conservation status, obscured coordinates, ranges

[Ken-ichi]

Hey all,

You may have noticed, but we've started keeping track of conservation
status for taxa in iNat. Right now we only have data from the IUCN
Red List, which includes most non-fish vertebrates, but we're hoping
to extend to other authorities. We've also begun obscuring the
coordinates of observations of threatened taxa, so if you observe
something that's between Near Threatened and Critically Endangered,
iNat will do a couple things to hide the location:

1) choose a random point within 5 KM of the true coordinates to show
on public maps
2) show a circular marker on maps instead of one with a stem
3) stop showing the lat/lon publicly

You should still be able to view the true coordinates when viewing
your own data, but in public places like the home page, project pages,
or /observations, they should be obscured. Here are some examples:

http://www.inaturalist.org/taxa/California_Condor
http://www.inaturalist.org/taxa/67016

Also related to munging data from IUCN, Scott has been hard at work
getting our amphibian taxonomy in order and importing range data, so
almost every amphibian in iNat now has a range map (click the Range
Map tab next to the description). He's also got all the mammals in
there, and other vertebrates are in the works. There's not as much
data out there for other taxa, but we're hoping to get some very rough
data on California plants soon.

-ken-ichi

[Winged Wolf]

Would you consider making this option available for people to select
voluntarily, for specific classes of animals? I know that in the
field herping community, a big problem with listing ANY observations
is the risk that a collector will come to the spot where you saw
something, and legally remove the animal. That must also be a risk
with fish. Being able to keep the data private, while still having it
in the system (and presumably available for scientific research) would
be tremendously beneficial in getting new members from the herping
community. At present, there is no good tracking site for herps in
the US, and range data is sorely needed.

--
Donna
http://www.EclipseExotics.com
EclipseExotics on Twitter

[Eric Hunt]

Agreed. Native flower bulbs are also highly poached - with organized teams from various countries around the world trolling Flickr and herbarium sites for locality info then coming to California (or wherever) to dig them up. They're not CITES I or II listed so they can cross international borders freely.

-Eric

> --
> You received this message because you are subscribed to the Google Groups "iNaturalist" group.
> To post to this group, send email to inatu...@googlegroups.com.
> To unsubscribe from this group, send email to inaturalist...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/inaturalist?hl=en.
>

[Scott Loarie]

I agree that giving users the option to obscure the location of one of their observations regardless of the taxa associated with it would be hugely useful. One example would be data from camera traps where you don't want people to come steal your camera or if you had a certain spot you want to keep private you could obscure the obs you get from there. This also fits with the iNat philosophy of letting users control their observations. But I think its important that we have a third party (e.g. the RedList) and not individual users setting which taxa are obscured in general. Otherwise it could get pretty arbitrary.

-Scott

--
You received this message because you are subscribed to the Google Groups "iNaturalist" group.
To post to this group, send email to inatu...@googlegroups.com.
To unsubscribe from this group, send email to inaturalist...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/inaturalist?hl=en.

--
--------------------------------------------------
Scott R. Loarie
Post Doctoral Fellow
Department of Global Ecology
Carnegie Institution for Science
260 Panama Street
Stanford, CA 94305 USA
www.stanford.edu/~loarie
Email: loa...@stanford.edu
Office: 650-232-6913
Cel: 415-278-1220
--------------------------------------------------

[Winged Wolf]

I had envisioned letting users obscure the data that they enter,
without affecting the data that others entered, but I'm not sure of
the technical problems that might cause for the programmer. It will
still be incredibly useful. I do think that if that isn't possible,
it should be considered whether taxa vulnerable to legal collectors or
poachers should be obscured. I know that I, and many other herpers,
would hesitate to put location info into the system at all if there
weren't a way to obscure it, particularly for species that aren't as
well protected as they should be in all areas (such as box turtles).

If it IS technically feasible, then a simple check button labeled
'obscure data' could be added to the observation form, allowing people
to choose to obscure data for any individual sighting they feel should
be protected.

[John Cramer]

It's already possible for any user to obscure the location of any observation.

To do so just enter a "vague" location like the name of the forrest or the zip code and hit return.

[Eric Hunt]

John,

Yes, and that's what I do with some of my bulb observations but unless you keep a separate, private database of locality info, then you've lost data.

I'm still undecided. In the end any data one puts into the cloud can be discovered by people you don't want to discover it. Websites grow up, are sold, and the new owners don't have to have the same values as the creators, and boom your previously obfuscated GPS coordinates are now public and/or the dataset is sold with the coordinates in cleartext.

-Eric

[Scott Loarie]

Good point - but it would be nice to allow the actual coordinates to be recorded privately so they can be used for science etc. I liked WW's suggestion of the checkbox. Of course it begs the next question of who's 'qualified' to access the raw data? I imagine admins/curators of a project that the user agrees to submit data to, assuming projects stay reputable and we don't see any 'where to poach critters' projects popping up!

[Mark AT]

I must say I was hugely disappointed when I discovered you were
offering pinpoint location information on Google maps. Not only is
this a collectors dream but it undoes the efforts many amphibian
hobbyists (like myself) have put in to keeping specific location
information off the internet. I welcome the idea of collecting this
useful data for scientific purposes but to make it publicly available
is in my mind irresponsible. You must realise that even for common
species there is a collection threat, so please, lets not make it easy
for them.

At the very most location information should be restricted to county
or state – no ability to drill down further should be allowed.

