Utah Birds, Utah Birding, and Utah Birders. Promoting the sharing of information, and the conservation of habitat for birds in Utah and elsewhere. We are a group of people who want to share what we know, and create a positive birding experience in Utah.


a blog by and for Utah Birders

eBird, show me those unverified reports

posted by Tim Avery at
on Friday, April 26, 2013 

If you know me, you know I am a huge eBird advocate.  I think its a great tool, and has only gotten better with time (as any good software with a large user base and a developer that takes feedback should).  There are still a few things I would like to see when it comes to exploring data--specifically ALL reports of a species, whether the report has been verified or not.  This would be very easy to show, and very useful to many birders.

A first fall female Cape May Warbler--whether or not the photo is conclusive, it's still a report worth being able to see.

My reasoning is simple: There are numerous checklists with rare birds that have not been verified--but the report very well may be correct.  This means that when exploring data on eBird maps there are species that don't necessarily show up on the maps in areas they have been reported.  And some of those reports are valid reports, but for any number of reasons the eBird reviewer has decided not to verify it.

I have seen 3 Cape May Warblers in Utah in the past 20 years, yet none show up on the eBird maps.

Now wait a second Tim, that reviewer is the one making sure the data pool isn't a mess? Yes, I do agree, but I also agree that if done correctly the information would be beneficial to everyone looking at it, and still be able to show what sightings are perhaps legitimate, versus ones that aren't.  Talking specifically about maps, I will give you an example

Let's say you want to look at HERMIT WARBLER sightings in Utah.  This species is rare but annual fall migrant.  It cuts across the southwest corner of the state in mixed Ponderosa woodlands--the Kolob area and Oak Grove Campground are two areas where this bird is reported.  Although the species is annual, there are very few verified reports in eBird.  This may just be because very few people report them--but let's pretend they are reported more frequently, but with questionable or no documentation, and the reviewer doesn't feel comfortable approving them.

This was the only photo I had of Kolob--during prescribed burns one summer.

At a top level when you come in with the maps if there were locations with verified and unverified sightings they would still show up how they currently do--as purple blocks.  If there were locations that had only unverified sightings, they would show up as red blocks.

How I envision unverified reports showing up from a high level.  Locations with ONLY unverified reports will show up red.

When you zoomed in to the actual location points and clicked on one, not much would change.  It would still display everything as is, with the verified reports showing up as they currently do. The major change would be that at the end of the list is a link that reads, "view unverified reports".

What you would see when you click on a way point--the same as the current view, with a link to pull up the unverified reports.

If you were to click on the link, the list of unverified reports would show up below the verified reports in the same fashion--however nothing would be bold so that it doesn't have as much prominence as the verified reports.

How the unverified reports would appear with the verified.

This would accomplish two things.  First, it would give a complete visual of the ACTUAL reports of a species in a  given area, or specific location.  In doing so it would make the data more valuable so that users could view the whole picture.  Separating the unverified reports out leaves the interpretation of the data to the user.  The majority of birders could look at that and be able to determine the obvious reasons reports weren't verified--versus reports that seem to actually make sense that weren't verified for a random reason here or there.  Secondly, and this will sound bad to some people, but for certain species that are over reported, and seemingly misidentified time and time again, it will make the observer more aware in a public space of their continued mistakes.  A lot of people put species into the database over an over that are misidentified.  A lot of these are probably common species, and it goes unnoticed, because in the grand scheme of things it falls into an acceptable margin of error.  However for other species, that maybe are continually reported, it will be an eye opener.  If you are the only birder who ever reports LEAST TERN from Lee Kay Ponds, and you seemingly report them 2 to 3 times each year, but the other 99% of birders only see one there every 5 years--that will be obvious in the data.

Not everyone self-polices, myself included.  I have numerous reports that have not been verified by the reviewer for some reason or another--however, many of those reports come with valuable information, some with undeniable photo proof of a species, and SHOULD be accessible with the rest of the data in the public records--not through some awful report you can generate from another website that is wildly impossible for your average person to look at and make sense of.  I would rather have my unverified sightings visible to the public, and let the users make their minds up about the sightings based off the notes and photos I provide.

How a location with only unverified reports would show up when the way point is clicked.

