", etc. On the other hand, having an article on "edible Brassicaceae species" is no different than "edible species whose names use the letter N", or "edible species with pink flowers". That is: for the purposes of discussing nutritonal and cultivation properties, these three sets are equally arbitrary and hence uninteresting. Makes sense? Jorge Stolfi 16:17, 26 February 2006 (UTC)
I understand your point more clearly now, and I disagree; it's emphatically different from "letter N", since plant families are (or at least plant taxonomists intend that they be) natural groups. It can be inferred that all species of the Brassicaceae are descendents of a common ancestor not shared with species outside the family. Food items in the Brassicaceae have a basic similarity in that they are all members of that natural group – they all have the same genetic heritage. This naturally influences their cellular structure, growth forms, ability to be selectively bred, biogeography, and phytochemistry, all of which influence their use as food plants. This is hardly arbitrary or uninteresting.
I would expect so myself, but unfortunately Mother Nature had other plans. If you look at any list of species of a clade (including Brassicaceae), you will see that even within the same genus -- no, even within the same species! -- one can find plants that are as different from each other as day from night, food- and culitvation-wise. One cultivar has big edible tubers, the next one has no tubers but lots of seeds good for oil, the next one is highly toxic, the next one lives in arid climates, the other is semi-aquatic, etc.. On the other hand, you find many plants from very different families that produce red sweet berry-like things to attract birds, and hence are interesting for jam or ice-ceam makers. The fact is that *some* very basic things *do* follow taxonomic lines, but "end features" like edibility or growing requirements do not. Jorge Stolfi 17:09, 26 February 2006 (UTC)
And yet the above-ground parts of few (none?) Solanaceae are edible, but those of most Brassicaceae are (although many are not palatable). One of the things that makes evolutionary biology continually interesting is the contrast between history and adaptation.--Curtis Clark 17:34, 26 February 2006 (UTC)
Off-topic, but for what it's worth, the list of plants with edible leaves has Capsicum annuum, Capsicum frutescens, Solanum aethiopicum, Solanum americanum, Solanum incanum, Solanum indicum, Solanum macrocarpon, Solanum melongena, Solanum nigrum, Solanum nodiflorum, Solanum scabrum, Solanum sessiliflorum, Solanum spirale, Solanum torvum, Solanum uporo, Solanum wendlandii, Solanum xanthocarpum, Nicotiana tabacum, and Physalis angulata. Leafeater 21:22, 23 June 2006 (UTC)
Agreed, those general properties are worth mentioning in the family articles. (I suppose you mean stems and leaves, since many Solanaceae have edible fruits, e.g. tomatoes and wolf apples.) BTW, are the toxic substanes of the Solanaceae chemically related?
It's been a while since I've looked into this, but I seem to remember one or a small number of classes of alkaloids being the main poisons.--Curtis Clark 15:50, 27 February 2006 (UTC)
None of this is to say that foods couldn't or shouldn't be arranged in other ways as well (and in fact they are). But arranging food by plant family is a common sorting device. Ethnobotanical lists and botanical works on herbs and spices (among others) are characteristically arranged by family. When I teach plant families, I ordinarily refer to the economically important plants (when they exist) to give the students a sense of the family from plants they may have already encountered. And even technical floras often list economically important members of each family.--Curtis Clark 16:38, 26 February 2006 (UTC)
Last year I noticed that many clade articles listed all the species under that clade. So the family article would have a big tree, down to species level; each subfamily article would have a subtree of that tree, again down to species; and so on. That seems nice for the reader, but when you need to change one species, you have to edit a dozen articles insetad of one. It seesm that people are now trimming down those trees, so that the article of each clade only lists the nodes at the next level, not its entire subtree.
After many such experiences in various contexts, I have learned the following "rule of thumb": if one lists the same imformation N times in a database, updates will cost N^2 times as much, and there will be N^2 more inconsistencies among all those copies. In other words, replicating information forthe sake of the user is counter-productive: instead, one should put each information in one place only -- and give the user a good search tool.
This is a very good point, and I've been bothered by the same thing. Search, though, doesn't solve everything; when I was a child, I would sit down and read volumes of an encyclopedia, seeing connections that a search would never provide.--Curtis Clark 17:34, 26 February 2006 (UTC)
In this case, that rule says that one should discuss the food properties of individual cultivars in their respective articles, and not even mention them in the higher clades (unless something can be said of the clade itself. Jorge Stolfi 17:09, 26 February 2006 (UTC)
And yet there is a case to be made for being easily able to look up the food articles for a single family. How about a general treatise in the family discussion (___aceae contains many edible members because of blah blah blah) and a category (say, "Edible members of the Brassicaceae") to accommodate the efforts that people such as Eupedia go to in assembling all the references.--Curtis Clark 17:34, 26 February 2006 (UTC)
Thank you for recognizing my contribution! It took a lot of learning for me to get the table right, etc. etc. (i have now applied/expanded this to dental notation of all things, to very nice effect, IMO). but to the question at hand, if i may weigh in: i started this list as an excel spreadsheet a year ago when i noticed how many important foods seemed to be so densely arranged phylogenetically, once i completed it, i was blown away! in this spirit, trimming it to brassica only would leave the most dramatic effect, even at risk of losing completeness, since, the ratio foods/taxon (assuming taxon is defined recursively) is certainly higher for brassica than for brassicaceae. furthermore, who's to say i shouldn't have gone up to brassicales? then again, if there is a convention already for grouping families, then perhaps we should stick with it. my main message here: despite being the author, i am flexible, and see it both ways, and will not be offended by or attached to either implementation. Eupedia 20:29, 27 February 2006 (UTC)
  1. The "Use as food" section of Brassicaceae is inadequate. Perhaps merging this in to replace that section would be useful, but calling this "Main article" in that section (per Wikipedia practice) would also be useful. If this were merged, the Brassicaceae article would be mainly food and a genus list, which is not necessarily a bad thing.--Curtis Clark 05:21, 26 February 2006 (UTC)
I would remove that section from Brassicaceae altogether, or reduce it to one paragraph like "this family contains the important food genera Brassica (including cabbage, cauliflower, canola, and over 20 other common vegetables and oil crops), Nasturtium (watercress), ..." Jorge Stolfi 09:29, 26 February 2006 (UTC)
Or you could look at all the plant families with edible members and bring those sections up to a common standard.--Curtis Clark 15:22, 26 February 2006 (UTC)
Retrieved from "http://en.wikipedia.org../../../c/r/u/Talk%7ECruciferous_vegetables_b75f.html"