- Mark

On May 19, 7:49 pm, Ken-ichi <kenichi.u...@gmail.com> wrote:
> Hey all,
>
> You may have noticed, but we've started keeping track of conservation
> status for taxa in iNat.  Right now we only have data from the IUCN
> Red List, which includes most non-fish vertebrates, but we're hoping
> to extend to other authorities.  We've also begun obscuring the
> coordinates of observations of threatened taxa, so if you observe
> something that's between Near Threatened and Critically Endangered,
> iNat will do a couple things to hide the location:
>
> 1) choose a random point within 5 KM of the true coordinates to show
> on public maps
> 2) show a circular marker on maps instead of one with a stem
> 3) stop showing the lat/lon publicly
>
> You should still be able to view the true coordinates when viewing
> your own data, but in public places like the home page, project pages,
> or /observations, they should be obscured.  Here are some examples:
>

> http://www.inaturalist.org/taxa/California_Condorhttp://www.inaturalist.org/taxa/67016

[Eric Hunt]

Mark,

I'm genuinely curious - is there a coordinated effort among research institutions to obfuscate this data when they move entire natural history collections/databases onto the internet? For plants it's a lost cause - more and more herbariums are digitizing their entire collections and posting them online, making it very easy for bulb collectors to find new populations to poach.

-Eric

[Scott Loarie]

Hi Mark,

We share your concerns about collecting. You'll find the locations of all threatened species are obscured. Regarding common species, I agree that there are cost benefits with making information public. Risk of collection must be balanced with the benefits for raising awareness and gathering data for science. Also there are many other examples of amphibian museum locations widely available on the internet through portals like GBIF (click 'show museums records' here: http://www.inaturalist.org/taxa/25591#taxon_range ) and on  photosharing sites like Facebook, Flickr, and Picassa that enable geo-tagging. Like these services, iNaturalist allows you to not include coordinates if you'd like or rather to just tag your observatons with a place name such as a city or state.

Thanks,

Scott

--
You received this message because you are subscribed to the Google Groups "iNaturalist" group.
To post to this group, send email to inatu...@googlegroups.com.
To unsubscribe from this group, send email to inaturalist...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/inaturalist?hl=en.

[Winged Wolf]

It sounds like you're not planning to offer that feature, then.  (This is too bad, as specific location information could be very valuable to scientists, but if it will fall into the hands of field collectors, recording it would be too risky). 

I've been gauging the location and interest level when I record herps.  If it's in an area where collecting is legal, and is a species desirable to collectors, I wouldn't list the location--if it's a species collectors don't bother with, or it's in an area where collecting isn't legal (such as Fontenelle Forest), then I go ahead and put the location down.  It's true there is some risk of poaching, but that's a risk everywhere, and usually wildlife preserves will tell you what species you can find in them anyhow.

Winged Wolf
Twitter:  WingedWolfPsion
My Stores:
http://www.EclipseMetaphysical.com
EclipseMeta on Twitter

[Ken-ichi]

Mark and Eirc (and anyone else concerned about this issue): do you
feel that 5 KM of uncertainty isn't enough? Mark, I'm not sure if you
noticed but we actually don't display exact coordinates for taxa that
we know are threatened according to the IUCN Red List. Instead, we
choose a random point within 5 KM of the true coordinates and display
that. 5 KM was a number I chose somewhat arbitrarily, but I suspect
I'm not as familiar with collecting/poaching pressure as you guys are,
so I would appreciate some input.

Regarding the feasibility of allowing users the ability to obscure or
hide the coordinates of any of their observations, it's totally
feasible and I want to build it, I just haven't gotten to it yet.

Also, I think Eric expressed some healthy skepticism above that we
should all keep in mind when sharing data on the Internet: sharing
data means losing at least some control over it. Even if you email
coordinates to a friend, you don't really know what that friend will
do with the data. Obviously we at iNat want you to share your data
with us and the world because we believe the benefits of doing so
(spreading awareness of biodiversity, helping scientists and resource
managers, etc) outweigh the potential costs (poaching,
over-exploitation), and we try to minimize the costs through practices
like obscuring coordinates. However, once the data leave your
computer, all you can really do is trust the people you send it to use
it responsibly. We hope you will trust us, and if you find reason not
to, we hope you will tell us about it so we can find ways to address
the problem. But you should always think about whether you trust the
parties with whom you're sharing data, and if you don't trust them,
don't share.

-ken-ichi

[Winged Wolf]

On Wed, May 25, 2011 at 12:08 PM, Ken-ichi <kenich...@gmail.com> wrote:
>
> Mark and Eirc (and anyone else concerned about this issue): do you
> feel that 5 KM of uncertainty isn't enough?  Mark, I'm not sure if you
> noticed but we actually don't display exact coordinates for taxa that
> we know are threatened according to the IUCN Red List.  Instead, we
> choose a random point within 5 KM of the true coordinates and display
> that.  5 KM was a number I chose somewhat arbitrarily, but I suspect
> I'm not as familiar with collecting/poaching pressure as you guys are,
> so I would appreciate some input.

No, I think that's perfect.  It's the taxa that aren't considered to
be threatened, and that can be legally collected, that I was more
concerned about.

>
> Regarding the feasibility of allowing users the ability to obscure or
> hide the coordinates of any of their observations, it's totally
> feasible and I want to build it, I just haven't gotten to it yet.

That's fantastic, and I definitely look forward to it!

