Date: 11/25/12 2:25 pm
From: Bill Hubick <bill_hubick...>
Subject: Re: [MDBirding] More on eBird Review


Jim, Tim, and others,

We appreciate your concerns, but I think we should take this thread off-line from here. We do want everyone to have the best possible experience while supporting eBird's important work.

If anyone wants to discuss off-line how we can improve, please feel free to e-mail the MD eBird team:
(Tyler Bell, Ron Gutberlet, Matt Hafner, Bill Hubick, Mikey Lutmerding, Rob Ostrowski, Jim Stasz.)

Our email addresses are:
<jtylerbell...>; <rlgutberlet...>; <mh1920...>; <bill_hubick...>; <mlutmerding...>; <rjostrowski...>; <jlstasz...>


Marshall Iliff is also CC'd and would gladly accept any feedback on MD eBird.

Good birding!

Bill


Bill Hubick
Pasadena, Maryland
<bill_hubick...>
http://www.billhubick.com
http://www.marylandbiodiversity.com




>________________________________
> From: Jim Wilkinson <lakekoshare...>
>To: <mdbirding...>
>Cc: Bill Hubick <bill_hubick...>; Marshall Iliff <miliff...>
>Sent: Sunday, November 25, 2012 4:21 PM
>Subject: Re: [MDBirding] More on eBird Review
>
>I agree with Tim.� This is not about unwillingness to provide information to ebird reviewers or rarities committees as implied by some posts--I try to provide complete data.� It is rather the lack of communication--not a problem until recently with ebird and also with the most recent MD Atlas where I submitted a record and found out 6 years later that it was rejected.�
>
>Regardless of how busy people are I think they could email or call at some point during a year or a several-year period if they have questions about a particular record.� If bird data collection efforts are to work, volunteers- both reviewers and observers- need communication.
>
>Jim Wilkinson
>Columbia, MD
>
>On Sunday, November 25, 2012 10:55:26 AM UTC-5, Timothy Houghton wrote:
>> Some anecdotal accounts suggest that everything is fine with MD eBird. Other anecdotal accounts suggest problems exist. I think that MD eBird, like any group, should take seriously the comments by concerned people and think about how policies and procedures can be improved. If improvement is called for, then how about focusing on that, on what, exactly, might be done. Perhaps nothing can be done; then so be it. The reviewers have a big task and work hard and everyone is grateful to them for their work. But I think that the guts and spirit of eBird are the mass of birders who take part in the eBird process, and I'd bet the Cornell people would agree.�
>>
>>
>>
>> So I think it would be helpful if a MD eBird representative would explain why some people whose birds are rejected are informed and others are not. Communication is important for everyone, more important, I would argue, than whether or not the bird has been rejected. People deserve clarity. Someone may know soon that his or her bird has been rejected--and perhaps even have the opportunity to clarify the find for further review. That person will at least know what's going on. Another individual who has not been informed will be wondering--for how long, weeks or months?--if his or her find is still in queue or has been rejected. Is he or she supposed to think this is ok when they know of others who've experienced more responsiveness from eBird?
>>
>>
>>
>> I also think eBird can't expect birders not to worry about whether their observations have been accepted or rejected. People aren't wired that way. But I think that 99% of people can understand a rejected bird if they think they are being treated like others and aren't held in limbo. Communication is the key. So if it's now impossible time-wise for everyone to be informed of a rejection, what can be done? More reviewers? A more efficient review process? What, exactly? If it's not possible to treat everyone the same way, then can we at least know at what point we can assume a find has been rejected? How long is too long? What's a reasonable amount of time? Everyone needs to be treated with respect.
>>
>>
>>
>> By the way, I hope people attend to my words and don't perhaps reject my comments as likely sour grapes of some kind. I only know of one of my reports that has been rejected, and even that one has only been tentatively rejected--as far as I know. My concerns are more general ones.
>>
>>
>>
>> Finally, up until the last couple of days, I was under the clear impression that everyone whose bird had been rejected was going to be informed. It doesn't help that I and others were wrong in that regard. It doesn't make sense to question the ultimate goal of eBird--quality data through filters and a fair review process; I don't think the "display everything" idea makes sense--but I think communication is a real issue and needs to be addressed in a way that makes the process clear and seem fair (as fair as possible)--arguable not anecdotally but as policy--to the hundreds of MD birders who submit checklists to eBird.
>>
>>
>>
>> Tim Houghton
>>
>> (Glen Arm)
>>
>>
>>
>>
>>
>> ________________________________________
>>
>> From: <mdbirding...> [<mdbirding...>] On Behalf Of Bill Hubick [<bill_hubick...>]
>>
>> Sent: Saturday, November 24, 2012 11:36 PM
>>
>> To: MDBirds
>>
>> Cc: Marshall Iliff
>>
>> Subject: [MDBirding] More on eBird Review
>>
>>
>>
>> Hi Everyone,
>>
>>
>>
>> Here are some follow-on thoughts based on feedback I've received following my earlier response. It's interesting that most people emphasizing concerns are writing about their own records not being validated (or validated quickly enough). I've wondered if we just haven't gotten the terminology right yet (I strongly prefer "confirmed" over "validated"). No one would balk if there were a "Photographed" column that was marked "No" when you provided no photos. Why be upset when a reviewer can't confirm a Yellow-bellied Flycatcher based on the comments "Seen well"? I've thought about this, and I think the more noble side of it is simply not wanting our efforts to be in vain. I assure you're they're not.
