Microsoft Caught Rigging ZD Net Poll

Dj writes "Microsoft have been found to be rigging a ZDNet poll". Apparently they didn't dig on the idea of .NET losing. Of course as anyone knows, never trust an online poll because this sort of stuff is obviosly happening all the time. I just wonder how many comments posted around the net are posted with the same goals in mind.

Re:hmmm. no suprise here

[el_doop]

Well, it sorta was...

Several of the voters evidently followed a link contained in an email, the subject line of which ran: "PLEASE STOP AND VOTE FOR .NET!" We know this, because our logs include the Web address where visitors browsed from; when people click there from a Microsoft Exchange email message, Exchange helpfully gives us the subject line and username. The people who followed that link all had email addresses in the microsoft.com domain.

Re:Did Microsoft bother...

[Anonymous Coward]

This is nothing new.

Our company has 40k employees and whenever there is a big poll somewhere involving any of our products, a memo is sent out to the entire company telling everyone to go vote for our products.

Needless to say, even a portion of that 40,000 can drastically change the skew of a poll.

I always vote for a competitor because I find it detestible that my employer would do this.

Re:Hohum

[LEPP]

You must not have read the article. It almost certainly was not a script. ZD was tracking what links were used to get to the page. Fortunatly MS Exchange gave the link with the subject line of the email and the unames. This is no great conspiracy but it is kind of funny. This kind of crap is commonplace but it is funny when they get caught.

LEPP

Re:Hohum

[rizzo242]

Second, Microsoft uses proxying for Internet-related stuff, which could make the multi-vote issue appear to be worse than it actually is, as many separate users would come from a single IP.

Uuh...but the article said the referer URLs to those votes included their usernames. They would therefore have no problem counting the per-person votes.

Seriously, people need to read the articles before posting.

Re:Did Microsoft bother...

[HMC CS Major]

uh, perl runs on windows? not only that, but writing a script to do it in VBS would be pretty simple too...


let's keep the anti-ms raving directed and on topic.

Re:Interesing 'privacy' note...

[jimboc]

it is the web based version. from it you can extrapolate the username (not necessarily their email address) and the subject of the message. Bill and his Boys make another schoolboy error :)

There is a note about this

[xanadu-xtroot.com]

On the poll archive page [zdnet.co.uk], there is this message at the top:

Poll Results
On 21 December, ZDNet posted a story [zdnet.co.uk] reporting the preliminary results of this poll, which showed a large majority of respondents who said they planned to deliver applications via Web services by the end of 2002 favoured Java for the job. At the time, Java outranked .Net by a factor of three in this poll. By early January, the position had reversed; the results are shown here. An investigation indicated that Microsoft employees used vote-rigging to distort the results. The full story can be found here [zdnet.co.uk].

Re:So why didn't ZDnet pull the poll?

[bungalow]

It appears that the article has been edited since you linked to it.

The new test reads thus:

On 21 December, ZDNet posted a story reporting the preliminary results of this poll, which showed a large majority of respondents who said they planned to deliver applications via Web services by the end of 2002 favoured Java for the job. At the time, Java outranked .Net by a factor of three in this poll. By early January, the position had reversed; the results are shown here. An investigation indicated that Microsoft employees used vote-rigging to distort the results. The full story can be found here.

Re:So why didn't ZDnet pull the poll? - THEY DID.

[MisterP]

Hmm. When i go to the poll link listed i get this:

"On 21 December, ZDNet posted a story reporting the preliminary results of this poll, which showed a large majority of respondents who said they planned to deliver applications via Web services by the end of 2002 favoured Java for the job. At the time, Java outranked .Net by a factor of three in this poll. By early January, the position had reversed; the results are shown here. An investigation indicated that Microsoft employees used vote-rigging to distort the results. The full story can be found here."

Re:Well

[jnana]

Here you go. It is one of the bulleted items in the article:

• There is also clear evidence of automated voting, with scripts attempting to post multiple times.