On the beta site, the "mini" progress graphs on the left side of each sub-item (Characters-writings, readings...) are missing now. Just a blank space.

My progress has been minimal for the past week, but not non-existent.

This occurs regardless of the view (week/month/year).

Hmm perhaps this was a temporary bug; I can't reproduce it now. I've been moving a lot of code around on the beta site so it was probably an issue with that that's since been fixed. Are you still seeing it? If so, what browser are you using?

I just did a "hard refresh" (ctrl-shift-R) with no change.
I'm using Firefox Aurora which is now Version 17. Last week was version 16 where I also had problems, but that only started happening about 2 or 3 weeks ago.

It works OK on IE9.

Checking into the page source, I find the IMG SRC (at chart.apis.google.com), and when I copy it to my address bar & press enter, it tries to switch to HTTPS, and then complains about an untrusted connection.
I used to have an add-on installed to switch to https whenever possible, but I do not see it on this computer, and I didn't (yet) find any option that deals with it.
I'll keep searching.

Maybe it's a Firefox bug, since the certificate says it should be valid on *.google.com. "(Error code: ssl_error_bad_cert_domain)"

OK, after more investigation, the *.google.com in the certificate can only match to 1 subdomain (apis.google.com) but not to a 2nd level. This is per the associated RFC.

if I change the URL from chart.apis.google.com to chart.googleapis.com (with the rest of the URL the same), it switches to https, but doesn't have certificate problems, since *.googleapis.com is covered by the certificate.

I still cannot identify why (or by whom) I'm getting redirected... might be a google setting or a Firefox setting. (in my IE test, when I log in to google, it doesn't redirect me, so probably a firefox issue, but haven't found the root cause!)

Thanks for changing the URL to googleapis.com! It's working quite well now.

No problem! Thanks for researching and identifying the issue, so it was really straightforward to fix. Right now it's only on the beta site, and will be on the stable site later this month.

Scott, further request completely separate from the above problem.
When you look at different graphs, the bottom right shows the retention rate in different colors depending on how good it is. (Red for under 80%, orange for 80 to 85, and green above 85%).
Can you change the color of the mini graphs to match those colors? Near as I can tell, it's just a matter of changing the "chco" parameter of the charts URL.

(I'd also prefer to see more color divisions (85-90, 90-95, 95-100), but 1 step at a time...)
The colors will change the aesthetics of the page, so maybe a checkbox or option for those who don't want it...

Hmm we can give that a try. My only concern is that it won't be immediately clear what the colors mean. Having forgotten more words than you remembered? At least with the larger graphs it's next to the number so that it's a bit clearer. Any others have any thoughts?

We do have a 95-100 division, which is orange because it's suboptimal. 85-95 is the optimal range in terms of how much you learn, so it doesn't make much sense to split them into two different colors since either side is about as good in terms of efficiency.

very true, and there isn't a nice way to put a legend or explanation anywhere. Perhaps "Colors by Retention: 0-80% 80-85% 85-95% 95-100%" with each range colored appropriately?
But I'm sure having the last range in orange will raise more questions than it answers since most people are not in that range (I assume SRS takes care of that unless they have their retention rate Set at 98%.

[perhaps leave the last range out, and answer the 1 or 2 people that inquire about it]

There are some who get items right often enough to be in that high range. The system will adjust, sure, but that doesn't prevent people from getting each thing right and garnering 100% retention.

We don't get a lot of questions about it, and if we did, we'd generally put the answers in the FAQ.

FYI, Firefox finally fixed this issue in the Aurora & Beta builds (https://bugzilla.mozilla.org/show_bug.cgi?id=786417).

The bug hadn't shown up in the current release builds, so those should still be OK.