In some instances this is important for birds that have never been reported otherwise in the state, or there have been a handful of reports but no verified sightings in eBird.  A prime example is GILDED FLICKER.  This species has now been reported a handful of times from the Mojave Desert in southwest Utah, but none of the reports have been verified in eBird--despite written descriptions and photos of the birds.  Now I don't know the reason for these reports not being accepted--but it is a disservice to other observers and in this case hides important information about the possible extreme northern limit of this species.

Female Gilded Flicker flying past me.  Other photos show other field marks for this species as well.

Not everyone is going to agree with this approach--specifically folks who have lots of unverified reports and might feel uncomfortable with that being out there.  Others, perhaps those in the strictly scientific side of things, might also not like this approach because they feel the person doing the verifying job is weeding out BAD data.  And I see both points.  However, this database is open to the public, and the data should all be easy to read an interpret for all reports.  The reviewer will weed out the majority of bad sightings, and the user will be able to look at that data and make their own assumptions based off of it.

Just my two cents, on something I would like to see to make the data pool even more useful.  What do others think about this kind of idea?  Or do others have ideas that would improve the user experience or data provided in some way or another?

Labels: , , ,

Blogger Stephanie Greenwood said...

I think you make a lot of valid points; I would definitely support a change to ebird like this.

April 26, 2013 at 12:03 PM  
Anonymous Christine Bastian said...

I usually am a "the more info the better" person, and this is no exception. Also because there have been proven changes in regions and migration routes, being able to go back and view reports that were unverified in previous years that have now become verified would give people a view of when these changes may have started taking place. Perhaps even pinpointing possible causes.

I have no idea what is considered verifiable since I sent in a picture of a Glaucous Gull that was ID'd by Cornell itself and still wasn't listed as "verified". I wasn't able to embed the photo in my report, so perhaps that was it. Either way it would be fantastic to see what people are seeing (or think they are seeing) and checking it out for ourselves.

I guess the only "con" I can think of is that a new birder like me may be less quick to write of a bird as "no that couldn't be it" if there is a possibility it is. But even that is only a "con" if I'm wrong! ;-)

April 26, 2013 at 12:39 PM  
Blogger shyloh monster said...

I like this idea. Perhaps an added checklist tool like the already available "hide from public output" option, where users could decide for themselves if they want to "hide/unhide unverified sightings from public output" that would then appear or not appear on the the map.

I'd like a quick method of looking up my own unverified observations. It would make self policing a whole lot easier when using the 'my ebird' tools.(If this is already available, please enlighten me!)

Hats off to the eBird reviewers out there. They do a helluva job. I've created my own option for seeing every eBird checklist submitted in my region. I can't imagine the hard work it must be to keep up and keep accurate.

a good topic for the ebirdtechtalk group.

April 26, 2013 at 2:19 PM  
Anonymous Drew Webe said...

It might be appropriate to figure out a way to display 'certain' unverified eBird sightings. However, a majority of sightings that don't make it into the public eye in my region (Pennsylvania) are flat out wrong. (Carolina Chickadees well north of the Appalachians, thrushes other than Hermit before mid-April, winter reports of Broad-winged Hawks). When a record is decidedly wrong, I don't think it should show up. However, there are other times when perhaps not enough info has been provided to the reviewer for them to comfortably verify it, and perhaps these types of records could be added to the map in some way. The one huge issue is that most people aren't going to critically look through records, they will see a dot on the map and assume that it is correct.

April 26, 2013 at 3:17 PM  
Anonymous Anonymous said...

I actually agree with most of this,and believe that all reports,even obviously wrong ones, should be visible some how,just give us the data,we don't need someone to tell us what to think. However,I also think that any review species,or first state record,should always be invalidated,no matter what proof is offered, unless a record is submitted to the records committee,because eBird should not take the place of the records committee.

April 26, 2013 at 11:54 PM  
Blogger Kenny Frisch said...

So you are proposing to purposely invalidate valid records even ones with sufficient documentation so that the records committee get the glory?

That just shows that in the 21st century that records committees have no place. If they are willing to tamper with the science that is going on in ebird out of spite, who says that they are going to tamper with other records that are submitted to them based? Who's to say that they won't reject a record with good evidence because of personal grudges or accept a record with not enough information because it was from someone on the committee or one of their friends.

Committees can be biased as the people serving on them. It's time for more open records.

