I mean I am not bemoaning you I am bemoaning the system. Sorry. I already made all my observations totally free use for these sorts of reasons I just hate how it all works
2 Likes
No problem and sorry for being over-defensive! I actually run a LERC and it’s quite frustrating. We can of course approach individuals and organisations but when we’re trying to aggregate tens of thousands of observations from multiple sources with only the benefit of nature in mind, be it trying to flag the presence of a protected species or inform a local strategy, it becomes quite the challenge and in reality not feasible.
3 Likes
@sbushes So if Seek is another app from iNaturalist why would it be any better than the Autosuggest feature in the iNaturalist app? What is the CV demo page? What is an RG ID? Sorry if I’m asking basic questions here!
upload page
Don’t know hat you mean by RG ID, RG is Research Grade status of observation.
Seek is a separate app which only uses Computer Vision to suggest IDs, no community involvement. It is possible to make observations via Seek and post them to iNaturalist, but by default they stay on your device.
I don’t know what the “CV demo page” is.
RG does not refer to a single identification, but to an entire observation. It stands for Research Grade, though that is a bit misleading as any observations can be used for research. Essentially, RG means there is community consensus on an identification. The basic way to get there is for two people to ID an observation as the same species, but it gets more complicated if there are disagreements. It’s explained here (I recommend checking out the rest of the Help page too, there’s some useful information there).
2 Likes
This is not true. iRecord first started importing observations from iNat in 2018/2019. This stopped at the end of 2019, but then restarted in 2021.
Again, this is not really true. The central purpose of iRecord is to act as a data warehouse, so it has been deliberately designed to take data from many different sources (as well as from its own web and phone portals). Every imported record passes through an independent verification and cleansing process, which will detect the vast majority of duplicates and reject them. Such rejected records will simply remain in the iRecord databases, and never be passed on to e.g. NBN.
The import of observations from iNat certainly does represent a challenge for iRecord in terms of scale. But the above-mentioned trial in 2018/2019 has already proved that the system is robust enough to deal with large numbers of duplicates from iNat (and a similar story can be told for data coming from iSpot). Even more importantly, it should be stressed that the iRecord verifiers are under no obligation to process records from iNat (or indeed from any other source) - and in fact, several verifiers have publicly stated that they intend to ignore all records coming from iNat, regardless of their apparent quality.
This was discussed at length during the 2018/2019 trial and the feedback from the site admins was that it does not present a huge problem and users shouldn’t worry about it. Personally, I still use both sites and I know many others do the same. I would much prefer to use only iNat - but that simply isn’t a viable option at present, because the integration between the two systems is far from complete.
The bottom line is this: if you use iNat alone, it will greatly decrease the chances of your records being accepted on iRecord. I mentioned one reason for this above, but another major issue is the different taxonomies used by each system. At present, iRecord will simply exclude any RG records for which it can’t find a match in the UK Species Inventory (and that is a fairly common occurence).
Back in 2018/2019, when I first discovered that my iNat records were being imported into iRecord, the potential issues with duplication (amongst other things), bothered me quite a lot. For a while, I seriously considered abandoning iRecord altogether. But I eventually came full circle and decided that continuing to use both was the only realistic approach. Currently, the only way to guarantee that all my records (regardless of their status on iNat) get stored in the iRecord databases, is to enter them directly via one of their own portals. My reasons for using iNat are largely orthognal to that, and, for the moment, I can’t see any good reason to stop using both.
2 Likes
For me I don’t see it as any better (or worse) in terms of using the autosuggest to get an ID.
That was my point. However, as @cthawley says, the AI alone certainly shouldn’t be relied on for a definitive ID … Although it may provide a useful starting point or placement for iRecord.
It’s not so simple, sadly. We have no idea which records are or are not imported and there is no easy way for iNat users to view them in iRecord or see the feedback from expertise as one would with direct upload. Choice over whether or not to import depends on the scheme and some are resolutely against using the system. Also some schemes are simply not even active on iRecord and prefer spreadsheet.
The best option I think is to ask the schemes directly what they prefer.
I hear what @bazwal is saying re: just adding duplicates regardless, but I think it’s best not to cause umbrage if we can help it. Some scheme leaders are already frustrated.
3 Likes
Dang, that sounds like a mess. Sorry they haven’t been able to design a better system!
2 Likes
This is a “problem” that is entirely of iRecord’s own making. If verifiers and scheme leaders are frustrated, they should direct their concerns to the site admins. It is not the responsibility of individual users to deal with such issues.
Before the previous trial in 2018/2019, the terms of use asked users not to “submit records to this site that have already been submitted to … another online wildlife recording system”. This statement has since been removed, and has not been replaced by any similar language or conditions. Clearly, the site has made a conscious decision to place no restrictions, implied or otherwise, on how and where users choose to upload their data. And that seems entirely right and proper.
2 Likes
Thanks for the advice @deedb8 . I have in fact got in touch with my Local Environmental Records Centre and they have said that they would encourage me to enter my records on iRecord even if they have already been entered on iNaturalist. But they also recommended creating an activity for the site I’m interested in on iRecord so that records could more easily be retrieved, and this seems to work well. Encouragingly they say “Your records should be taken into account should the site come up for development”. So here’s hoping!
2 Likes
See the UK update on the usage of iNaturalist observations within iRecord in the latest NFBR Newsletter by Martin Harvey - http://www.nfbr.org.uk/sites/default/files/newsletters/NFBR%20Issue%2062.pdf
Extract here:
How iNaturalist users can help iRecord verifiers:
There are a number of steps that iNaturalist users can take if they wish to make their records as compatible as possible with the UK recording schemes:
• Choose an open licence for your records: CC0 or CC BY will enable your records to be used as widely as possible; CC BY-NC (non-commercial) can pre-vent records being used by some schemes and records centres. Other licence choices (such as SA and ND) are difficult to interpret for individual records, and cannot be used in iRecord or the NBN Atlas (nor on GBIF). More info is on the iNaturalistUK blog (https://uk.inaturalist.org/blog/58298-licenced-to-share).
• Provide your real name if possible; this can be added as the “Display name” in your iNaturalist profile, and will then be used as the recorder name on iRecord
• Avoid obscuring locations unless absolutely necessary, as this can prevent them being linked to grid references of suitable precision for recording scheme use
• Now that the record import is in place, it is helpful if you can avoid adding the same record to both iNaturalist and iRecord, to avoid duplication of both records and of verifiers’ time
2 Likes
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.