Reality check on Flash visitors

I am seeing surprising numbers in my Visitors > Statistics > Plug-ins data. The number of visitors whose browsers support cookies is typically about twice the number of visitor whose browsers support Flash. Has there been a major revolt against Flash that I don’t know about, or is it that Piwik only reports the most recent version of Flash?

I don’t think it’s an iPhone/iPad issue. While we do get a significant number of iPhone visitors, it’s still “only” about 4 to 5% of our site traffic. (www.sfmta.com serves San Francisco, so we tend to be ahead of the hardware curve)

maybe you have a lot of IE visitors and at the moment only the Java-plugin is tracked on IE.

btw: this tracking is something that should be enhanced.

Yeah, I would also appreciate plugin detection on IE.
I wonder how other tracking tools are doing it if it causes that much trouble with IE.

[quote=Thomas Seifert @ Jun 8 2010, 09:49 AM]Yeah, I would also appreciate plugin detection on IE.
I wonder how other tracking tools are doing it if it causes that much trouble with IE.[/quote]

There must be a solution without any promt.
I have one small page tracked and I got already 45% of my visits with IE in 2010.

I’ve already spent an inordinate amount of time researching this to come up with a workaround. Plugin detection on IE relied on instantiating the add-on’s ActiveXObject. On earlier versions of Windows+IE and when the add-on is up-to-date, this was fine. Problems start occurring when there are (or have been) multiple versions of an add-on, the add-on phones home to check for updates, or visitors are using a newer version of Windows and/or IE with “enhanced” security…

Ultimately, the decision was made to no longer detect plugins on IE because of the pop-ups that could appear: e.g., Reader opening a blank window, prompts to upgrade an add-on (e.g., Quicktime, Flash, Real player), or security alerts.

Tracking should be unobtrusive.

how does Google track the plugins?

GA instantiates the ActiveXObject on IE. It would be subject to the popup problem that I mentioned.

If IE is not going to be included, it would be good for that particular statistic, and any other such statistic to have an asterisk, with a “*-excludes IE” footnote.