[Eric Hunt]

Ken,

I think that 5km is fine. What I don't understand is the distinction between a red-listed species and a common species. Maybe you've never been around a collector - it's almost like a mental illness - nothing will stop someone who wants to collect *ALL* of a certain thing. It's a compulsion.

Bulb and orchid poaching is truly staggering. I regularly visit populations of plants one year and visit them again in future years only to find gaping holes in the ground.

-Eric

[Eric Hunt]

Scott,

You make the counter-point that I've made. The cat is out of the bag for a lot of this stuff. Herbariums, other hobbyists, etc., are all posting stuff online without obscuring it.

If the purpose of iNaturalist is to enable the public to make observations of the world around them, I think giving people the option to obscure coordinates is in keeping with that purpose. Otherwise it seems you would have to seriously think about some of the previous discussions here - data integrity, not allowing people to record 'bogus' observations, etc.

-Eric

[Eric Hunt]

> Regarding the feasibility of allowing users the ability to obscure or
> hide the coordinates of any of their observations, it's totally
> feasible and I want to build it, I just haven't gotten to it yet

Ahh - I sped-read right over this. Cool!

-Eric

[Julie Ray]

In some manner locations should be obscured or species that are known
to be rare or highly collected must be passed by an administer before
they are posted. It is a great idea and learning tool to make data
available on species like amphibians and it puts a lot more people out
there documentng species. This is important as money and time are
limited, especially when species are declining as amphibians are.
However, it is too easy for collectors an poachers to narrow down a
site. Furthermore, photos should be taken so they do not announce the
exact site (i.e. in front of a sign or obvious landmark).
It is important to maintain populations in the wild, yet allow people
to see and enjoy them, plus study them. In some manner locations
should be obscured as the amphibians are the most important here.
Thanks!
Julie Ray
La MICA Biological Station
El Cope, Cocle, Republic of Panama

On 19 mayo, 13:49, Ken-ichi <kenichi.u...@gmail.com> wrote:
> Hey all,
>
> You may have noticed, but we've started keeping track of conservation
> status for taxa in iNat.  Right now we only have data from the IUCN
> Red List, which includes most non-fish vertebrates, but we're hoping
> to extend to other authorities.  We've also begun obscuring the
> coordinates of observations of threatened taxa, so if you observe
> something that's between Near Threatened and Critically Endangered,
> iNat will do a couple things to hide the location:
>
> 1) choose a random point within 5 KM of the true coordinates to show
> on public maps
> 2) show a circular marker on maps instead of one with a stem
> 3) stop showing the lat/lon publicly
>
> You should still be able to view the true coordinates when viewing
> your own data, but in public places like the home page, project pages,
> or /observations, they should be obscured.  Here are some examples:
>

> http://www.inaturalist.org/taxa/California_Condorhttp://www.inaturalist.org/taxa/67016

[JP]

I would really prefer an option to obscure data. Personally, I work
on a rather common species of poison dart frog (the Strawberry Dart
Frog), which, according to the IUCN is a Least Concern species, so
points on where I found it would show up. The problem is that this
species is one of the most heavily imported dart frogs, so any locale
information that collectors can get is highly prized because new
populations offer great amounts of money to collectors.

One of my biggest complaints is that, because of this, scientists are
unwilling to share information on population locations, which results
in reinventing the wheel over and over. I would love a central
database that researchers could access so we know where and what is
known about the species, but this comes at a risk to the information
getting into the wrong hands. Having an option to obscure data would
give me a great deal of relief in submitting the data.

In regards to access of the data, I would suggest a series of checks.
People wanting access to data or information should be able to
demonstrate that they have a legitimate scientific interest in it
(ongoing project, grants, etc). Not only that, but should I obscure
data, people wanting access to my data should have to ask permission
to access my data after they have passed these checks. I think that
would prevent, as best as possible, the ability of the data getting
into the wrong hands.

-JP

> >http://www.inaturalist.org/taxa/25591#taxon_range) and on  photosharing

> > sites like Facebook, Flickr, and Picassa that enable geo-tagging. Like these
> > services, iNaturalist allows you to not include coordinates if you'd like or
> > rather to just tag your observatons with a place name such as a city or
> > state.
>
> > Thanks,
>
> > Scott
>

> > On Wed, May 25, 2011 at 9:05 AM, Mark AT <mark.aartset...@googlemail.com>

> >> >http://www.inaturalist.org/taxa/California_Condorhttp://www.inaturali...

[Adam R. B. Jack]

Hey Folks,

The approach taken at WildObs is that although as much exact location details as possible is recorded in the database (for research purposes) no exact location is published to anybody other than the original observer (and any research programs that get the data.) Instead the local city or (say) park name is shown. This is for all species. The hope for WildObs is to entice people to go out looking for critters, 'cos getting people out exploring in nature is good in many ways.  That, and knowing what species are "around them", for which exact location isn't important.

(FWIIW: A record can also be marked private, for personal/research consumption only. Further a species can be marked as 'not published' so no records show for it, although that data has not yet been populated.)

My question to this group is this ... is a park name good enough of a data fudge? Most parks are larger, but some might be less than 5K. Is the main goal not to give exactly lat/lng, so collectors know they'd have to do real work to hunt something down? Or, are there some cases where there are only so many habitats for a species within a park, and that information is too much to share? If the later, is there a way we could get (or build) some "risk of collection" information by species? That, or ought certain places be not listed?

