Tuesday, 28 January 2014

Background work

As I mentioned in my previous post, I have started work on my Donegal Wildlife database, and I now have around 200 of my 1500+ species in place. The structure of the database has changed a little due to feedback and also due to my experiences putting the site together, and the work (which effectively summarises over 11 years of recording) has taught me a few things:

1) I have a lot of photographs
2) Over the years I have progressed quite a bit and I can now identify species that I couldn't identify previously. This means that some of my old 'unidentified slug' pictures can now get a name. The by-product of this is that I'm actually increasing the number of species in my work, so my species total has increased without me leaving the study.
3) I'm now understanding taxonomy more clearly than I had before.
4) Some species had slipped through the net and although they were documented, they were missing from the species list. So again, my species list has grown a bit. (The obscure side-effect of this is that I can start working on the project, add a few species and then discover that I still have a higher number of species to add than I started with.)
5) Some of the formal names for species have changed in the past 11 years. This is to be expected as we find out more and more about how species are related, but it means that this kind of work will need to be updated on a fairly regular basis.
6) I still haven't decided how to document unnamed hybrids (thinking of Willows and Dactyls particularly, here).
7) Google takes around 10 days to reach a site once it has been informed that it exists.
8) I still have huge gaps in my coverage, especially in relation to spiders, bugs and beetles.

I now have a dilemma: I feel the strong urge to share what I have done so far with a much wider audience, but I am acutely aware that I'm only 13% through the work, so there are still many more gaps and blind links than entries. So far I have 'completed' butterflies, mosses, sawflies, bees, wasps, leafminers and molluscs, and made a start on trees, flowering plants, animals, birds, fungi and moths, with a scattering of other partially-completed groups. Groups not yet started include flies, beetles and lichens.

Access methods include:

1) Deciduous Woodland (as an example of "search by habitat")
2) Common names in groupings (e.g. Butterflies, Sawflies, Leafminers, Fungi)
3) Taxonomic tree, which is complete for every entered species, since that is how the species are stored in the database.

There is also a text search, which lags about 10 days, so that it isn't fully useful during the development (but will be extremely useful when the site is stable). And finally, there is a 'random page' button that takes you to a err...random species in the website.

The website also contains a guest book for comments and a working forum to discuss the site and wildlife in general.

Please use the comment facility on this blog to let me know what you want, and I'll decide how to proceed based on that.


Peter Archdale said...

Sounds a huge task Stuart, but the result will be worth it I'm sure. One area you might eventually consider is linking to a GIS. This would allow people to search by locality. I realise this may be a huge additional task to complete, but maybe you can build the functionality into the database and then add the interface anon.

I'm considering how to do this with my own info, but am at very early stages. As well as the species info I plan to have fixed point photography so that I can access views as my site changes.

stuart dunlop said...

It is a huge task Peter, and I reckon it will take perhaps 2000 hours of work in total. You're not the first to mention location: another suggestion is to access the data by townland. I'm considering a number of options, including links with e.g. google maps.

Gill said...

Fantastic project, Stuart. Keep up the good work.

"I still haven't decided how to document unnamed hybrids" um, "willow hybrid" or "spotted orchid hybrid" ? :-)

I'm intrigued to know how you add a text search to a site.

stuart dunlop said...

Gill: the difficulty is how to build a taxonomic route to each hybrid. Maybe I should put a 'spare' entry at the end of each relevant taxonomic genus with a "species page" called 'genus-hybrids', where genus can be Salix, Dactylorhiza, etc. That would work.

In terms of text search, Google has a little-used facility which restricts a google search to a single domain (website). The format of a typical search is:

site:donegal-wildlife.blogspot.com moth

That google search returns a list of every page on this blog that contains the word 'moth'. Try it!

All you need is a script on your site that performs that command.

I can't add the script in this comment (scripts are barred for good reason), but you can get the script from the index page of the new site.