I’m too unfamiliar with how google handles these things to really comment on this, I just want to say, that it would still be a good solution for the short term, in any case. Also, I assume that a PDF that has at least some downloads will still be available in the way current hard-to-get keys are available (by messaging people who have them). Of course, that’s not ideal, but it’s better than nothing at least.
Also, while I agree that this kind of information should be as widely available as possible, it is kind of fun to collect good rare and obscure books, papers, etc. (whether digital or physical) :P
I have edited a lot on Wikipedia (>100k edits over several years) and can confirm that it is against the editing guidelines to add guidebook style identification information such as ID tips to Wikipedia pages. The WP:NOTGUIDE policy (here) deters the addition of anything that resembles a guidebook or identification key for articles. Additional project guidelines (birds, fish, plants, etc) have examples of species article layouts that encourage the addition of morphological descriptions that do not reach the level of a guidebook.
ID tips and links on BugGuide are great, but do not extend beyond arthropods and are limited only to North America.
The “best” location for ID tips on iNat has been discussed a lot in the past and many forum users would like to see an editable identification guide near the “Similar Species” and “About” tabs for each individual taxon page.
Yeah, that is seen to cross the line between a description and an ID tip/manual/guide. There are also many ID tips that are only regionally specific/useful, that would not be pertinent to a Wikipedia article.
OK, so we obviously need to make a wiki then separate from wikipedia if that’s true. I would appreciate clarification as to why there appear to be some disagreements above with that.
I’m a little confused whether or not Wikipedia is adequate based on the feedback so far here, but leaning that it’s not. If so, does anyone else want to try something new on our own? I don’t understand why we’d wait for iNaturalist to do it themselves, or why it’s important for it to be part of iNaturalist. (If it works well it could eventually be added to about pages as the main article, just like wikipedia already is, for example)
In my writing of and reviewing others’ hundreds of thousands of changes to taxon articles on Wikipedia, no one has ever deleted appropriately sourced content that is similarly stated to what you wrote. It’s not a “how-to”, as it shouldn’t be phrased like “If you need help identifying the species, look for the horn…”
I got that impression from the policy as well; that you can just change the phrasing in many cases:
Describing to the reader how people or things use or do something is encyclopedic; instructing the reader in the imperative mood about how to use or do something is not.[4] Wording can easily be modified to avoid advising the reader: Do not give aspirin ... ⇒ The WHO advises against the use of aspirin
To re-use an example, you could give a description like this which doesn’t mention any other species or advise how to distinguish them:
That successfully eliminates all the other similar Melanerpes species, but it becomes sort of a logic puzzle for the reader to figure out which individual feature doesn’t match if they’re looking at an observation of one of those species. It’s much more intuitive to add another paragraph:
They differ from other species in that Red-crowned Woodpecker is very similar but is only found between Costa Rica and Venezuela, Gila Woodpecker has no red on the nape… etc.
But that might be over the line given that the Wikipedia articles for those species don’t include anything like that? When you get into more intricate ID challenges with multiple variables to compare, those comparisons are almost essential.
iNat curator and long-time (but somewhat lapsed) Wikipedia contributor here.
My view is that the ideal tool we’re all looking for (and may need to build) is some type of community-edited multi-access key framework, but that doesn’t mean that we should not also add properly sourced identification information to Wikipedia articles.
Here’s an article I wrote on Sisyrinchium albilapidense, a narrowly distributed Iridaceae species from Brazil. The Taxonomy section has a couple of short paragraphs on how to distinguish the species from three related species. That’s all fine, but it’s still suboptimal on several counts:
It requires quite a lot of effort per species
It’s a lot harder to use than a well-designed key
It doesn’t include any photos or audio (and doing so would require even more effort)
Wikipedia’s source-citing requirements may be a little too onerous for our purposes
Here is what I’d love iNat (or the iNat community) to develop
The ability to “chain” keys, to lead users from family to genus to species
Some ability to create a user-friendly narrative and UI around the keys, with the goal of the site being usable by novice naturalists, just like iNat.
Ability to link to external ID resources (field guides, other keys) with context.
Supports images (to show diagnostic characters) and audio files (e.g. for birds, bats, stridulating insects)
“Community-owned” in much the same way as WP and iNat, so contributors agree to license their contributions under a Creative Commons license.
Appropriate policies on sourcing and review. These might be a blend of WP and iNat policies.
On that last point, I have in mind something like this:
I’d suggest that all contributions need to be sourced to published data, but with an emphasis on productivity and flexibility. So you add a published paper to the internal database, and then cite it just by choosing it from your “favorites” as many times as you need to use it.
Some type of review for new and changed contributions, to minimize vandals and spammers, possibly with “sampling” of contributions for more experienced contributors.
For peoples’ observations, sometimes I will explain my ID choice, similar species, and common mushroom IDing practices - especially if I see that they are relatively new to iNat, or are new to citizen mycology or observing fungi in general. It is sometimes difficult to summarize the nuances of identification so that it fits into cleanly into non-multi-paragraph comment :)
I really appreciate this response–the clarification of the limits of Wikipedia is great, and it also illustrates that the difference between Wikipedia and our ideal “inat wiki” isn’t necessarily the kind of information that is allowed but the way the information is organized and accessed.