Thanks for raising & contributing to this topic. The benefits of engaging people with nature do not outweigh the risks of collectors, but nature sure needs all the benefits we can give it.

regards,

Adam

--

Find your nature: http://wildobs.com

[Ted R. Kahn]

Hi Adam, we are in discussion. Close to getting our apples in a row per se. I think the park point is a good one, and I think we are leaning to a 10km range, but your suggestions are good too. Marking data as "research only" would be good imo, but the curators capacity has to be taken into consideration, and we want to engage the layman, whose data may be important and unknown to them how so. We already have an example of a rare sought after taxa where information was put out there, we caught it... All an all, lots to think about. And I am sure we will be posting here in the hours and days ahead!

Thanks,

Ted AKA Sapito

 

Ted R. Kahn, Executive Director

Neotropical Conservation Foundation

http://neotropicalconservation.wordpress.com/

6407 32nd Street NW

Washington, DC 20015-2305

U.S.A.

Tel: +202.374.3794

Fax: +202.363.5651

Skype: ted.r.kahn

Email: tedr...@gmail.com

 

Partner

Smithsonian Institution, National Museum of Natural History, Division of Amphibians and Reptiles (NMNH)

Smithsonian Institution, Laboratory of Analytical Biology (LAB, NMNH)

 

Senior Research Scientist

Latin American & Caribbean Social Science Research Network

 

Academic Member

International League of Conservation Writers

 

Member

Amphibian Specialist Group

Scientific Consultant

http://www.iucnredlist.org/amphibians

Neotropical Conservation Foundation supports the United Nations Decade on Biodiversity.

[Nick D Waters]

In my opinion, if people are going to obfuscate a locality then a reasonable, relatively non-arbitrary method is just to the level of County.

Giving a park name, to an unscrupulous collector, would be sufficient for them to target a search.

Warm Regards,
Nick

[Mark AT]

Yes, I believe there is a general effort amongst herpetological
discussion forums to avoid posting specific location information. As
an administrator of the largest and longest running amphibian forum on
the internet, caudata.org, I can confirm we’ve been regulating
location data for the past 10 years. I know many other large and well
respected forums take the same stance. I do not know if the same
stance is adopted by the scientific community.

I’m not sure which corner of the world you are in but here in Europe
many of our amphibian species are protected from collection for trade
(although the practice still persists in some countries). In the US,
for example, amphibians have less protection and a quick browse
through the US kingsnake.com classifieds will show many wild collected
US native animals being traded for good money. Where protection is low
the one thing that stands between a commercial collector and the
animals is the not knowing exactly where to collect from. I imagine
collection is an even bigger problem in developing countries where
rules are more lax i.e. China. There are many species which
fortunately do not make it into the pet trade, something in itself
which make them rare and valuable to collectors. I cannot see how
publishing location data can help these species.

Additionally, restricting the location information to common species
is not enough in my opinion. Some are on the red list simply because
of collection. For example Neurergus kaiseri , listed as CITES
appendix 1 only last year, collected to within the brink of
extinction. Just because an animal isn’t threatened at present doesn’t
mean it isn’t desirable to collectors.

I agree with JP. Dart frogs are a massive hobby and silly money is
paid for morphs from a specific location. Publishing location
information would almost be a death sentence in some cases.

- Mark

[Dr John Clare]

Dear Group,

I share Mark's concerns in this matter. I think iNat is a fabulous
program but the location aspect is not meant for the cold hard world
in which we love. As someone who spends a great deal of time chasing
amphibians in the wild for photography and personal interest, I know
that 5 km often isn't enough for most species. Many times I've been
able to find species with a little intellectual elbow grease, based
solely on vague location details like you propose. In several cases
in the past few years, I've located rare species based solely on
county with a little bit of extra info, such as habitat description or
a photo.

It is true to say that there are databases on the Internet. However
even they remove location data occasionally. For example, Camp et al.
discovered the first new genus of vertebrate in North America in
2007. The location details are not available through any publicly
accessible database, despite specimens being listed there. The
species in question has a very limited distribution and it would
certainly run the risk of extintion had they not taken these
measures.

Mark speaks from a position of experience in this area. He is quite
correct that there are many commercial harvesters out there who
regularly decimate our wildlife resources, particularly amphibians and
reptiles, all for a quick dollar in the pet trade. These people are
experienced field collectors, so they can accomplish their goals with
little location data.

I implore you to rethink your location approach. It is true that
there are databases with publicly accessible location information, and
in my opinion these are to blame for many local species declines due
to commercial harvesters (cf. the likely extirpation of Neurergus
kaiseri). However, we do not need to add to this travesty by giving
out new location data or restating the location data already available
if you know where and how to look. Please don't add fuel to the fire.

John P. Clare, Ph.D.
Founder of Caudata.org

[Nick D Waters]

Expanding the general topic, there is more to this matter that must be considered.  There are many issues which need to be addressed beyond what I've seen thus far.  Keep in mind that I think this site has a great deal of potential, otherwise I wouldn't have bothered to write this note.

I know you good folks do not know me.  I have close to 30 years of experience as a naturalist, field biologist.  I've worked for agencies and private industry, on my own in conjunction with others on a wide variety of topics, including ecology, malacology, and herpetology.  I have not published, but have written a lot of gray literature... but am sitting on loads of data which... I'm actively working on.  I was a programmer at a fortune 500 financial institution for several years, involved in information security before I got sick of commercial programming and returned to my true calling.

