John Bratton's picture

What exactly is an agreement?

If you look at http://www.ispot.org.uk/node/317103
I have suggested this might be Agabus bipustulatus but I don't think enough detail can be seen to rule out, for example, Ilybius montanus or Rhantus grapii. Two people have agreed. So are they saying they can see enough detail to be sure it is A. bipustulatus, or are they just agreeing it might be this? I thought an agreement meant you were confirming the id., however speculative it was.

John Bratton

Reply

Comments

DavidHowdon's picture

A few discussions

There have been a few discussions relate to this and suggestions that there should be a way to "weakly agree". I believe team iSpot are looking at a way of doing this (possibly a RAG type system).

dejayM's picture

agree

Sometims we (I) agree because it looks like the one we know so well and have been brought up to know.
But after agreeing, someone might point out that it's impossible to tell this from that, they may even modify the ID to a point of safety.
In those cases we (I mean I really) are left more confused than educated because we are still certain, in our limited way, that it IS what we have always known it as, but now discover there are three or four varieties that can only be separated with the aid of a microscope (say).
There are plenty of examples of that - my ID of a greenbottle is perhaps one - http://www.ispot.org.uk/node/318047
I'd sometimes rather leave my naive Agreement but I also have, recently, removed one (maybe three) or agreed on both the "you cannot separate them" one and my instinctive, 'everone knows it is' one.

Ray Turner's picture

Personally ...

When I agree an observation I take into account the confidence level. I fairly often decline to agree because someone has given an “I’m sure ...” ID and I’m not, whereas I do think they may be right and can see why they think what they do so would have agreed had they expressed a lower confidence level. If we don’t do this then what is the point of a confidence level at all?

In your example I personally would be agreeing “It might be this” (assuming I knew anything about diving beetles, which I don’t) and implicit within that agreement would be it may well be something else too.

Ray

Ray

John Bratton's picture

The confidence level is

The confidence level is attached to the first identification. There is no confidence level option for agreements. If an agreement is assumed to be at the same level of confidence as the original id., you would never get confirmation of a doubtful id.

Ray Turner's picture

Grey Area

I agree this is a grey area. However looking at the confidence level is pretty standard tool when deciding whether or not to agree particularly where the evidence is meagre, say a poor photo. A bird seen is silhouette against the sun may not allow for a ‘Sure as can be’ ID and (unless possessing a very distinctive jizz) would probably not get many agreements if it did. However many more people would be happy to agree to a ‘Might be’.

I guess the logical follow-on from this stance is if you view an observation at a ‘Likely’ level, say, which you know for certain is a particular thing then you should submit a new ID at the greater level. This would seem unfair on the original person posting though and so is deemed poor etiquette. Therefore it should not be considered wrong to agree because you thought the confidence level too cautious.

Mentally I make the jump in confidence level if I see 15 people, several of whom are expert, agreeing with an observation (Though interestingly if you look around you will occasionally find the very rare observation with multiple agreements etc that is plainly wrong; usually spotted by a total novice asking why everyone thinks it is such and such.) at a low confidence. iSpot sort of does this when selecting ‘Likely ID’.

Ray

Ray

dejayM's picture

last para

The very last paragraph is the key to some of this. 4 or 5 experts may allow me, in a slighly doubtful mood, to tick I Agree (how can they all be wrong?)
However, let's not kid ourselves, there are degrees of 'correct' in ID.
Green Bottle http://www.ispot.org.uk/node/318047
was right but it may have been any one of five 'scientifics'.
And there are plenty of others like Mainedhair Spleenwort http://www.ispot.org.uk/node/317748

How can we agree at all if we cannot tell the difference between any of those - but it is SILL a Greenbottle or Asplenium (say).
We are not going to sway the World with 'slightly-off' agreements, particularly 'novice' ones, so it's a cumulative agreement that matters, which is why, in my opinion, there should be FAR more aggreements made (is the maximum allowed 14?)

DavidHowdon's picture

I suspect

really one should not say "I agree" unless you agree from your own knowledge so just because a expert has agreed with something I'll not click "I agree" unless I do agree (and would have done so had no one else agreed).

In practice of course some of my knowledge comes from the experts here (for instance if they explain a key ID feature which I can then see in the image) so it is not always that clear cut.

dejayM's picture

I agree

"....unless you agree from your own knowledge.."
Yes, David, I Agree.
Derek

dejayM's picture

Blind agreements

I do think a lot of people blindly agree - I can't put my thumb on why.
But take this http://www.ispot.org.uk/node/319189
so obviously NOT a redwing yet, at the time of writing this, 6 people think so, despite pleas from real experts.
One of the problems is that people are obviously NOT reading comments - surely the most important part of iSpot.
So - what about a note on each ID which says "Before agreeing, look at ID notes and comments, if they exist".
I expect, in the case I mention here, most of those six will withdraw their agreement once they see the second ID and any one of the comments.
Derek

Mydaea's picture

Yes - and another equally

Yes - and another equally perverse case at http://www.ispot.org.uk/node/319192.

