I posted a problem at http://forums.nbn.org.uk/viewtopic.php?id=6600 the other day. The reponse from NBN Support seemed to imply that I was mistaken about the nature of the taxonomic information available on ALS. I am expanding on that problem now.

Two examples will demonstrate precisely why ALS is utterly unfit for purpose in allowing clarification of taxonomic problems, something which G5 does quickly and efficiently, and that making it fit for purpose requires vastly more than superficial fiddling - a complete assessment of the site and re-design of some major components of the coding is needed.

Example 1. An old name for a particular cuckoo-bumblebee is Bombus distinctus.  Someone extracting data from an old text, for example, might come on this name and not recognise it.  In G5, type ‘bom dis’ into the ‘Browse species’ page, and (if you are fortunate enough to hit a time when G5 is working), Bombus distinctus appears as #2 in the list, and you are immediately aware that the current name is B. bohemicus.

Do the same exercise in ALS by typing ‘bom dis’ into the box on the page headed ‘Search for taxa’, and we get five results, thus:

species: Hesperis matronalis L.  – Dame's Violet
Locality: Bom Beaver
Locality: School (dis)
Locality: Filton Airfield (dis)
Locality: Hillside Specialist School For Autism Spectrum Dis

One taxon, four localities. From a *taxon* search!

At the top of the results page, there is another search box with the string ‘bom dis’ helpfully entered. Click in that, and – wonderful! – Bombus distinctus appears at the top of a drop-down list. Click on that, hit Enter, and you get 33 results over 4 pages.  Not one of them indicates Bombus distinctus (remember – that is what I searched for!), not even under the name B. bohemicus.  To get any further (and there is absolutely no reason why anyone would consider it obvious to do) you have to click on the B. bohemicus link, then on ‘Names’, and there, at last, it appears under Synonyms.

I have to assume that whoever wrote the code for this function had well-thought-out reasons for designing this particular result. I am sure I am not the only one who would welcome being let into the secret.

Anyway, to cut to the nub, ALS is no help at all in clarifiying the status of the binomial ‘Bombus distinctus’.

***************************
Example 2. This relates to a muscid fly whose nomenclatural muddle caught me out a few years ago. In G5, type ‘neo cor’ in the ‘Browse species’ page. Two of the top three returns are just what we are looking for, as you would if that name was included in an R6 import:

Neomyia cornicina    Neomyia cornicina (Fabricius, 1781)
Neomyia cornicina    Neomyia viridescens (Robineau-Desvoidy, 1830)

In other words, the binomial refers to two totally different species, and it is fairly important to a data manager that the correct one is chosen. He/she can then do further checks if need be to establish which species, cornicina or viridescens, is the subject of the record.

Repeat the exercise in ALS. Searching for ‘neo cor’ gives 25 results over 3 pages, only 9 of them actually species, and no mention of Neomyia cornicina anywhere.  Try again with the box at the top of the results page, N. cornicina is at number 2 in the drop-down, search, and – Bingo! – that species appears at the top of the list of four returns. Its doppelganger, N. viridescens, is last, but there is no indication at all that cornicina is an old synonym of viridescens, and therefore no incentive for anyone to click on viridescens, go to ‘Names’, and see the confusion.

***********************

I have gone into considerable detail here, so that the severity of the problem is blindingly obvious to anyone, and anyone can repeat the steps for themselves.  In importing records via R6, this sort of problem occurs frequently, and unless conflicts are resolved, data quality is compromised.

Any further comment seems utterly superfluous.

Murdo