April 27, 2013 at 4:49 AM  
Blogger matthew pendleton said...

Tim, I Couldn't agree More. I Have Had Similar Thoughts And I Am Glad You Bring It Up. I Too Love Ebird And I Continually Advertise It In Hopes That More And More People Will Use It. I Agree With Most Of the good Points That Have Been Made Except Changing Ebird Data So The Records Committee Can Deal With It. That Just Doesn't Make Sense.

April 27, 2013 at 9:28 AM  
Anonymous Matt Hafner said...

I agree with Drew Weber. Most of the "invalidated" sightings we have here are either truly wrong or at least lacking in sufficient documentation. Identifiable photos would certainly be accepted without issue. I'm guessing you have already talked about this with the eBird reviewer for Utah.

April 29, 2013 at 8:25 AM  
Blogger Tim Avery said...

@Stephanie: Thanks for the comments Stephanie!

@Christine: eBird reviewers either validate a sighting or leave it as unverified based on a filter system that weeds out rare, or out of season sightings. Usually the eBird reviewer will contact you to seek out documentation to try and make sure the sighting makes sense. If the documentation proves enough the sighting will be validated. If not then it is left as is and does not show up in the public data pool.

@Shyloh: I don't like the idea of letting the user decided whether or not their unverified sightings appear in the data pool--that gives the user the power to decide which of their sightings. Why would you want to hide your unverified sightings, if you think it is valuable scientific data? I agree with the 2nd comment though, I wish I could easily pull up all my unverified reports, so that I could specifically go in an add details--but only if a system were in place where the public could see them. I agree the eBird reviewers do an amazing job--perhaps one thing that would make it easier in Utah is having multiple reviewers. 8 years ago there were less than 100 people using it in Utah so it was easy to manage. Now it's a whole other store. Feel free to share this post with the ebirdtechtalk group.

April 29, 2013 at 11:46 AM  
Blogger Tim Avery said...

@Drew: I kind of agree here. You make a really valid point as there is plenty of data like that for Utah. There really is no point in showing a report of 9 Lapland Longspurs in May in Utah--it's clearly a bad ID, and if the observer isn't willing to change it in their checklist, it is just bad data. I guess this would be a setting on the reviewer side where the reviewer doing a thorough job could make the call based off a set of rules that makes sense. The dot on the map can be handled via a different colored dot and icon that helps users understand the difference between validated sightings, and unverified.

@Anonymous: Thanks for the comments. I strongly disagree with your connection to records committees though. A records committee IS NOT a governing body for eBird or science in general. They are loosely organized groups of birders, with NO official governance of birding and science in general, and in my opinion only muddy the waters of actual science due to a number of internal issues plaguing these groups in various states. Just do a search on our blog and you can see my past discussions on this subject. If out of 100 birders, only 10 are submitting records to a committee, what about the sightings of the other 90% of birders? Records committees at least the one here in Utah do no outreach to the birding community, and have done nothing to make them seem like a valuable tool for us to use. eBird is replacing records committees and they will one day in the near future be a relic of the pre-digital years...

@Kenny: preach on brother man!

@Matt P: thanks for the comments dude!

@Matt Hafner: Yes I have--we have been friends for almost 20 years and I openly discuss this kind of stuff with him often. And I agree with you and Drew--but there are definitely some things that could be done to improve data quality and transparency. A good example might be something like this: someone reports broad-winged hawks during migration more than any other person in Utah and from a location where others aren't seeing them. There are no photos, and broad-wingeds are known migrants but the observer is new to birding and insists its what the birds were. This is a case where I would like to see the reports--just my two cents though, well that and the two-cents of a few others around here apparently :)

April 29, 2013 at 12:01 PM  
Anonymous Anonymous said...

Now Tim, do totally agree with you as concerning this with Ebird. Yes I myself would love to see those Unverified Reports! Also do think that this with not showing the unverified reports and having so much verification could work against them in some ways. If there is so much verification needed on certain species, along with number seen and the dates, how much also could this work against them. For some birders do think will withhold that data to just themselves and not post that data to not go thru all the hassle. Yes there definately needs to be verification! But on the other hand have seen where people who are in authority often look down upon those who are not in authority. We the average person is Not as dumb as those in authority would like to think. So if there was this one addition, do think it would improve things on Ebird. Also do Love this UtahBirders Blog Spot. Wishing You the Best! Kayla in Wyoming

April 29, 2013 at 1:16 PM  
Anonymous Terry Bronson, Morgantown, WV said...