I think it would be better to create a new wiki because then we could make it more specific to this particular case. Wikipedia has pages for almost everything so it can’t really have specific features, this will also hopefully invite some consistency. For example, Verditer Flycatcher (a well-known bird that I’m familiar with) has no habitat on Wikipedia, even though it does have one on eBird.
Also, I’m pretty sure the UX-UI can be better than Wikipedia.
BugGuide is limited to a small taxonomic group and to only 2 countries. This doesn’t cover the diversity, and also it is hard to contribute to BugGuide because of restrictions (you need to be an editor!)
We definitely need something that covers the bigger scope.
It’s definitely an option, but a bad one compared to what could be done within iNat itself.
Then, Wikipedia is also tedious to contribute to. Here are some points against it:
Requirements to write articles in broad terms. That means we’re gonna have a hard time writing about specifics, and in the style which must be essentially different to cover the needs as an “Species ID Wiki”
Because of the requirements, it may be tedious to write the article in general. Style, specific formatting, - it could be made simpler for an iNat wiki.
Pictures must be a certain license. CC-BY-NC and CC BY-ND are basically forbidden. On the other hand, iNat could basically be the place where in theory you might grab someone’s observation photo (even if “All rights reserved” - very common thing even for observation data; this sucks, I think the issue might be bad onboarding for new users) and place it directly in your wiki page, since it’s within the same site.
Integration. iNat has way more possibilities for making something fit well within the system if they implement their own wiki, rather than relying on Wikipedia. One example would be - we already have a taxonomy here, this doesn’t have to be recreated somewhere else. Then the wiki can even be used to train the genAI thing if they really want… given it will be a tool helpful in the same way as Computer Vision.
Additional point might be dependency issue. iNat is trying to depend on Wikipedia but it’s purpose also differs from what iNat identifiers/users need. Having the built-in wiki will allow for more flexibility…
Please see this feature request about “editable identification guide”
There were other ones but it’s the most popular one.
We’re currently at 102 votes - it would be great if anyone could also place a vote!
The idea is to have that identification guide / wiki in the ‘About’ page (or maybe in ‘Similar species’ as suggested).
I think wikipedia uses open access software that we could copy? Do we have enough tech-savvy iNat volunteers that we could find a group qualified to set it up?
Wikipedia uses MediaWiki (free software), basically the backbone behind it.
Have people tried doing this in the past?
Apparently yes, with the “Species-ID”. They seem to use the same technology, which is a variation of MediaWiki, you may see it here at the bottom (“Powered by MediaWiki”, etc.). However, the project seems to have simply died, not clear why. https://web.archive.org/web/20160527031232/http://species-id.net/wiki/
How hard would this be to make and what are the obstacles?
I’m pretty sure there is a tutorial on how to set it up, so given enough time it may be straightforward.
In terms of obstacles:
Need to consider the possible cost of hosting the site
May need to learn how to use such software and set it up further
Need to consider how extensible the wiki should be (are there specific tools within MediaWiki?) - although if we simply need to host separate articles, it should do…
It would be great or much better to have this integrated into iNat itself, otherwise there will be less people contributing an even knowing about such resource. Also iNat having custom software can make it fit better within the system and for the specific need of “Identification guide / wiki”
I have a lasting impression that iNat is extremely slow on implementing what users ask for, some requests even get dismissed for years.
Even something extremely important like this:
(reveal, slightly off-topic)
I would say instead “removed ASAP”, until they get to implement it the correct way…
Staff said: “Anonymization option(s) is something we want to do, but as to when it would potentially be available I don’t know.”
But:
Some features like this must have critical priority over the whatever else non-essential feature that “may” help users, if it works out at all… (e.g. the genAI idea, it’s just a test demo)
I think what would help greatly at least is having a public feature roadmap, it will be much clearer if they are busy with some priority feature but still plan to implement X, let’s say 1 year from now on.
There have been features that were looked at by staff, but then no clear response followed.
Of course wiki itself doesn’t have to be a part of iNat, but then this will probably have much less support from community, will be harder to maintain and integrate within iNat (if they would even consider adding this community site to “About” pages)
I don’t feel it should be necessary for all identification tips to have a published citation attached. This would exclude all identification tips sitting in the heads of expert iNaturalist users. I can see that a citation gives an element of reliability, but published keys have plenty of errors or couplets which are at best unreliable. If tips were to require a citation, many skilled identifiers won’t get around to writing a published note on everything they know, and when they do it could be a year or more before the article is published.
The trouble with giving id tips under an individual species, as suggested above, is you generally need tips to separate two species. Saying how to identify species A requires you to guess all the other species someone could confuse it with. Better to invite tips on how to separate A from B. Then there is the question of how to index the tips.
For identification tips, you don’t necessarily need a key to a whole genus, just to two or three difficult species. And the difficult species may not be in the same genus. E.g., the fly you are most likely to confuse with Phaonia pallida in Britain is Thricops diaphanus.
There will remain the issues that tips may simply be wrong, and that they may only work in certain parts of the world. So I suggest a tip needs to have a field describing the geographical area in which it is valid; and there should be a voting system rather like the current RG system which allows iNatters to agree or down-vote it, preferably backed up by comments.