Fenwickfield's picture

reputation and agreements

I have read all the comments above and agree with the comments.I think this happens quite often and I too will not agree at times with people who put sure as can be when the only way to be sure is if it is examined under a microscope.I see this a lot with fungi and if they had put might be or likely I probably would have agreed.I think there are a lot of mis- identified observations.There is also the problem of just putting a photo on with no information about size habitat ect which is very important with certain subject's.I truly feel that ispot is starting to lose what it was originally started up for and is becoming more about point scoring.I have started to lose interest with some area's.I thought it was about sharing and developing knowledge but I am not so sure now.

Fenwickfield

dejayM's picture

Obvious postings

I agree with most of that.
What is written near the end ("There is also the problem..."..."point scoring"..."losing interest..") is revealing - for the Admin Guys.

I have also noticed that even VERY experienced observers are not adding ID descriptions nor location-type - worse, some are posting 'obvious' pictures to apparently attract multiple agreements.
I hesitate to give examples but they are not difficult to find in recent postings.

ophrys's picture

'despite pleas from real experts'

There are no 'real experts' saying any such thing, are there?

Don't forget, too, that iSpot is a site for all...children included, so it seems a bit churlish to worry about blind agreements. It does not matter if people agree with a wrong ID...eventually the correct ID will win out.

Ian
_________________

My Flickr photos...

http://www.flickr.com/photos/52163027@N02/

dejayM's picture

Experts and children

Oh yes Ian "despite pleas from real experts."
is here http://www.ispot.org.uk/node/319189
it is quite explicit - 'this is NOT a redwing'
But this is off-subject and I apologise if I have taken us there.
Your point about children is taken to heart by me and I should have been (will be) more cautious.
But the topic here is supposed be Agreements and my remark is intended to alert Admin to the possible need to direct agreeers (is that a word?) to Notes and Comments before agreeing, otherwise they might blindly agree without weighing up additional evidence and expert opinion.
"Before agreeing, look at ID notes and comments, if they exist".
Derek

chrisbrooks's picture

Experts

Hi, as this was my comment I should just point out that "experts" have yellow badges, I am far from an expert but knew on this occasion this was NOT a Redwing. Reputation points should not be confused with expertise as a general rule.

Thistle's picture

I was one!

I was one of those who agreed with the initial ID of Redwing. Until I came across this thread I was unaware of the comments which I guess were made after I added my agreement.

Why did I agree? Because, given my limited knowledge, I thought the ID was correct. It is different from most Redwings but I didn't think it was so far out as to be impossible. In my ignorance I didn't know there was such a thing as a Black-breasted thrush.

Having been back to the thread and read the comments, I am happy to remove my agreement to Redwing. As already noted, I'm not familiar with the Black-breasted thrush, so I'm not going to agree with that ID.

Getting back to the original topic, I agree with something if I think it is correct. I readily admit that I can (often?) be wrong.

Perhaps if the "Likely ID" is changed from the one that someone has agreed with, that could be flagged up in their "My Spot"? Not only would this help to reduce the number of incorrect agreements, it could also help the learning process of those whose apparent errors are pointed out.

Ian

dejayM's picture

Mistake

Yes, Ian.
It was not my idea to put people (anyone) on the defence and, like you I have made this sort of mistake (if it is one).
But blind agreements, I have noticed, are few-and-far between (blind posting is increasing - ha!)
Flagging is already in place for anyone who contributes an ID, comment or agreement. I would expect all those who agreed to Redwing to have removed their agreement by now - you have, I would. But, you know, it all adds to the power of the site.
On your point of 'not familiar'. I often research, via my books, a friend or two and the dreaded Internet, those things I am not familiar with. I gradually become SO in a agreement, that I agree. Take this -
http://www.ispot.org.uk/node/320025
as an example.
In the Redwing case, I posted the Internet evidence BEFORE 5 of the six agreements were made.
Derek

Thistle's picture

Ah!

I thought that there was no flagging for agreements. I see that I am wrong and that I've simply got it switched off. I'll switch it on and see how it goes but I'm a little wary of getting lots of pretty redundant flags.

Ian

dejayM's picture

bombarding....

It'll drive you crazy - so do less agreeing; be selective on the best or most interesting posts with good ID Notes!! Ignore blind posting!
http://www.ispot.org.uk/node/319702
Oh dear, what HAVE I said....?
ðerek

Thistle's picture

Crazy?

Too late to drive me crazy ... I got there long ago ;)

I've now got 1500-2000 flags to clear :( See http://www.ispot.org.uk/node/321564.

I really don't want to stop agreeing. Agreements are part of the lifeblood of iSpot and are perhaps particularly important for those who don't have the inclination / time / knowledge / confidence / etc to add notes to their observations.

Off to be an agreeable chap again!

Ian

chrisbrooks's picture

Agreements

I agree with Ian, without agreements iSpot would cease to function.

chrisbrooks's picture

ISpot

I have read the above comments with interest. It's probably best to keep it simple, if you agree then agree and if not then don't.

I think iSpot can be what you want it to be, each to their own. It's a great learning medium for all, experts alike. As for multiple postings, some just like to showcase their images and why not.

At the end of the day does it really matter what each individuals agenda is, just enjoy a great website. All postings also add valuable information to national recording statistics.