An idea which may work is suggested. Zero-sum use: Users provide some means of providing more secure information to establish identity, such as a fishing license, research collecting permit, credit card information or paypal account.  This will invariably drive some people away who may otherwise contribute, or perhaps the opposite.  If we establish a way to balance contributors of information from "takers" of information then there is incentive for use beyond altruism.  A penny in, a penny earned.  Make it a zero-sum situation for people.  Providing information earns money, and using information spends it.  Those viewing a lot of information, particularly those flagged as rare species, or those suddenly requested often, or those known to be trafficked in the pet trade, will provide trackable information as to the users identity and their intentions which can be investigated by law enforcement.  Those both viewing and providing information are more likely to be the target naturalist audience.  That said, no matter what security method you employ, fraudsters will always find a way to get around the checks and balances.

Perhaps ask Bruce Schneier, security expert extraordinaire.  He may have useful suggestions.  His basic stance is that security by obscurity does not work.  It is only natural to think along these lines. But hiding information is doomed to failure, and is what I see being proposed here.

The matter of sharing information can be extended further.  Providing data on species, however superficially innocent, will encourage many to visit reported sites to observe species on their "Life List".  This increases likelihood of introducing invasives, habitat destruction, disease, and good old fashioned depletion.  People will take specimens for personal interest.  People will distribute diseases and minute critters on their clothing and shoes. The only way to avoid problems like this is for all of us to stay in our homes and labs.  Not going to happen.

There is no simple answer to these problems. Many sites provide accurate locality data.  Calphotos, CalFlora, AmphibiaWeb.org, Flickr, Picasa, Eol, Caudata.org, FieldHerps, natural history museums, and so on regurgitate the full level of information provided by the user, either in comments or detailed database fields. Anyone can ask for information from the users even if it is not posted on the more casual sites. Any experienced programmer can write interfaces to cull this information into a private system, just like iNaturalist has done, and like I have done in the past (but iNaturalist is much prettier!)  To naturalists and "serious" researchers such as myself, what is the utility and incentive of a site such as iNaturalist if only portions of the data are obtainable. I can readily obtain records from any number of curators at institutions around the world directly online or by calling them.

How well funded is this effort?  I don't think the organization wishes to evolve into an agency, right?  Is it feasible to employ the typical approval process which grants levels of data provided based on trust or good old boy methods?  It requires money to hire customer service, a lengthy review process, letters of recommendation, on and on.  As more users sign on how do we determine someone is a bad guy?  One can impersonate other researchers, provide phone numbers and contact information to other bogus individuals to "vouch" for them.  Anyone motivated enough will find a way if the information is broadcast on the web without some means of knowing "who" they are. People talk and share information offline.

Nick Waters

P.S. Sorry for the long note. But I highly value this matter.

--

[Ken-ichi]

Thanks again to everyone who has commented. This is a great and
enlightening discussion, and we appreciated all the time everyone's
put into airing their opinions. A lot of interesting ideas have been
mentioned that I'd love to discuss, but for now let's just stick to
the visibility of spatial coordinates.

To address some of the concerns, we've just pushed out some changes to
how spatial coordinates are displayed on the site:

1) First of all, threatened taxa are now obscured by 10 KM instead of
5 (this applies to everyone's observations of these taxa). I know
showing *any* coordinates alarms some of you, so...

2) You can now manually control the "geoprivacy" of your observations.
If you set the geoprivacy to "obscured," the coordinates will be
obscured by 10KM as they are for threatened taxa, regardless of what
taxon you observed. If you set the geoprivacy to "private," the
coordinates won't show publicly and your observation won't show up on
any public map. For both "private" and "obscured" observations, the
true coordinates will still be visible to the curators of the projects
to which you add the observations, so you can still contribute
observations of sensitive taxa to common causes (but make sure you
trust the curators of the project).

As far as the taxon ranges on the taxon pages go, I don't see why we
should remove those data since we got them from the IUCN, where they
are publicly available for download.

Note that for non-threatened taxa, coordinates are still visible by
default, so it's up to you to change the geoprivacy if you think it
needs changing, or suggest that others do so. We're also thinking
about allowing site curators (people who have volunteered to help
manage our taxonomy) to mark taxa as "threatened" even if the IUCN
doesn't think they are in order to handle the many taxa the IUCN Red
List doesn't cover. If there's a taxon that isn't Red Listed *should*
have its coordinates obscured, please flag it to bring it to the
curators' attention.

Thanks again for all the great discussion. Please let us know if you
feel these changes address your concerns, or if there's more we can
do. Also please holler if you find bugs!

-ken-ichi

[Scott Loarie]

Hi all (sorry for posting this twice, but in retrospect I wanted it to be associated with this thread),

As Ken-ichi mentioned in his previous post, iNaturalist now has
functionality that enables you to keep the locations of your observations
private. However, you still have the option to make your data available for
scientific and conservation purposes by contributing them to a project that
you find to be reputable.

The Global Amphibian Blitz is an example research project which is a joint
effort by AmphibiaWeb at UC Berkeley, the Smithsonian Conservation Biology
Institute, the IUCN/SSC Amphibian Specialist Group, Amphibian Ark, and the
Center for Biological Diversity. The aim of the project is to census the
world's amphibians to gather data on where different taxa persist, explore
how the rate at which species are recorded correlates with other rarity
measures such as range size, and to develop a baseline census to compare
with future censuses to explore changes in amphibian rarity. The project is
run by a team of curators including myself. The terms and conditions of the
project posted at
http://www.inaturalist.org/project/global-amphibian-blitz/terms lists the
academic and conservation credentials and affiliations of all 10 curators.
The document also outlines our Data Use Policy for the Global Amphibian
Blitz. In brief, if a third party requests the data, the curators will first
assess the scientific and conservation merit of the proposal and whether the
third party will respect the sensitive nature of the data. The data will
only be shared if the curators reach a favorable consensus on both
questions.

