Fae brought this up: https://bugzilla.wikimedia.org/show_bug.cgi?id=67477
Clearly, mailing the hard drive would be the most efficient way of
transferring the data. If the hard drive can make it to close enough to our
data center's network, anyway, which might have security implications. Is
there any precedence to doing something like this?
Here’s an update on the Media Viewer project, with new metrics and remaining issues.
1. Weekly metrics:
Global disable events - logged in users 37 (3%)
Global disable events - logged out (anons) 1,126 (97%)
Total disable events 1,163 (100%)
See the new spreadsheet for disable metrics (1) — and its corresponding opt-in/out dashboard (2).
These results were collected on June 30, 2014. Since we just rolled out this new metric and dashboard, we don’t have any data to compare it to yet. And we don’t know either what’s an acceptable disable rate for this type of feature. For now, here’s one way to put these numbers in context: total disables represent 0.0049% of image views that day.
Comparing English and German disable rates is also informative, as shown in the spreadsheet: English users are disabling a bit more than German users as a percent of image views (0.0064% vs. 0.0048%). It’s also worth noting that a larger percentage of English logged-in users (10%) disabled Media Viewer, compared to their German counterparts (2%).
This new metric gives us another view on overall acceptance of this feature. It is not perfect, but seems more objective and less time-consuming than the short-term surveys we used to collect feedback in recent months.
As discussed, we have now removed Media Viewer surveys in most languages — except for English and German, which will be turned off next week.
To wrap up our survey reports, we added the last weekly approval rates as of June 23 on this updated worksheet (3). Overall, we observed an upward trend on most wikis last week -- especially on the English Wikipedia, with weekend approval growing to 52% (vs. 38% ‘nos').
These surveys provided invaluable feedback to identify the most important issues across all user groups. Now that we have addressed most of these issues, we plan to end these surveys next week.
2. Top outstanding issues:
* More info above the fold (3%) > click on truncated info to see more
* More image sizes (8%) > add more sizes in download panel (we are doing a quick survey now to better understand what’s needed)
* iOS issues (iPhone/iPad) (1%) > we’re punting on these, because the new tablet redirect only impacts a tiny user group now
3. Top issues addressed this week:
* Want full resolution images (2%) > show tooltip when they click on images
* Show more attribution + license info (2%) > show attribution credits in download tool
* Can't find download (3%) > add text label ‘Use this file’
* Can't find Commons link (2%) > move Commons link to the left
* Too slow (2%) > ask ops to implement back-end image rendering
4. Community RfCs:
We are now monitoring 3 Requests for Comments on these sites:
* English RfC (4): Unfavorable so far, with 48 responses to disable for logged-in users, 5 to enable (similar ratio for logged-out) — ends on July 7.
* Commons RfC (5): Too soon to tell, with 4 responses to disable for logged-in users, 3 to enable (similar response for logged-out) — just started June 27.
* German RfC (6): Hasn’t started yet, but 10 people now think it’s needed — so it’s expected to start soon.
We will send another update about the RfCs in coming days.
Please let us know if you have any questions about this update.
All the best,
(1) Weekly Disables Worksheet:
(2) Opt-in/out Dashboard:
(3) Weekly Survey Worksheet:
(4) English RfC:
(5) Commons RfC:
(6) German RfC: