Hi Matt,
No such thing as a 'quick question' as you can see from my lengthy answer below!  You've raised a good point and I hope this answers it: 

In the current system there is no control over full dataset downloads which occur as part of a site search or species search.    This is relevant to data providers who supply datasets for a single species, a single site (e.g. a National Trust site), or for a very small geographic area that falls completely within a 10km square or designated site boundary.   
There is currently a technical limitation on downloading records for large geographic areas such as LRC boundaries – the maximum size of a polygon for which data can be downloaded is 900km2.  However, under the new system, this limitation is unlikely to apply.
We therefore propose to add a separate control over full dataset downloads and to apply this to ALL requests for whole datasets, whether they occur as a result of a direct request or incidentally as part of a species or site search.  This will also apply if the request occurs via a custom download or via Web Services.
We propose that data providers would receive an email asking for permission to download their whole dataset and would have 21 days in which to accept or refuse the request.  We propose that permission should be granted automatically if the request was not answered within 21 days.  This is detailed in the flow chart: http://www.nbn.org.uk/nbn_wide/media/Documents/Data/Single-dataset-download-flow-chart-screenshot.pdf

The data provider will have the following options when setting ‘public’ access to their dataset:
a)    Allow download of the whole dataset without delay.  These datasets will automatically be available via GBIF and Web Services.   The data provider will not have to deal with requests for full dataset downloads, but will receive feedback on who has downloaded their dataset and for what purpose.
b)    Allow download of the whole dataset with a 21 day delay.  If this option is selected, the data provider will be given two further options:
b1)    Allow export of the dataset to GBIF
b2)    Allow an exemption to the 21 delay to enable the dataset to be accessed through Web Services
The reasons for these separate controls are:
b1)     It is possible to download full datasets via GBIF (although there is a current limit of 100,000 records per download) and it is not possible to set a control over this on GBIF.  Some data providers have advised us that they want control over who downloads their whole dataset via the Gateway, but they are keen to make their dataset available via GBIF.
b2)    It is not possible to integrate a 21 day delay into Web Services, so if a Web Service request for data on a single species or single site calls for a whole dataset, and the data provider has set a control over whole dataset downloads, that dataset would simply be excluded from the Web Service return.  Some data providers have indicated that they want control over who downloads their whole dataset for use in their own in-house database (due to concerns over duplication), but they have no objection to their data being made available via Web Services.  Indeed, many data providers are also users of Web Services.  Please note that this will only affect Web Services that supply details of the records. Web Services that just return map images, such as the gridmap Web Service will not be affected
Most data providers supply multi-species, multi-site datasets so their whole dataset could never be downloaded through a single species or single site data search via the Gateway or via Web Services.   We therefore hope that the majority of data providers will agree to let their data be accessed via Web Services.
From a user’s point of view, if you download records following a site search or species search on the Gateway you will be notified if your download includes any whole datasets for which the data provider has set a control over full dataset download.  You will see a list of these datasets, enabling you to apply for permission to download them, and you will also have the option of downloading the rest of the data immediately.
Under the proposed new system, you will be able to apply for access to multiple datasets with a single request, so applying for permission to download the datasets will not be a lengthy task.
From a data provider’s point of view, please note the following:
•    Records flagged as ‘sensitive’ will always be excluded if a dataset is downloaded at the ‘public’ level of access.
•    If full dataset download is permitted at the ‘public’ level of access (with or without the 21 day control) the user will only be able to download the data at the level of resolution that has been set for ‘public’ access.
•    The ‘enhanced’ level of access to a dataset will always include the ability to download the whole dataset.
•    If a user has ‘enhanced’ access to a geographic/taxonomic/temporal subset of a dataset, and if the data provider has chosen not to allow full dataset download at the ‘public’ level of access, the user would need to apply for full access to the whole dataset in order to download the whole dataset.  In practice, this is unlikely to be an issue, as if they are only interested in a geographic/taxonomic/temporal subset then they are unlikely to want to download the whole dataset.