There is one other aspect to unverified records--those from areas where there are no eBird reviewers, which primarily affects foreign countries. Think Uzbekistan, New Guinea, Bolivia. There seems to a threshold in eBird for certain species, but because there is not an eBird reviewer records that surpass this threshold linger in eBird limbo for a long time. I have eBird records from Sweden, Panama, and Peru that don't show up for this reason--some two years old. One involves the first-ever reported sighting of an Orange-crowned Oriole west of the Panama Canal; two involve thousand-bird-plus sightings of Barnacle Geese and Great Cormorants in Sweden, some involve 500,000-plus sightings of seabirds on offshore Peru islands. These were easily identifiable species, though I concede estimates of 500,000 birds are just gross estimates. I think such sightings should be reported and applaud the proposal for an unverified section in the eBird output.

April 30, 2013 at 4:39 AM  
Blogger David Lawrence said...

Here in Idaho we recently had an episode involving Snowy Plover. Over a 2 day period it was seen and reported by as many as 8 observers. Identifiable photos were submitted with at least one report, yet the reviewer not only chose to not confirm the reports, and chose not to leave them as unconfirmed, but actually took the step of totally invalidating the reports. Thus they even disappeared from the 7-day rare bird reports for the state of Idaho. I personally contacted other reviewers in the state and they finally over-ruled the initial reviewer and reinstated the reports, at least as unconfirmed. Just not the way it should be.

April 30, 2013 at 7:35 AM  
Blogger dmorin said...

I was just going to add that local reviewers will fall/have fallen into the same traps the record committees do. The personal politics that record committees were renown for have started to pop up in ebird. I`m personally thinking of going back to just using AviSys or a private bird-record keeping program just to not have to deal with `reviewers`.

April 30, 2013 at 8:02 AM  
Blogger Tim Avery said...

@Kayla: thanks for the comments!

@Terry Bronson: You bring up a great point about these areas. Peru is a great example--my biggest headache with eBird Peru is actually the fact that the checklist given to you to fill out shows all the species of Peru instead of being broken down by region, etc. This is a case where more moderation is needed by someone willing to put in the time to make sure these lists are easier for the user to use, and give more accurate data. Some funny things show up in the wrong habitat/locations due to this. At the same time, great sightings can be left out for along periods of time because of this lack of moderation.

@David Lawrence: This type of action on the reviewers behalf completely makes the public data useless. If valid sightings are being removed by the reviewer, because they feel they are the end all be all they are doing a disservice to the community. How any reviewer in their right mind would do something when valid documentation is obtained is beyond me--and shows a real lack of care for making sure the data is being shown accurately. This is prime example of a scenario where the system I propose would make perfect sense.

@dmorin: Thanks for the comment. The biggest thing to remember about eBird is they are trying to have a "good data set". They need to be able to have verified and unverified sightings--I am okay with that. eBird wasn't created to be our personal life life keeper. It was created specifically to allow citizen scientists to gather data for scientific purposes. In that sense what I am proposing could be ignored completely on the basis of "we have a system in place to handle 'potentially' bad data." But what would be better is having a way for data to sit in both verified and unverified status and still be visible to the public to make their own assumptions. Scientists using the data can then still weed out data they want to ignore--but for the purpose of birding, bird finding, learning status, distribution, range, timing, etc, it provides a wider array of data to the users.

April 30, 2013 at 9:19 AM  
Anonymous Matt Hafner said...

Tim, I guess I'm a little confused. If there were notes/photos supporting the identification of the BWHAs (or GIFLs), why wouldn't they be accepted? Doesn't everyone put in notes for their rare birds? In Maryland, we try to require notes from everyone for flagged species (within reason, filters are not perfect). We've gotten most of our top birders to lead by example and show that everyone should submit documentation. Ebird data is forever and we want those field notes preserved for the distant future when it's unclear who was/was not a good birder.

April 30, 2013 at 9:28 AM  
Blogger Tim Avery said...