We hope that this policy strikes the right balance between protecting
sensitive data from falling into the hands of those who would exploit rare
amphibians and making data available to scientists who will use it to
promote amphibian conservation. I sincerely hope you will join the Blitz and
contribute your data so that we may work together to find new strategies for
conserving imperiled amphibian species, but if you remain unconvinced, lets
continue this important discussion here.

Thanks very much,

Scott

[George]

This is a great discussion. It shows that iNat is evolving into an
excellent resource.. I agree that some species locations need to be
disguised. I've done that for the sensitive (not even endangered)
species I've submitted -- establishing the location in the middle of a
lake, say, in the basin it occurs. I would think an automated system
should do at least 10km and that in addition to the default for
identified endangered species, there be an optional one. Perhaps there
could be two entries: the actual coordinates that are not shown but
made part of the record and the public one, which shows the point on
the map.

This, of course, sets off a whole cascade of administrative
responsibilities and protocols in being able to release the accurate
information to legitimate researchers. But there are probably
University protocols established that could be adopted. Still, another
layer of complication for the people running iNat.

George

On May 19, 11:49 am, Ken-ichi <kenichi.u...@gmail.com> wrote:
> Hey all,
>
> You may have noticed, but we've started keeping track of conservation
> status for taxa in iNat.  Right now we only have data from the IUCN
> Red List, which includes most non-fish vertebrates, but we're hoping
> to extend to other authorities.  We've also begun obscuring the
> coordinates of observations of threatened taxa, so if you observe
> something that's between Near Threatened and Critically Endangered,
> iNat will do a couple things to hide the location:
>
> 1) choose a random point within 5 KM of the true coordinates to show
> on public maps
> 2) show a circular marker on maps instead of one with a stem
> 3) stop showing the lat/lon publicly
>
> You should still be able to view the true coordinates when viewing
> your own data, but in public places like the home page, project pages,
> or /observations, they should be obscured.  Here are some examples:
>

> http://www.inaturalist.org/taxa/California_Condorhttp://www.inaturalist.org/taxa/67016

[Killarney Park]

I have some questions and concerns with how species at risk are obscured.

1.  Would it be possible to obscure the description as well? Or are there reasons for not wanting to do this?  There may be little true cause for concern here except that, in promoting others to use iNaturalist, one major focus of criticism is on how well species at risk locations are obscured.  Even though I have found iNaturalist to be at least as good as other databases for obscuring coordinates, every little detail or possibility for trouble seems to be able to convince many to not use it. 

2.  In the situation where the species was seen on a shoreline - is there anyway to have the 10 km arbitrary point placed on the land form?  The area of possible location becomes very small when the point chosen is almost 10 km from any land and the species in question is a land-dwelling one.

Thank you,

Nicole

[Scott Loarie]

Hi Nicole,

1.  Would it be possible to obscure the description as well? Or are there reasons for not wanting to do this?  There may be little true cause for concern here except that, in promoting others to use iNaturalist, one major focus of criticism is on how well species at risk locations are obscured.  Even though I have found iNaturalist to be at least as good as other databases for obscuring coordinates, every little detail or possibility for trouble seems to be able to convince many to not use it. 

can you point to any examples where the description on an obscured observation is revealing the exact location? And if so can you leave a comment expressing your concerns and asking the owner to remove those details and let us know what happens?

2.  In the situation where the species was seen on a shoreline - is there anyway to have the 10 km arbitrary point placed on the land form?  The area of possible location becomes very small when the point chosen is almost 10 km from any land and the species in question is a land-dwelling one.

Again can you provide an example? On average, the area that would have to be searched to locate a land dwelling critter on a shoreline from an obscured observation is 157 km2 which is larger than all of San Francisco County. I don't know if its fair to call this very small.

Best,

Scott

 

--

You received this message because you are subscribed to the Google Groups "iNaturalist" group.

To unsubscribe from this group and stop receiving emails from it, send an email to inaturalist...@googlegroups.com.

To post to this group, send email to inatu...@googlegroups.com.

Visit this group at http://groups.google.com/group/inaturalist.
For more options, visit https://groups.google.com/d/optout.

--

--------------------------------------------------
Scott R. Loarie, Ph.D.
Co-director, iNaturalist.org
California Academy of Sciences
55 Music Concourse Dr
San Francisco, CA 94118
--------------------------------------------------

[Charlie Hohn]

Yeah, not to be a cynic but for those people who aren't 'convinced' that iNat is a good thing because of elaborate "blind points" in the geoprivacy... in my experience most of them will not be swayed by evidence. They don't like the technology (or any new technology) for whatever reason and if a magic 100% perfect geoprivacy scheme appeared they would find another reason not to use iNaturalist. These days I just ignore those people. Sometimes they come around, usually they don't, at least not for years. 

The geoprivacy will never be perfect, it is really good and has come a long way but it isn't perfect. I do think sometimes Google populates exact location descriptions in the location text (i.e. "chickering fen", a fen of a few acres appears in the text). I do think it might be worth making sure that obscures, though it might already be done. As for the randomly obscured points I know it has been discussed to switch to just rounding up the coords so they all cluster in the same spots like a grid (I think others do this)... this would not be perfect but would be a good thing, if not urgent. For instance I generalize all of the stuff at my house to our town because otherwise there would just be a huge bulls eye of observations around my house because I've taken hundreds here. The system is not perfect.

