Silverlight stat quotes

Wall Street Journal had an article today, and it has already been picked up at Seeking Alpha and other sites. Here is the line which sounded strange to me:

“Adobe’s Flash player is installed on about 98% of Internet-connected PCs, and Silverlight is only installed on about 25%, according to Adobe and Microsoft.”

The first part is okay — it’s a rare computer which doesn’t have Adobe Flash Player 9 or better already installed, and almost everybody already has H.264 support — it’s the “SL at 25%” which was news to me, particularly when attributed to Adobe.

Closest match I’ve seen was the equivocal “Over 1 in 4 computers on the Internet now have some version of Silverlight installed” quote from Scott Guthrie last month. shows SL2 support at 10% on the general sites they sample.

I checked with my partners here this morning, and the Adobe speakers apparently did not offer such a Silverlight quote. Best guess is that the sentence would be more clearly phrased as “Adobe’s Flash Player 9 is installed on about 98% of Internet-connected PCs, according to Adobe, and some version of Silverlight is only installed on about 25%, according to Microsoft.”

Whether 10% or 25% doesn’t matter much in the larger scheme of things — the core question is “What does it cost your audience to use?”, and we’ve only seen new Player apps start to spontaneously appear when up above 80% existing consumer support. In case you were wondering about the WSJ quote or its echoes, it seems to be a typo.

There’s also an interesting bit on the pay-for-play angles of trophy-site adoption: “When CBS Corp.’s college sports group decided to build its Web site using Silverlight earlier this year, Microsoft chipped in free development and support that ‘reduced our costs tremendously,’ said Tom Buffolano, the CBS business unit’s former chief. A CBS spokesman declined to comment.”

Summary: No big thing, but we didn’t say it was such a big thing, in case you were wondering…. 😉

5 Responses to Silverlight stat quotes

  1. Zeh says:

    I needed that kind of information a month ago and couldn’t find any reliable source either after days of search. Here’s some additional reading though, some third-party analysis:
    17.6%, with no mention of version.
    MS really needs to start measuring that kind of stuff. My guess is that they won’t do that so soon though and will prefer to use vague quotes like that – I mean, there’s NO source mentioned or even hinted at in Scott’s post.
    Nobody expects a deep penetration so soon in the game so I don’t think the 15%-ish numbers are that bad, but that MS prefers to hide the reality is not really surprising and a sad indication of how MS still sees developers.

  2. Mihai says:

    It depends a lot on what site is measured:
    Looks like about 62% of the C++/.NET developers have Silverlight installed.
    And you need programmers to write the first wave of compelling application before you attract regular users…

  3. Dan Rayburn says:

    There is no source mentioned for Adobe’s market share numbers either. Lets not take this out of context. Both Microsoft and Adobe give out very generic market share and install numbers without providing any of the data behind where those numbers come from, let alone giving users the ability to review that data. They don’t break out market share by vertical and don’t say how many installs are unique.
    Both companies have been doing this for years. Adobe can’t say Microsoft’s numbers are questionable and Microsoft can’t say Adobe’s are questionable when both companies are doing the exact same, vague reporting.
    [jd sez: ??? Dan, I provided the link to the quarterly Millward-Brown consumer audits of realworld viewability in the article up above. And I sourced the MS quote, as best I was able, rather than saying it was unsourced or “questionable”. An attempted “moral equivalence” argument won’t play!]

  4. Dan Rayburn says:

    You have to be kidding me right?
    This is a “proprietary study commissioned by Adobe Systems” which means no one except Adobe has access to the data. No where is there a link to what this survey looked like, what was asked, what percentage were just consumers, what kind of video was tested, (streaming, progressive download, download, live), what their connection speed was, etc…. there is absolutely no data to actually review.
    [jd sez: Methodology & sample test are at the cited link, and this study has been going on for eight or so years now. It presents the simplest, most basic use of video, and asks “Can you see this right now, without installing anything new?” (Hmm, looks like the “Sample Survey” link is broken, let me check.)]
    Not to mention the methodology says “The survey presents respondents with several pages, each with a graphic image in a different media file format, and asks respondents to indicate whether or not they can see each image.” Why would a “graphic image” be presented if we are talking about video?
    This is not a knock on Adobe and is directed to ALL of the platform providers. But if Adobe, Microsoft, Apple or any other company with a player/plugin wants us to take their numbers seriously, then why isn’t the data available for review? Why can’t you click on a link and see what the actual survey looked like?
    Does Adobe Flash for video have massive penetration, yes, but does it really matter if you have 90% or 99%? Absolutely not. The amount of time and money that companies put into these types of surveys, which like this one was a very_small sample size, could be spent to educate the market in a lot better manner.

  5. Scott Barnes says:

    The 25% number did not come from Microsoft, and it’s just editorial control via the journalist.
    Furthermore, Adobe & Microsoft from time to time do help vendors out with production costs where it suites, but the intent is always to provide a strong ROI for both parties (typically a 1/3rd investment from all folks involved).
    NBC Olympics was a great story here, whereby all folks not only invested, but both parties came out with a profitable venture. Writing blank cheques which the quote above eludes to is simply colorful reporting.
    One thing I find astonishing with Adobe at the end of the day, is it’s stated there is 98% Flash installed, but yet Adobe reports 18million downloads per day on average. 1.4billion people online, so my math maybe off but I raise doubts about the 4600 folks surveyed representing 98%.
    [jd sez: That incidental “18M” quote was the average daily successfully-completed installation results during July 2008, within a larger article where Emmy tried to show why such extrapolations were folly.]
    Furthermore, in the 4600 Adobe doesn’t disqualify with folks whom are at “work”, and yet cross-reference the data against folks connected to the internet “at home” census data.
    Point is this, the whole notion that ubiquity is the important piece here is somewhat misguided overall. 18 million people installing Flash per day and that’s successful installs (more with if and you factor in abanonemnt rates) clearly indicates that the assumed “plug-in fatigue” is actually not as major as folks think.
    It really comes back to how you market your RIA/Branded Experience as whilst someone will willingly install a plugin, will they around to watch the solution as well?

    Scott Barnes
    Rich Platforms Product Manager