@Matt Hafner: Good question? This is where the flaw of the eBird reviewer comes in full force. These guys are just birders and volunteers, and some are far from experts. Be it any number of reasons a reviewer may not want to validate a sighting, and even if there seems to be great supporting evidence there are numerous sightings missing from Utah in eBird--and probably every other state. Even if there isn't great supporting documentation, valid sightings exist and this is where my idea would make perfect sense. The reviewer doesn't need to verify the sighting--just leave it and let you or me make our own educated decisions based off the data provided.

April 30, 2013 at 9:34 AM  
Blogger Marshall Iliff said...

Tim, do you plan to embed the Cape May photo you show above in the eBird checklist that reports it? Without documentation in eBird, I am not sure the record should be reversed. I agree that the photo above shows an unequivocal Cape May, but I am looking at the checklist now and it has no notes or embedded photos: http://ebird.org/ebird/view/checklist?subID=S5470698. With some documentation, I will gladly contact the reviewer to suggest it be accepted. I agree that we should find a way to make it more obvious which checklists need notes/photos.

I am one of the 3 eBird project managers and we have all read your post and comments, which have spurred good discussion. I hope everyone who has commented here has a chance to read this article about the review process: http://help.ebird.org/customer/portal/articles/1055676

Please understand:

1) Pending records and records treated as "not valid" are different. Some records are not displayed simply due to time lag in the review. We work hard to minimize this across 500+ reviewers globally, but there are certainly review regions that have time lags. Since we don't pay reviewers, we continue to be thankful for the volunteer help they provide. If they were being paid, then setting review deadlines might be appropriate.

2) Re: Peru, we do not yet have an active team reviewing (also in many other countries). Each week we add new people to the global team though, and we are working to solidify partnerships in Europe and beyond to really ramp up review/filter creation. Building a global community takes time and we continue to work towards this goal. We do apologize for areas (like certain departments in Peru) that have inappropriately long lists. eBird review is fully fleshed out in USA and Canada, but outside of those areas it can be spotty (e.g., Africa). In those areas, you essentially are seeing all records submitted without vetting, and in those areas there is no way to tell the good data from the bad data.

3) Here's an example: check out the Crested Caracara map: http://ebird.org/ebird/map/crecar1. The species does not occur south of the Amazon, but due to taxonomic confusion, it is reported widely from Brazil. Brazil has both species, but Crested only in the far north, and Southern in the south (http://ebird.org/ebird/map/soucar1). The filters are not yet regional in Brazil, so all records go through and are treated as valid. Soon we will implement these filters and flag the records of Crested reported from areas where only Southern occurs, and the map will show the ranges better. There are hundreds of similar cases where the filters and reviewers are doing their job and the maps are "clean". Do we really all want a system where known errors are shown? I, for one, would rather see a reasonably accurate map of the two caracaras in eBird, rather than a confused mess where both species are reported throughout South America.

4) I understand the point that no one wants valid, well-documented records to be excluded from public output. I hope everyone is aware that with good documentation, a case can be made to correct such treatments. No record is EVER deleted from eBird, so your Cape May can be corrected, especially since the photos are diagnostic. We do need to improve the process to have such records reassessed. But my twist would be to encourage more complete documentation so that there are fewer cases where well-photographed records (like the Cape May and Gilded Flicker examples) are treated as not valid. On the eBird side, perhaps we can improve ways to display records in your account that need additional documentation.

5) If there are problems with individual reviewers, please contact us at ebird@cornell.edu and describe the problem. We agree that the process needs to be fair and we can remove reviewers that are not polite or not performing the job in the spirit of eBird. That said, I know of very very few cases where a reviewer is not following our policies.

April 30, 2013 at 10:22 AM  
Blogger Tim Avery said...

Marshall: Thanks for the comments, and thorough/thoughtful reply. Just as an FYI, this post was purely intended to talk about ways to improve the availability of ALL the data in eBird, not just sightings that have been validated. I happened to use real life examples to make my point--as for the warbler, it is my bad that it isn't on the checklist, I have recently spent considerable time going back and adding documentation from my website to my checklists to try and remedy this. I think the majority of the folks commenting are probably aware of the review process. I am well aware of it, and have been a reviewer during the GBBC in the past. As to your points, I just want to comment back on each one:

1) Pending records and records treated as "not valid" are obviously different--however, they do not show up to the public--making them "the same" in the eye of the user. It is understandable that there would be time lags, but with a system in place like what I am proposing this could all be handled via statuses and how the records are displayed on the map.