>>
>>
>>
>> Whenever we submit an unusual record, it is always saved and available for future research. Just like a reviewer not confirming a flagged report can never remove it from your lists, no record is ever removed from the database. Although we as reviewers provide a coarse flag of evidence for a record, a researcher in the future will almost certainly pull all associated records for their subject. If a researcher were reviewing early Eastern Wood-Pewee arrivals or Mid-Atlantic Northern Goshawks, they're going to deep dive into the data. In that review, they are likely to find confirmed records that raise their eyebrows and that they will still toss out. They could also see records that weren't confirmed that seem more reasonable in their expertise (or in a broader context re-reviewing older records; see below) and include them. Ultimately, all data you submit are permanently archived and will have whatever value they have to researchers studying that
species. With that in mind, I regularly add comments even for uncommon species that are not flagged for the place/date.
>>
>>
>>
>> The best way to ensure any sighting is used for years to come is to document it. The more unusual the sighting, the more important documentation is. Take the initial example of arriving Eastern Wood-Pewees. This species is so consistent in its arrivals that solid documentation is required on 4/18, but some sightings are expected locally on 4/25. Let's say we receive a report from someone's yard on 4/17 that says "Heard very clearly. Totally positive." Each year, some people will be 100% certain that these calls were not European Starlings; however, years of data show that Eastern Wood-Pewees are simply not arriving here yet. The species may just be touching down in the Gulf states. When we as reviewers accept a report, we are saying that we personally stand by it, and we are likely to be called out by eBird coordinators like Marshall or other state reviewers. And when we confirm something that is rare, we are setting a new precedent. When we validate
a new early arrival date for a species without documentation, we are carelessly changing what we claim to know about that species. The next year, that record could be used to defend still earlier casual reports, creating a feedback cycle that damages the data set and its trustworthiness. And of course these sorts of things are being reviewed in the context of environmental change. Not only must we get these things right, but our review process must be rigorous enough to defend the data integrity of the data. The more unusual something is, the more important it is that we have it right.
>>
>>
>>
>> Going back to confirming or not, I maintain that a 4/19 record of Eastern Wood-Pewee without documentation would be tossed out regardless of its current review status. On the other hand, records with notes can withstand the test of time. Let's say you submitted a Mississippi Kite in late April in 1991 and provided a paragraph of comments, but the Mississippi Kite phenomenon was still new and blowing everyone's minds. Your paragraph describing the bird might not have been enough to rule out Plumbeous Kite, so it wasn't confirmed. If I encountered such a report now, I would re-review and validate it. In fact, I use that example because some of Maryland's early Mississippi Kite records weren't accepted due to the high bar that seemed appropriate at the time. With the benefit of hindsight, we can now re-review, which is an item on our long to-do list for MD/DCRC. This applies to eBird as well as BRCs. Records can be reassessed based on new information at
any time. Anne Arundel ravens were crazy rare two years ago, and the Swan Creek birds (on the Bay?!?? no....) blew our minds. Although they're still flagged, I now mark any Swan Creek Common Raven as "Species--Known to be at location". We learn and adjust, but we didn't adjust in the case above until the detailed notes and photos were submitted.
>>
>>
>>
>> A logical comeback to all of this is "Why not just display everything?" This seems reasonable enough after all, at least for a moment. The problem is - I'm going to say it - people are wrong. Everyone is. A lot. So much. Without the filters, without the review process, we would absolutely find that Black-capped Chickadee is an abundant resident across all of Maryland. A review process - a very thorough one at that - is required for this project to work. Everyone involved in Maryland review has poured entirely too much of their being into Maryland bird distribution minutiae, and we are all nice people who want to do it right. Please, please embrace the review process and try not to worry about what is and isn't ultimately confirmed. It's all on your list and the researchers delving into the data will see exactly what you submitted. I have gone back and invalidated some of my own historical records, and would gladly do it again if there were concerns
raised. The data set is better for it.
>>
>>
>>
>> I hope that's helpful.
>>
>>
>>
>> Bill
>>
>>
>>
>> Winner, 2012 Mark Hoffman "Longest E-mails" Award
>>
>>
>>
>> Bill Hubick
>>
>> Pasadena, Maryland
>>
>> <bill_hubick...>
>>
>> http://www.billhubick.com
>>
>> http://www.marylandbiodiversity.com
>>
>>
>>
>> --
>>
>> -- You received this message because you are subscribed to the Google Group 'Maryland & DC Birding'.
>>
>> To view group guidelines or change email preferences, visit this group on the web at http://www.mdbirding.com
>>
>> Posts can be sent to the group by sending an email to <mdbirding...>
>
>--
>-- You received this message because you are subscribed to the Google Group 'Maryland & DC Birding'.
>To view group guidelines or change email preferences, visit this group on the web at http://www.mdbirding.com
>Posts can be sent to the group by sending an email to <mdbirding...>
>
>
>
>
>

--