Closed Bug 473856 Opened 17 years ago Closed 16 years ago

Set-up flash version tracking in Omniture

Categories

(www.mozilla.org :: General, defect)

All
Other
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: kkovash, Assigned: oremj)

Details

We want the ability to see the breakdown of flash versions for traffic at mozilla.com. For whatever reason, Omniture doesn't provide this out-of-the-box, but their support people provided the following info: Eric Hansen: Unfortunately, there is no default way in SiteCatalyst to actually track the version, but this could be easily remedied by a customized solution...passing the version into an eVar or s.prop Eric Hansen: Adobe's detection kit would help you in this area Eric Hansen: http://www.adobe.com/products/flashplayer/download/detection_kit/ Eric Hansen: Here is another example site that uses javascript code to detect the flash version Eric Hansen: http://web.mit.edu/is/web/reference/guidelines/flash/example.html You: so we'd need to work with our account manager to set this up? Eric Hansen: if you want, but the examples I just gave you are pretty easy to implement Eric Hansen: you could set up the s.prop for flash version tracking yourself Eric Hansen: simply make changes to your s_code.js file and/or your page code Can this be easily implemented?
Wil, what is your opinion on including either of the scripts, mentioned in comment 0, on mozilla.com.
Looks like they are both built off the same script (AC_OETags.js). If it's important I think it'll be fine. Just more stuff for the browser to execute on every page load. :-/
Could we approach this like funnelcake, and sample every once in a while but most days not have this code active? It may be a PITA to push the code for a day and then back it out, weighed against having this run all the time on every page. The adobe kit is complete and browser neutral, but it does involve running a flash snippet. I'd be fairly happy with a simple logging of navigator.plugin["Shockwave Flash"].description That's probably Mozilla-only, and isn't alwys as detailed as the in-flash version, but it's good enough for me really. Do we need to know the Flash version of IE-users? might be interesting, but the IE-installed flash isn't the same one as the plugin version so the two don't necessarily match. I'm primarily interested in how potentially at-risk Firefox users are. Users of other browsers are already at risk of not running Firefox :-)
Assignee: server-ops → oremj
is there some middle ground solution that would allow us to always have the code active? for example, could we have the code run all the time on a single page (e.g., the whatsnew page)?
Group: infra
Component: Server Operations → www.mozilla.com
Product: mozilla.org → Websites
QA Contact: mrz → www-mozilla-com
What's the verdict here?
Dan -- are you working on an alternate solution, or can we implement something in Omniture?
What's the action on this bug?
I sent an email to Veditz. I'd like us to implement some version of this idea if possible.
I spoke with Veditz. There are no objections here. We're good to implement either script/solution from comment 0.
What has to happen to close this bug out? Just implement?
Correct. I think we're just waiting on implementation.
oremj, can this be done this week? Does it require a downtime window?
Committed in r23704 and pushed live.
Status: NEW → RESOLVED
Closed: 16 years ago
Resolution: --- → FIXED
Jeremy -- did you double check the reporting? I'm looking at Custom Traffic -> Flash Version, and some of the popular "Flash Versions" are listed as "-1" or "WIN 10.0.22.87". Are those supposed to be appearing that way?
Yep, those are the values that the http://www.adobe.com/support/flash/detection.html javascript returns. -1 means the browser does not have flash installed or, at least, a version could not be detected. IE reports flash version differently, e.g., "WIN 10.0.22.87".
Component: www.mozilla.org/firefox → www.mozilla.org
Component: www.mozilla.org → General
Product: Websites → www.mozilla.org
You need to log in before you can comment on or make changes to this bug.