Amadan's picture

Use of "Get Recommended" shown?

It's a bit embarrassing if you click on "get recommended" when posting an observation, only to find yourself corrected by an expert because the systematic name iSpot provided is apparently incorrect. I'm not talking about wholly incorrect identifications, just those where iSpot appears to be (like me) behind the taxonomic times.
I'd feel a bit better about it if there were an icon on the observation to show that I had followed the simple "Get Recommended" protocol: or is it a reasonable expectation that observers will check the systematic name offered?



ophrys's picture


I would relax and not worry about it. Martin Harvey has posted at length about this, and the suggestion is that you go with what iSpot is using at the time. I used to correct scientific names regularly, but I can see what Martin says and now just go with what comes up. The commonest to be updated seem to be Lesser Celandine, Stonechat or Black-headed Gull, but you are OK with using the Get recommended button. I think people would be better adding the updated name in the comments rather than as a new ID.

No need to feel embarrassed!


My Flickr photos...

Masked Marvel's picture

Expert editing

This is where I think it would be good to allow experts some limited editing power to correct things like this without causing embarrassment of causing people to lose reputation points when they got it right.

Dioctria's picture


That would certainly be better than a new ID, but it doesn't stop the same thing happening again to the next user. Which must be just as frustrating for the experts having to make the same revision over and over again as it is for new users feeling let down by the system.

A new name in the comments would be less awkward but then we have the old problem of comments not being noticed.

Taxonomy is in a constant state of flux so perhaps we just need to accept that the NBN database is the authority for iSpot.

If the database is particularly behind the times then maybe the experts could take it up with the NBN? Or iSpot could maintain an extra database of deviations from NBN which would be checked during the Get Recommended process?


Syrphus's picture

The NHM dictionary that iSpot

The NHM dictionary that iSpot uses is regularly revised and updated, and the current taxonomy and nomenclature is visible on NBN Gateway. However, other users of the dictionary (e.g. users of R6 and iSpot) need to update their copy with each revision. The current version used by iSpot is not the most recent, so that does cause some problems as it contains errors that have already been recognised and corrected.

It seems the updating the dictionary on the iSpot server is a major task, hence it has not been done as often as perhaps is desirable. It is an iSpot matter, not an NHM one.

The primary reference on names should be the current NHM dictionary, but in practical terms inexperienced users should go with the 'get recommended' button and should not be corrected for e.g. using Larus rather than Chroicocephalus or one of the older names for Stonechat. My view is that if the binomial used is unambiguous and correct, it should be left.

If the correction is for a mis-spelled binomial, that is a different matter, and a correction is appropriate. If taxonomy (rather than nomenclature) moves on, as it has in the Formica rufa ants and the Bombus lucorum bees it is different again, and my approach is to revise e.g. F. rufa to the genus on the grounds that the binomial is ambiguous and the species ID is not possible. I hope that folk see that as educational, but I can understand that some may object. It is not very different in nature from the posts of female 'Panorpa communis', which are routinely and correctly revised to genus.



recording wildlife with The Recorder's Year on

Dioctria's picture


Apologies if it sounded like I was pointing a finger at the NHM, that wasn't my intention as I don't know the detail. I was just trying to suggest that solving the issue among the expert/professional community might be preferable to making new users feel they are wrong. I see you are of the same mind and I agree with your other points too. (Including your quiet correction of my loose use of taxonomy vs nomenclature! :) )

On the technical side, whilst I can see why iSpot might want to maintain its own copy of the NHM dictionary for fast look-up on auto-complete, 'Get Recommended' sounds ideal for a web service at the NBN end which would solve the stale data issue at a stroke. iSpot could always fall back on its own dictionary if the NBN site was down. Although I imagine this might raise its own issues with funding and cross-organisational development.


miked's picture

I think the web services idea

I think the web services idea has been actively explored for some time although possibly not likely to be implemented until the summer when there should be a range of updates on ispot site

Dioctria's picture

Thanks Mike, good to hear

Thanks Mike, good to hear it's a possibility.


Amadan's picture

Thanks for the comments -

Maybe an alternative type of revision would be useful, but I expect that would be quite onerous to program.
I'd noticed the frequent exchanges about BH-Gulls - as noted above, it must be frustrating for the experts too.
Taxonomy can be taxing...

Refugee's picture

Another way

What we might be able to have if we ask nicely is an expert tab with another "group" under it where the name changes can be entered as and when they are found. This could be added to the Get recommended button on a second priority basis with the choice being offered. This would "tie up" observations with both old and new names. I have found one or two common names that do not produce a result from get recommended and as a result i have had to use search I-spot and paste the name from an observation that has a decent amount of agreement on it.
I have got back following a long period living in town and can at times get a partial latin name and google it and match it with pictures there in order to complete it. A bit slow and not often needed.