2) Re: Peru -- you need not apologize, this is understandable. But surely anyone with even a remote knowledge of the birds of Peru could take care of setting up even base filters by district in a rather short period. Or why not approach a "local expert" like Gunnar Englbom and see if he would mind at least doing the ground work?

3) This is a bad example though, because Southern Cararara and Crested Caracara were once considered one species--this is where these lists need to be manually updated, or there needs to be a script that eBird can run to change these sightings based off region. Think about Sage Sparrow and what will happen when the AOU officially splits them this year (if, I guess I should day)? Same type of issue, that will work itself out over time.

4) One person mentioned this, and I agree it would be amazing to have a link to a list of invalidated/pending review species, so that we can easily go back and add documentation. It was also make some users astutely aware of mistakes they are making when that box is clogged with 200 reports of a species.

5) I would like to take this time to applaud Colby for the work he has done for eBird in Utah. He has done an amazing job with handling the sheer number of reports as they come in over the past decade. The usage has gained crazy momentum and I think my only recommendation would be more reviewers for states like Utah where in the past we only needed one, now we probably need it broken down into regions to take the burden off one individual.

April 30, 2013 at 10:48 AM  
Blogger Drew said...

>>If valid sightings are being removed by the reviewer, because they feel they are the end all be all they are doing a disservice to the community.

Agreed. And if that is happening it should definitely be addressed. But I don't think a couple rogue reviewers (I don't know the situations described so this is hypothetical) should justify flooding the eBird public output with a lot of bad data.

April 30, 2013 at 10:49 AM  
Blogger Tim Avery said...

@Drew: There were several examples of this mentioned in the comments, which is why I brought this up--that and the fact that the reviewer is the end-all, be-all. They make the final decision on whether or not data is validated, and I am okay with that. The example given above about the Snowy Plovers sure seems like someone on a power trip. But the same could be said for a reviewer who invalidates any sighting from a user who consistently submits reports without evidence--and I guarantee that is happening to. You and I both can look at an eBird checklist and know what seems out of place--which is why there is a need for it. But there are plenty of ways to show the bad data as well, so that it doesn't have the same prominence as the validated data, but is still readily available to view in the large picture. Hell, it could be as simple as adding a checkbox to the maps to "show/hide" invalidated data. As long as it is done in a smart way it really wouldn't foul up the data pool--it would add another interesting dynamic.

April 30, 2013 at 11:05 AM  
Anonymous Anonymous said...

Tim, again how I support you in what you are saying on this as concerning reports and ebird. now have seen many times when someone who gets some authority goes on some power trip. I think that is dispicable of what that reviewer tried to do with those Snowy Plovers in Idaho like Dmorin said. Now if Ebird put themselves in such an authoritarian type of status as to reviewers of peoples bird sightings, then who will oversee Ebird? Now have to ask that who runs Ebird and who are the reviewers to begin with? I just came across Ebird the other day and have mixed feelings. I personally have been birdwatching for near 35 years. And most of the time am back deep in the backcountry wilderness areas here in the west where few people go. I personally have quite a few birding records for these back area like the headwaters of the Yellowstone River region in NW Wyoming and the Escalante River region in Southern Utah. For many years have just kept these records to myself and gone about pursuing what I enjoy - living deep in the wilderness. On long trips one will always come upon some rarity it seems at sometime. Seeing at least some bird that is marked as rare on the checklist at times on some trip. But have noticed on Ebird that any bird that is not common and uncommon for one area and is rare, Ebird wants the data and withholds the bird for review. Now on my latest trip, I saw a Painted Redstart in Zion NP which did submit to Ebird but they have yet to get back to me or verify it while others who saw the bird at the sametime this spring, they verified. This puts a very bad taste in my mouth as concerning Ebird. And like others, I just might go back to keeping all my records to myself and have nothing to do with Ebird if they are going to be this way. Who are they to be my boss and say what I have seen and not seen when they were not there. But what you propose Tim is a great compromise do think. We people are smarter then lots of those who put themselves in authority and try to be Lords over the people think that we are. Would love to meet up with and talk with other birders bigtime. But just as for myself personally, Ebird will have to prove themselves to me if they can be trusted or not. Wishing Everyone the Best! Kayla in Wyoming

April 30, 2013 at 11:32 AM  
Anonymous Anonymous said...

