User reports inaccurate toals

[steveraden]

I was forwarded this note from Barbara Thiers via Guralnick.

darryluk, one of our top transcribers, is upset because he says his totals (Macrofungi) are not being recorded accurately. He says he messaged twice on this.

I will post details here as I obtain them.

[joanball]

The problem is still occurring and daryluk is still complaining about it. Any updates on this issue?

[steveraden]

I just saw the details of his earlier complaint on talk.notesfromnature.org and messaged him there http://talk.notesfromnature.org/#/boards/BNN0000002/discussions/DNN00001o0

The details on his fungi totals are probably as described here but confirming through above. http://talk.notesfromnature.org/#/boards/BNN0000002/discussions/DNN00001mm

I'm going to take a look later today to see if it's easy to see confirm what's gong on.

[robgur]

Thanks, Steve. I think you are likely to quickly track down the problems here.... On Dec 18, 2013 2:13 PM, "sraden" notifications@github.com wrote:

I just saw the details of his earlier complaint on talk.notesfromnature.org and messaged him there http://talk.notesfromnature.org/#/boards/BNN0000002/discussions/DNN00001o0

The details on his fungi totals are probably as described here but confirming through above. http://talk.notesfromnature.org/#/boards/BNN0000002/discussions/DNN00001mm

I'm going to take a look later today to see if it's easy to see confirm what's gong on.

— Reply to this email directly or view it on GitHubhttps://github.com/zooniverse/notesFromNature/issues/307#issuecomment-30870957 .

[steveraden]

The user may have a bug on sending classifications or getting them reported now, but it's a little contradictory. What's constant is that his overall total is the same as what it on 12/11 when he said that it was pegged at 7750 However, the one archive sub-total has incremented since he reported a problem and Barbara looked at it.

From looking at the database, I can confirm that 7750 is his transcription total with a last transcription date of 11/27.

However, he is sub-total for fungi is 122 He also said (11/25) that his displayed total for the Fungi archive was only 94, but Barbara confirmed (11/24 file) that he had 103. This subtotal is now 122. Given his prolific activity a jump in 20 records (final count for that archive) from 11/24 to 11/27 is pretty feasible.

So I have to discuss with colleagues the best way to try to debug this further, but we may need to monitor a transcription from the user. I'll reply back to the user in talk once I understand precise things we may ask him to do.

[chrissnyder]

Does your transcription counts go up if you send one?

[joanball]

Not sure if you check discussion forum or get email updates, but darryluk responded to your last question:

darryluk: The totals had moved on a little since the 94 on macrofungi, and they add up correctly to 7750. But whatever I do they won't increase any further. So all the totals stay the same and therefore the 7750 stays the same. I haven't done any for a while, just waiting for this to be sorted.

[steveraden]

Didn't see his update. We're going to need to follow up with him next week and try to get him to help us debug his problem.

On Fri, Dec 20, 2013 at 5:49 PM, joanball notifications@github.com wrote:

Not sure if you check discussion forum or get email updates, but darryluk responded to your last question:

darryluk: The totals had moved on a little since the 94 on macrofungi, and they add up correctly to 7750. But whatever I do they won't increase any further. So all the totals stay the same and therefore the 7750 stays the same. I haven't done any for a while, just waiting for this to be sorted.

— Reply to this email directly or view it on GitHubhttps://github.com/zooniverse/notesFromNature/issues/307#issuecomment-31050356 .

[chrissnyder]

I know the site does correctly send back classifications, so it's not a systemic problem. It's also highly unlikely to be an issue with the API, as other users (such as myself) correctly see classification counts increment.

If we could ask darryluk to open his browser debugger and see if any errors occur in the process of transcribing, that would be useful. Possible that his particular setup causes an error to raise that doesn't result in any obvious UI failures, but stops classifications from sending? Again unlikely, but worth asking.

[steveraden]

okay, I'm going to talk to him on talk for this.

On Mon, Dec 23, 2013 at 8:50 AM, Chris Snyder notifications@github.comwrote:

I know the site does correctly send back classifications, so it's not a systemic problem. It's also highly unlikely to be an issue with the API, as other users (such as myself) correctly see classification counts increment.

If we could ask darryluk to open his browser debugger and see if any errors occur in the process of transcribing, that would be useful. Possible that his particular setup causes an error to raise that doesn't result in any obvious UI failures, but stops classifications from sending? Again unlikely, but worth asking.