Of course, like you said iNat is still better than other things out there. I remember someone flipping out about how a rare plant on one mountain was possibly inferrable on iNaturalist only to Google the mountain name and find a botanical society newsletter describing exactly where it is. Then there are geolocated tweets on twitter, instagram, you name it of rare plants or animals. Often the person doesn't even know it is rare because unlike iNat no one tells them. We are better off with a large and strong community here than to scatter amateur naturalists amongst a million websites where no one will be bringing up these concerns. 

There is a subset of the older population who are a shining example of willingness to try and learn new things. You find many of them at the top of the leaderboard here each month. I hope I  become someone like that as I age. For the rest of the people of the older generation, and a fair number of the younger ones too, they have just made up their mind. They whine about the 'distraction of smartphones' then go home and watch a sitcom. They fret over geoprivacy despite there being a deep history of plant 'borrowing' from within the naturalist community that probably has done more harm than random outsiders. Some come around eventually... but in my three years working where I work I have made no progress in getting other biologists onboard. Meanwhile, hundreds of other people in Vermont have jumped on and posted tens of thousands of observations... but basically none from where I work. It's a bummer... but we in Vermont can count ourselves lucky having the Vermont Center for Ecostudies who came and promoted iNaturalist in a way I didn't have the audience to do.

And of course... there are certain things that should just not be posted on the Internet. I never post locations of ginseng on here ,or rare orchids, because I don't post them on the Internet. Even wild ramps and fiddlehead I usually obscure. I don't post unobscured locations of a secret fishing hole or deer hunting spot here or anywhere else. Of course I am new and not good at fishing or hunting so no one wants my 'spots' anyway. If I went to Africa and saw a rhino, I probably wouldn't post the exact location. It's a bummer that the world is that way but for now it is.

Anyway, sorry for the wandering message.

============================
Charlie Hohn
Montpelier, Vermont

coyot.es/slowwatermovement

[Nicole R]

Hi Scott,

Sorry for the delay.  If you look at our park project, Killarney Provincial Park Wildlife Project,  you can see a couple points floating in Georgian Bay but that were probably not found in the water.  It isn't an immediate concern, just something I noticed that may be more of a concern with certain species or in certain areas.  In terms of the description, again, not an immediate concern, your idea for commenting on descriptions if they are too forthcoming is good enough in most cases.  

[jesse rorabaugh]

If they are not already, dates should also be obscured. At most the month of the observation should be posted. This is because it is often possible to deduce where something is based on what observations it is right next to.

[Faerthen Felix]

That's true, Jesse. I've worried about that, too. Want to find that rare orchid or carnivorous plant with the obscured location? Just go to the observer's calendar page and see what else they uploaded that day, and from where.

Maybe don't show sensitive species in the calendar or running life list. Or make a separate list for those, which is disconnected and out of context with any other observations (no dates).

FF

[Charlie Hohn]

The life list doesn't update that fast anyway. I think that one is ok. 

This is all a balance. These things are important to consider but also just education and there are some things people just shouldn't post. 

Is there any data on how frequent collection or harassment is for non targeted species? Right now we are obscuring things like California fan palm or super obscure sedges no one can tell apart from a common one without a key and hand lens. Are harassment and collection even remotely relevant to those? Maybe we should tighten our focus and also tighten the security of those things that matter. 

--

You received this message because you are subscribed to the Google Groups "iNaturalist" group.
To unsubscribe from this group and stop receiving emails from it, send an email to inaturalist...@googlegroups.com.
To post to this group, send email to inatu...@googlegroups.com.
Visit this group at http://groups.google.com/group/inaturalist.
For more options, visit https://groups.google.com/d/optout.

--
Sent from Gmail Mobile

[Lilly Adzler]

I've thought of that, and my solution is to obscure everything on that date and location.

[Charlie Hohn]

My solution is to just not add most rare plants unless i am certain there is no risk... so basically rarely. Not sure that is optimal but I work for Fish and Wildlife so I have other ways of tracking them anyway (i.e. they aren't lost to science or land management)

--

You received this message because you are subscribed to the Google Groups "iNaturalist" group.
To unsubscribe from this group and stop receiving emails from it, send an email to inaturalist...@googlegroups.com.
To post to this group, send email to inatu...@googlegroups.com.
Visit this group at http://groups.google.com/group/inaturalist.
For more options, visit https://groups.google.com/d/optout.

--

[James Bailey]

I upload rare plants while uploading sightings from other locations or much later, with a different location "title", to avoid people tracking locations from what else I've seen that day.

James

[Faerthen Felix]

Sure, but I have no control over obs of rare plants by other people working in the basin, whether I choose to enter into contortions like that personally, or not.

[Scott Loarie]

Hi Nicole - can you add a URL to the observation(s) you're referring to? But generally, I'm not sure the landcover type and taxa-landcover-association data needed to inform that kind of algorithm exists for all taxa globally. Research on that would be an important prerequisite to exploring what would be involved in coming up with an obscuring algorithm that considers those kinds of factors.

All - re: obscuring the date in addition to the location - I added an 'improvement issue' here https://github.com/inaturalist/inaturalist/issues/678 but limited it to obscuring to day rather than to month. Would that be a good start?