Now in thinking need to make another comment. Ebird says that they have 500 or so volunteers who are the reviewers if I understand it right. Now who are actually these reviewers? What are the stats to those who will be scrutinizing and reviewing some birdwatchers trip report? Maybe Ebird needs to come clean and reveal themselves and their reviewers to us in the birdwatching community if they want our trust. It would be simple for them to make a page on the biographies of them that run ebird, and their reviewers, and those that are affiliated with them. Maybe with all my time in the wilderness, I personally would like to see and know who are those that put themselves in such a place as Ebird is seemingly now putting themselves into. Just my two cents I guess. Wishing Everyone the Best! Kayla in Wyomming

April 30, 2013 at 12:56 PM  
Blogger Tim Avery said...

@Kayla: Your state eBird reviewer is someone form your state (or multiples in larger states), who is either an expert, has some affiliation with Cornell, or has volunteered because there was a need. The vast majority are very good and knowledgable birders, and are doing a good job. It is no secret who they are. If they are doing their job, every time you report a rarity you will get an email from them to obtain documentation. ebird doesn't have some ulterior motive, and isn't out to "get the newbies"--it is a great platform that you can use by CHOICE. Just because I don't like some of the features lacking doesn't mean that I don't feel it is an awesome piece of software, and I will continue to use it no matter what.

April 30, 2013 at 1:08 PM  
Anonymous Anonymous said...

@Kayla - Your Painted Redstart records will be validated. It just requires a little patience. It won't happen overnight, but in general most sightings are reviewed within a week of their input. Sometimes sightings get lost temporarily in ones awaiting review or when we go back through and run old data through a new filter, which can sometimes mean recoreds will wait a couple weeks, but in general they will make it there. Please email me at crn23@cornell.edu if you ever have any questions about something being reviewed in Utah.

-Colby (Utah ebird reviewer outside Cache County).

April 30, 2013 at 1:23 PM  
Anonymous Paul Ostler said...

It's very nice to see a reply from Marshall Iliff on this topic.

Marshall, you mentioned that you would rather see a reasonably accurate map of the two caracaras in eBird, rather than a confused mess where both species are reported throughout South America.

Here's a solution. Have a checkbox on the maps that, if checked, would enable the display of unverified records, as described by Tim. But if unchecked, everything would be displayed as today, with no unverified records.

Here in Idaho there has been much discussion on this topic, because rare bird reports do not show up on the eBird maps until confirmed by our state records committee and this is a lengthy process that can sometimes take a year or more.

From what I understand Team eBird encourages reviewers to defer to state record committees, and this can be a lengthy process. This also puts undue pressure on the record committees because most birders would like to be able to query for these sightings on the eBird maps.

Given that Team eBird leverages the efforts of the state record committees, then perhaps Team eBird could be persuaded that there is some scientific value even in unconfirmed records, and that they could fill a collective need for us here in Idaho by providing a checkbox on their maps to include unconfirmed sightings. This could help take the pressure off our records committee for speedy reviews, and given the volume of information in the eBird database there is scientific value even in unconfirmed records.

April 30, 2013 at 10:24 PM  
Anonymous Anonymous said...

I'm pretty sure that if eBird does not provide this capability, other more birder-friendly software will eventually have to replace it.

April 30, 2013 at 10:55 PM  
Anonymous Paul Ostler said...

I think Kayla makes some excellent points. Why is an eBird review committee a good thing, but a state records committee a bad thing? Here in Idaho, the state records committee is comprised of more experienced birders than the eBird reviewers. The fact that eBird withholds reports from experienced birders like Kayla so that an eBird reviewer with far fewer years of experience can approve it strikes me as wrong. At the very least, Kayla's and other's reports should appear immediately on the maps, but tagged as unconfirmed. Let the user decide which reports to trust. This gives the decision power to the people, not to the eBird reviewer. Giving an eBird reviewer the power to obliterate a sighting is wrong.

May 2, 2013 at 8:49 AM  
Anonymous Paul Ostler said...

eBird reviewers should be accountable. The best way to make them accountable is for eBird to be up front open with their records, and make it readily transparent the records they have not approved.

May 2, 2013 at 9:22 AM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Back to Previous