— Reply to this email directly or view it on GitHubhttps://github.com/zooniverse/notesFromNature/issues/307#issuecomment-31122637 .

[joanball]

Not sure if you saw his responses to your questions, but he replied yesterday...

http://talk.notesfromnature.org/#/boards/BNN0000002/discussions/DNN00001o0?page=2&comment_id=52ceeceb383cd709b7000510

[steveraden]

Debug notes from user darryluk show failure with api proxy.

Uncaught TypeError: Cannot set property 'badges' of null application-2013-12-19_20-59-11.js:1 Not filling because tld mismatch between https://api.zooniverse.org/proxy and notesfromnature.orgonloadwff.js:79 Not filling because tld mismatch between https://api.zooniverse.org/proxy and notesfromnature.org onloadwff.js:79 Not filling because tld mismatch between https://api.zooniverse.org/proxyand notesfromnature.org onloadwff.js:79 Not filling because tld mismatch between https://api.zooniverse.org/proxy and notesfromnature.orgonloadwff.js:79

That's all I got in the console, but it appeared when I was logging in nothing came up when transcribing an entry. I did a calbug one and the total went to 4813, then when I went to 'My Account' it was back on 4812 again where it has stayed! I don't know if this helps, it's confusing!

18:34 Greenwich Meantime (UK) Google Chrome Version 31.0.1650.63m running Windows 7 Home Premium Service Pack 1

On Fri, Jan 10, 2014 at 11:45 AM, joanball notifications@github.com wrote:

Not sure if you saw his responses to your questions, but he replied yesterday...

http://talk.notesfromnature.org/#/boards/BNN0000002/discussions/DNN00001o0?page=2&comment_id=52ceeceb383cd709b7000510

— Reply to this email directly or view it on GitHubhttps://github.com/zooniverse/notesFromNature/issues/307#issuecomment-32048771 .

[steveraden]

We asked him to change a setting in Chrome

Settings > Advanced Settings > Content Settings > Cookies Section and verify if “Block third-party cookies and site data”

If this is set, this alone would cause.

On Fri, Jan 10, 2014 at 11:54 AM, Steve Raden steve@zooniverse.org wrote:

Debug notes from user darryluk show failure with api proxy.

Uncaught TypeError: Cannot set property 'badges' of null application-2013-12-19_20-59-11.js:1 Not filling because tld mismatch between https://api.zooniverse.org/proxy and notesfromnature.orgonloadwff.js:79 Not filling because tld mismatch between https://api.zooniverse.org/proxy and notesfromnature.org onloadwff.js:79 Not filling because tld mismatch between https://api.zooniverse.org/proxyand notesfromnature.org onloadwff.js:79 Not filling because tld mismatch between https://api.zooniverse.org/proxy and notesfromnature.orgonloadwff.js:79

That's all I got in the console, but it appeared when I was logging in nothing came up when transcribing an entry. I did a calbug one and the total went to 4813, then when I went to 'My Account' it was back on 4812 again where it has stayed! I don't know if this helps, it's confusing!

18:34 Greenwich Meantime (UK) Google Chrome Version 31.0.1650.63m running Windows 7 Home Premium Service Pack 1

On Fri, Jan 10, 2014 at 11:45 AM, joanball notifications@github.comwrote:

Not sure if you saw his responses to your questions, but he replied yesterday...

http://talk.notesfromnature.org/#/boards/BNN0000002/discussions/DNN00001o0?page=2&comment_id=52ceeceb383cd709b7000510

— Reply to this email directly or view it on GitHubhttps://github.com/zooniverse/notesFromNature/issues/307#issuecomment-32048771 .

[joanball]

Another user (wreness) has just reported inaccurate totals. He sent me a private message saying:

"have done about 160 records. i did 100 yesterday night. The total was on 36 when I signed in, but went up to 159 when I stopped. Today it's back down to 36. It's tallying them up as I do more, so now I'm at 38, but if I sign out, then back in, it's down to 36 again."

@sturad , Is this the same problem?

[steveraden]

So we have a strong suspicion that darryluk's problem has to do with a browser extension, Last Pass. (based on the error message we saw that referenced onloadwff.js, a Last Pass file). He confirmed he's using it, and said he would get back to us to retest.

[steveraden]

Haven't hear back from him. Going to assume that LastPass was the culprit and close the issue.