General Bugs & Enhancements
-
Link from UI to FAQ.
Show a flag by the "Discuss" button, to show if a record already has any discussion.
Faster update to the bulletin board: new posts and tags don't always appear particularly quickly.
The "Discuss" link always accessible (it isn't in Ornithology).
Posted
-
Hi Bonnie, I think you make some really good points here. I'm very glad that there's a programmer coming on board, and I like it that we seem to be getting more feedback on bulletin boards from scientists; these are all changes in the right direction. Even more would, like you say, be good.
Is part of the problem lack of funding/overload in the scientists' lives? It'd be great if someone were just able to check in once a day and comment on new stuff.
I agree about "transcribe once and validate" - it makes me uneasy. I'd feel happier if we knew some research had been done on the quality of the data that results.
Hey - just keep flapping, just keep flapping, what do we do, we just keep flapping...
Posted
-
by
robgur
scientist, admin
Hi everyone, sorry for the long silence. I recently moved positions, and along with that, I also moved across the country, and that and starting a brand new job (at the Florida Museum of Natural History) has left little time for, well, anything.
As you hopefully heard, we brought on board a new developer to work on Notes from Nature, and 20% of his time is devoted to fixing up what has been, as you ALL know, a less than perfect Zooniverse project. We hope to start by getting a new batch of Herbarium images up and online, and then we plan a set of fixes that I think EVERYONE will like, and that have been a long time coming. This includes dealing with counting and badges issues once and for all. And a lot more.
There is possibly even better news for Notes from Nature on the horizon, so stay tuned. Its been a long road, but I see some sunshine now, and again, thanks so much for staying invested and involved.
I have some further thoughts on the issue with "transcribe once and validate" --- I am a proponent of testing whether this will work and what the costs and benefits will be. We haven't been able to run those tests let alone get much of anything done to fix Notes from Nature, for many reasons, simply due to capacity. If and when there are some more available funds to get creative, I guess I think we ought to test it out. We want to do the thing that balances efficiency and quality, and maybe there is no perfect solution, but I'd like to test the hypothesis that can we find an optimum.
Posted
-
by
Mr._Kevvy
Thanks for the update. :^) Looking forward to new Herbarium images as I've been rather bored recently!
As well as tapping away here, I've been involved with BOINC-based distributed scientific computing for a long while. The system that was settled on for that in preference to many others would be best expressed as "transcribe twice, more if required". Every work unit is sent to two random volunteers. If their results agree, then the unit is finalized in the database as completed. If they don't agree, it's sent to a third volunteer and the process repeats until a maximum of ten. If there is no agreement by then it's either flagged for manual intervention or removed from the database (only the former would occur here.)
A similar system here may also prove to be an ideally efficient and still reliable solution.
Posted
-
by
am.zooni
Hurray! I'm bored too. I only started helping on the Herbarium project shortly before the last set of images was completed, and had just gotten hooked enough to move from 'I have a little free time so I guess I'll fill it by transcribing' to making a point to reserve a bit of time. After waiting a few weeks, I was just beginning to consider having another go at finding a different (active!) project to work on instead. This message will make me be patient a bit longer.
It appears there are very few of us working on this project. So please do what you can to keep any of us from wandering off to a different project or other interest during the hiatus!. An estimate of when the new images will be posted would be helpful, but even an occasional 'more images coming soon' message tells us that there's someone there paying attention, so thanks for this.
Interesting puzzle to determine the minimum number of transcribers and degree of agreement you need to finalize the results for an image.
It's more complicated to solve on this project, because you really need a bigger group of volunteers working on it before you run that kind of test. It's difficult to get meaningful test results if you only have handful of people whose output you can analyze. You need enough to get some randomness (or at least variety!) in the of pairs (or trios, quartets) of transcribers whose results are being compared.
Posted
-
Hello! @robgur, good luck with your new job! Great to hear from you - it was starting to feel a bit radio-silency over here.
I'm in a quandary... I was starting to quite enjoy the ornithology ledgers. Does that count as going over to the dark side?
Posted
-
by
Mr._Kevvy
in response to HelenBennett57's comment.
re: "I was starting to quite enjoy the ornithology ledgers. Does that count as going over to the dark side?"
For Ornithology...
Posted
-
by
robgur
scientist, admin
in response to Mr. Kevvy's comment.
I really like this idea, Mr. Kevvy. Passing it along to our new developer and the rest of our team.
Posted
-
by
robgur
scientist, admin
in response to HelenBennett57's comment.
Ha! The ornithology ledgers are less the Dark Side and more the Opaque Side. If you have that special blend of patience, willingness to deal with the less than ideal interface, the broken counting, and the extremely hard to decipher handwriting, well, you are among the (very) few who find Zen in that.
Posted
-
by
robgur
scientist, admin
in response to Mr. Kevvy's comment.
Ha! Wow!
Posted
-
by
HelenBennett57
in response to Mr. Kevvy's comment.
Hahahahaha!
Posted
-
Question/enhancement: how do you find a tag group once it has been created? I just created #snakes and now can't find it to follow...
Posted
-
by
am.zooni
I just clicked on one of the images you tagged with #snakes and when it opened up, on top center there was a section titled 'Collections', and snakes (with 3 images) was listed there. I clicked on that and when it opened, there was a 'follow' button in the upper right corner. So now I am following 'snakes'. But what's up with snakes anyway? Are there snakes in one of the projects?
Posted
-
No, but there are a few references to finding snakes whilst innocently botanising!
Managed it now... I think it was just a caching issue. [going to add 'faster update of forum' to wishlist] Cheers.
Posted
-
by
reddder
in response to Bonnie123's comment.
you & others have raised your points b4 w/no response from the bridge -- very discouraging. And, while it will have no effect upon my undying loyalty to NfN, I would like to know how many are participating in NfN. Maybe Ed Snowden could give us the answer?!
Posted
-
by
DZM
admin
in response to dmbrgn's comment.
I've tried to comment where I can, @dmbrgn -- I can't speak to the science, but I can definitely help you here! 😃
There have been over 800,000 total classifications by over 7,500 total users.
There are currently about 50-90 users each week, and 2,000-3,000 classifications done each week.
Does that help?
Posted
-
by
md68135
scientist
@dmbrgn -- I (we) are working to be more responsive, we know we can do a better job at that! Thank you for your efforts here.
Posted
-
by
reddder
in response to DZM's comment.
Yes, it certainly does help. Thx for filling in the blanks.
Posted
-
by
reddder
in response to md68135's comment.
Gr8 -- we're in this together.
Posted
-
by
CharlotteBio
I have just been dumped/dumped myself out of the second page I was doing today 😦. Is there any way to retrieve them so that I can finish them? PLEASE....
Posted
-
Left this in another topic but it would go here too... (problems bugs)
- The input layer/pane sits directly in front of the specimen picture. which means we have to move it around to see the label on the specimen.
- Not back or restart in the input layer UI incase you need to correct something.
- There needs to be a re-display of the inputs for one last check prior to submission and allow for corrections. Or make the input boxes all visible, we fill them out make sure its correct and submit.
- where do multiple collector names go?
- For other plant groups and fungi, multiple specimens are often in the same collection specimen. How is that suppose to be dealt with.
Posted