Talk:Comscore/Archives/2010

Article improvements

I wrote the initial (non-stub) version of the article. There are a few things about the article I'm not perfectly happy with, or that I think could be improved:

1. The current article doesn't spell out whether comScore provides a single overarching service, or whether it provides several clearly distinct products and services. Its services are almost certainly related to some extent, since they probably all rely on the underlying 1.5-2 million pool of sampled users, but there may be different classes of end products with important distinctions between them.
2. Some of the (non-historical) references are up to 5 years old. It would be good to get newer sources and make sure the article reflects the current state of the company.
3. The company tends to be a bit secretive about its methods (eg. it's somewhat difficult to definitively determine that comScore is related to its distribution websites OpinionSquare and PermissionResearch; only Ernst & Young's .pdf does this). As a result, I ended up using two flawed .ppt sources. Both are hosted at unreliable URL's (it's not out of the realm of possibility that they're modified by other users, or possibly even faked), and one is marked "proprietary and confidential". Nonetheless, they were found in the wild using only Google, and they broadly concur with with other sources. Still, if better sources could be found for their detailed information, that would be good.

--Interiot 19:05, 21 November 2006 (UTC)

To point # 3 above, the privacy policies on both permissionresearch.com and opinionsquare.com contain the following line: "The information that you contribute is used by comScore Networks, Inc., a U.S.-based market research company that is a nationally-recognized authority on Internet and general economic trends, whose data are routinely cited by major media outlets such as the New York Times, the Wall Street Journal, CNN, and CNBC, and is extensively used by the largest Internet services companies and scores of Fortune 500 companies." --Bibbendum, 4 January 2007

Well, the things that the PPTs clarify/backup are 1) how the proxy servers and RDD are related (though, as the article now notes, the proxy servers may no longer be used), and 2) that comScore directly confirmed geography/income/age from the proxy server data, and 3) that geography/income/age were the pieces of data that were used to correlate the proxy server and RDD data. --Interiot 17:53, 4 January 2007 (UTC)

Proxy servers currently used?

The article currently says that comScore no longer uses proxy servers. It would be good to get reliable sources to back this up. The most recent source cited by the article (the Dec 2006 Forbes article) says comScore "takes virtual photos of every Web page viewed by its 1 million participants, even transactions completed in secure sessions", which strongly implies that Forbes thinks comScore has recently been using proxy servers with the man-in-the-middle SSL certs. Has comScore put out a press release about this, or has a mainstream publication covered the change? --Interiot 18:02, 4 January 2007 (UTC)

comScore might be going public, which would hopefully make this information much easier to come by. In its first SEC filing, comScore noted it changed its methodology in 2002, and that it was no longer able to generate all the reports it was previously able to (which caused comScore to lose some customers). [1] However, it also notes that its software is often classified as spyware. That's not an answer to the above question, but hopefully an answer will be coming soon, with more public filings and with the latest news spike. --Interiot 18:29, 3 April 2007 (UTC)

Corrections to entry

I have a few corrections/ suggestions for the content of this article. Thought I'd chime in here prior to undertaking edits (that's the way I'm supposed to do it, right?).

In the criticism section, it should be noted that the downloads were not authorized by comScore.

Per the Forbes article "ComScore admits that the company engaged in partnership negotiations with DollarRevenue, even going as far as giving the company test software, says privacy officer Chris Lin. But the discussions stopped there, and the companies never signed a contract. Then, several months ago, comScore software installed by DollarRevenue started reporting back to company servers, says Lin.

Lin insists that the unauthorized software did not violate anyone's privacy. The company quickly cut the cord between the software and the servers. "This is the only issue that we have had with a potential distributor in the six years that our company has been in operation," says Lin. DollarRevenue said it "never really worked" with comScore but did not answer further questions about the unauthorized downloads. ComScore said it never observed any illicit downloads from PacerD or MediaMotor and has no relationship with either company."

Suggestions: 1) Move "In the past, the software forwarded users' internet traffic through comScore proxy servers, which caused criticism among IT professionals. [2] [3] As a result, several universities and banks took steps to block the proxy servers. [4] [5] In response to these concerns, comScore no longer uses this technology.[citation needed]" to the "criticism" section of the article.

2) Add "comScore was selected by the World Economic Forum, Davos, as a 2007 Technology Pioneer." (reference www.weforum.org/techpioneers/2007)

ComScore 19:28, 18 May 2007 (UTC)

Thanks for posting here. Yes, Wikipedia:Conflict of interest does say that it's best if companies who wish to clarify/correct their articles can discuss proposed edits on the talk page (eg. here) before making them to the article. If you get no feedback or positive feedback, feel free to make the changes yourself, but in this case, I made most of the changes you suggested...

Regarding DollarRevenue/distribution, I updated that, as you suggested: [2]

Regarding #1: Do you know if there are any reliable sources (news articles from CNN/Forbes/etc. or university analysis or anything like that) that confirms that proxies are no longer being used? (and perhaps, more importantly, if you're implying this, that analysis by reliable sources concludes that the new data-collection method has less impact on user's privacy as the proxy method had) If there are reliable sources that confirm a new data collection method is being used, then I definitely agree that the #1 should be moved to the criticisms section. Otherwise though, our most recent reliable source from Dec 2006 says that the proxy servers were in use, and it's a fairly integral part of the company's methodology, and the rest of the paragraph really starts with that as a basis. But if reliable sources can be found, then it should be moved.

Regarding #2: It's possible somebody else might remove it as looking like too much self-promotion, but the website does a decent job of explaining what it's about, so I went ahead and added it. [3] --Interiot 19:01, 19 May 2007 (UTC)

[4] So are there reliable sources that say the proxies aren't being used? If not, then it's an integral part of the explanation of how data is collected, and at least I personally don't think it should be moved out of the "data collection and reporting" section because it's useful prerequisite information that helps the reader understand the latter half of that section. --Interiot 02:33, 7 June 2007 (UTC)

Technical Limitations

Might also be worth noting that their audience is also limited in scope by technical limitations, they only allow Windows users, using one of the popular browsers, this limits the scope of the audience. —Dreamsorcerer 19:12, 10 April 2008 (UTC)

STILL installs without asking

I have found that the web survey RelevantKnowledge still installs on user's computers without their consent. I have seen several cases of this lately. Also, many antivirus programs are blocking the sofware and/or website now. Perhaps something like this should be added to the complaints section? Bryzal (talk) 16:13, 9 June 2010 (UTC)