Moderators: please read
-
by SandersClan
Hi! Maybe you can sort out a small dispute regarding educated guessing. As you can tell, lots of us are pretty committed to NfN and want to make sure we do what's best for the project. This educated guessing subject has come up several times. Will you please look at the comments for EMEC662408 in 'Recent' and let us all know the best course of action? Thanks so much!
Posted
-
I'm all for educated guessing, as long as I get to decide which guesses are correct! LOL!
Seriously, I still think there need to be better instructions/guidance in the areas where most of the guessing is happening. One that is still a problem is multiple labels and differing info on them. I've seen it go from; enter it all separated by commas to just stick with the original label and ignore the others. I tend to stick with what is obvious/legible and use the "Skip this field" function on what I can't make out. I don't know how many times each specimen is examined/transcribed, but guesses may do nothing more than create a pile of garbage. However, a majority of "Skipped" fields on the same specimen may be much more pleasant to work with. Just depends on the sense of smell of the scientists, I guess (no pun intended, of course!) 8)
Posted
-
by joanball scientist
First, I have to apologize for the late responses. Just about everyone from Calbug is away in the field right now for an Essig Museum Collecting trip! But will start returning later this week. So, it will get much better! I also suggest adding a hashtag #moderator for issues you think need a moderator's attention.
About interpreting the locality information: Our initial plan was to have locality information entered verbatim. This is because people will inevitably interpret (the extremely variable) locality information differently and we will end up with multiple entries for the same location. However, we also recognize that many of you enjoy and are very good at interpreting the information. So, we will work on a way to accommodate this, maybe with another field. That way, we could use any interpretations that you offer to help with our final interpretations during the next stage where we assign latitudes and longitudes to these descriptions. This is called georeferencing, and we will write a blog about it soon!
For now, please enter the locality information verbatim, but keep looking up the counties when they are missing or filling in the country and state information through your own knowledge or google searches. Warning: keep a lookout for localities with multiple potential counties. When this happens, leave out the county!
We will keep you posted through blog entries when we make changes to the interface that hopefully will help clarify these issues that keep coming up (like interpreting and the host field). There is a delay in changing things with the interface right now, with people away and because the scientists and the website developers are different people! But, we'll try to make this happen as soon as possible. And, we welcome any suggestions.
Thank you so much for your help and good work!
Posted
-
by SandersClan
Thanks, Joan!
Posted
-
by geckzilla
Thanks for that, Joan, especially the part about multiple potential counties. I have been very confused about that.
Posted
-
by geckzilla
What about when things are misspelled?
Posted
-
by joanball scientist
I think it would help to correct misspellings when you are certain of it.
These are all really great questions. We will use all in efforts to clarify data entry help text and improve the fields we are using.
Posted
-
by SandersClan
I think my big question regards abbreviations. For instance, if we know beyond doubt that 'Cr.' is 'Creek', do we leave it as is or write 'Creek'?
Posted
-
by joanball scientist
For now, write abbreviations in the locality field exactly as-is (i.e. "Cr.").
But, If there is an obvious spelling mistake that's not an abbreviation, such as Santa Barbra and you know it should be Santa Barbara, then you could correct that. The bottom line is that we are trying to figure out the best way to do things so that we have less work to do in processing the data later. The locality field is obviously the most variable, with so many abbreviations, etc... which is why we think that entering verbatim for this field will generally make things easier in the end.
Posted
-
Thanks for the info! ๐
Posted
-
by nosenabook in response to joanball's comment.
Oh good because that's what I've been doing - as is.
Posted
-
by Pict
Thank you Joan
Posted
-
by SandersClan
Thanks!
Posted
-
by jamac41
When you write the collation programme for the data, will it be ignoring full stops and spaces? There are some things (such as collector names in CalBug, especially handwritten ones, or cramped location entries) where spaces are ambiguous, and I was wondering if we need to do our best to work out where spaces exist, or whether it will all be irrelevant come the processing stage.
Posted
-
by geckzilla
I'm not saying that's how it's done but that's how I would do it. Remove the spaces and punctuation and do it case-insensitive. a and รก for instance would also be equivalent.
Posted
-
by robgur scientist, admin
We are just about to get the second data dump of transcriptions and how to best process those has been a topic of conversation among the team here. The short answer is that we don't quite know what to do yet but will be working on some processing scripts etc. over the next weeks/months to help get the highest quality consensus outputs. We will also be blogging a bunch (we hope) about how that works, etc. Ideas and thoughts gratefully received!
Posted
-
by Dragonwalker
I have been typing verbatim, however one had R.R. Cyn. I looked through other records and noticed that when they were entered by whom I assume to be moderators it was typed as Railroad Canyon, so in the NOTES section I made a note, so that when someone has to add the official record they will know what it stands for without having to look it up. Also there was an instant where there was no collector information, however after some research I found that there were two collectors at that location that day on record, so I added this to the notes as well. Not sure if this is okay, but maybe having an additional notes section for us to be able to add information that isn't as clear on the labels but is found with further investigation could possibly expedite the work of those that are going to be finalising the records. It does not seem to make sense to have some that say R.R. Cyn. and then Railroad Canyon in others so I assume that this will be corrected anyway at some point.
Posted
-
by Dragonwalker
in addition, one thing is how to format names. I have been writing it as J E Powell for example. I could have done J. E. Powell, or J.E. Powell. I saw another person in the notes type it the first way, so I decided to stick with that as it is clear. I definitely suggest an update on how to write the names at least. Thank you project scientists for letting us help you!!!
Posted
-
by rhiannonHB
I do think that a set of conventions and basic instructions would be extremely helpful in standardising data where possible and to save time in answering what seems to be multiple queries appearing on similar topics.
Posted
-
by reddder
Unless I'm mistaken -- & i have been in the past, believe it or not -- at the rate the work is getting done on Calbug, it could take many years to finish the project. Your comments would be appreciated.
Posted
-
by bretarn
I stick as much as possible to original abbreviations, when it comes guessing about a location, I'm using Google o find out, and, by this way makes me travelling through your beautiful country.
๐Posted