While obscuring date to day would yield some benefit at little cost, obscuring date to month would add a great deal of costs (complexity/unknowns) to all sorts of functionality on the site - most notably 24-hour BioBlitzes which depend on using date to produce counts - for unknown additional benefit. I can't think of an easy way to provide a service for 24-hour BioBlitzes via iNat if obscured observations had no day associated with them (for example - nearly half of the species reported by the recent Hawaii Volcanoes NPS BioBlitz are threatened and there is a strong desire from many for them to be included in the BioBlitz counts). So IMO we should wait for some more general progress/evidence/discussion on the 'Costs/benefits to conservation of publicly accessible biodiversity occurrence data' debate before making such a costly change to iNat's services.

Curious to hear feedback, but I'd also note that we're trying to thread a needle here between offering a service for recording/sharing occurrences while still being respectful of sensitive data for conservation - and as in all good compromises that means there's going to be some dissatisfaction** on both ends of the spectrum.

-Scott

**And as always, if you're concerned about keeping YOUR data private, iNat might not be the best tool for you, remember 'It's NOT a way to collect secret information'.  Also if you are interested in moving forward the more general 'Costs/benefits to conservation of publicly accessible biodiversity occurrence data' debate please post a link to your article/arguments/research here so we can learn more, but I don't think this forum is the right place to have that debate. First, such an argument would have to include bigger players like GBIF, eBird, or Flickr that account for a much larger contribution of publicly accessible geo-referenced biodiversity occurrence data than just tiny iNat. Second, such an argument should compare iNat's approach to balancing what to obscure with the approaches take by these other players (which I understand is more risk averse than that taken by many others). Both of these arguments would be out of scope for this iNat-centric forum.

--

You received this message because you are subscribed to the Google Groups "iNaturalist" group.
To unsubscribe from this group and stop receiving emails from it, send an email to inaturalist...@googlegroups.com.
To post to this group, send email to inatu...@googlegroups.com.
Visit this group at http://groups.google.com/group/inaturalist.
For more options, visit https://groups.google.com/d/optout.

--

[Nicole R]

Hi Scott,

Here are some examples.

Monarch:  http://www.inaturalist.ca/observations/1871143

Blanding's: http://www.inaturalist.ca/observations/1645540

Smooth Greensnake:http://www.inaturalist.ca/observations/805247

Again, it is not a major concern for the observations we have on our project right now, (especially as the photographs give more information than the location does anyway), but maybe in the future if it a solution is found.

Nicole

[Matthew Muir]

For those interested in this topic, here's a recent piece published in the Guardian: Poachers using science papers to target newly discovered species: Journals begin withholding locations after warnings the data is helping smugglers drive lizards, snakes and frogs to ‘near-extinction’

This is a contentious issue, but I think it's so important. Even where we disagree (hopefully civilly!), we need to stay informed of new developments/perspectives as a community and consider approaches to this problem. For example, I learned from the article that the IUCN Red List has the following policy on sensitive location data:

For the record, I think the iNat team and community have so far treated the issue of obscuring species data seriously and transparently. As access to data on conservation status and threats allow, I would also love to see us align closer to the policy above (or be even more conservative like considering country- and state-level assessments as well) and/or coordinate with the wider conservation community. If there are new rules and standards up for debate (the IUCN link specifies that the policy is for 2012-2016 so it may be up for consideration again this year), it would be great if iNat could be involved.

There is at least one other thread on the topic of conservation status and obscured coordinates --> "Curator Talk: Let's have a discussion about geoprivacy" <-- but this is the original thread so am tagging here.

Cheers, Matt

[Ken-ichi]

Do you know if the IUCN publishes the data needed to assess these criteria in an API? Some of our decisions about geoprivacy for species are driven by what's best for the organisms, but if we don't have data (e.g. anything in section 3), we can't really do anything other than make policies for our curators to follow when manually curating taxa.

Visit this group at https://groups.google.com/group/inaturalist.

[Andrew Simon]

It would be great to get the conservation rankings of the CDC recognized by iNat.

http://www.env.gov.bc.ca/cdc/

Let me know if I can help in any way.

[Ken-ichi]

We have already imported NatureServe statuses for for Canada and all
its provinces. Do you know if there are any discrepancies with CDC
rankings? Also, assuming you're referring to the data presented at
http://a100.gov.bc.ca/pub/eswp/search.do?method=process&searchType=COMBINED,
would you want COSEWIC, SARA, and/or CF Priority statuses?

[Charlie Hohn]

Sorry to rehash geoprivacy yet again... but previously we in Vermont (Heritage, VCE, etc) had long discussed and settled on which species should be obscured. Recently, however, more natureserve statuses have been added to iNat for Vermont (and probably other states). That's a good thing. However they also were all set to obscure. This is of course important for anything that might have collection risk. However a lot of edge-of-range plants are being obscured... species that are very common elsewhere. Edge of range data of a precise sort is very important and useful, and I don't think we should be obscuring state-ranked plants that are globally common unless we have reason to believe there's a collection issue. And we can throw that net pretty broad. But if you can concoct a scenario where tulip poplar poachers will strike Vermont, or where a sedge will get poached when it looks identical to other common species except for peregynia hairs that only are visible for 3 weeks out of the year... I'd be very surprised. Likewise a ton of tree species are obscured in Quebec which are super common in Vermont. It makes it hard for me to see their range and habitat requirements up that way, and does nothing to protect them. Who's going to be harassing butternut trees in Quebec City? No one.

More broadly, can we have discussions about making these wide ranging changes rather than suddenly obscuring (or unobscuring) large numbers of species without talking to people using the data? It's frustrating.