Comments on: Unity08 Candidate Rankings

By: Michael van der Galiën

Michael van der Galiën — Mon, 17 Sep 2007 18:19:55 +0000

Yeah, I’ll soon found out what it’s like to lead a major effort
The offer for a guest post still stands

By: Bob Roth

Bob Roth — Mon, 17 Sep 2007 18:04:52 +0000

We only reported a sampling of the more secure versions. The entire results are not yet available. It is a massive amount of data and some of the questions differed between studies (none of the varied questions were reported, they were mostly specific to Unity08 itself).

By: Tully

Tully — Mon, 17 Sep 2007 00:18:56 +0000

Bob,

Query: Did you report only the more secure version? And did you report the results of it in entirety? That point is left unclear in your response.

By: Bob Roth

Bob Roth — Thu, 13 Sep 2007 18:26:27 +0000

Thank you, Michael. You would not believe how busy we are… we’re trying hard to keep overhead low, yet still build the tools and provide the support that our members expect. Not sleeping a whole lot around here. In fact, right now I am holed up in a Starbucks on the west coast refueling and trying to get caught up on email. Lovin’ it, though.

By: Michael van der Galiën

Michael van der Galiën — Thu, 13 Sep 2007 18:16:25 +0000

Bob, it’s good to hear from you again.

By: Interested

Interested — Thu, 13 Sep 2007 17:09:33 +0000

existing technologies is far different from available technologies.

By: Simon

Simon — Thu, 13 Sep 2007 17:02:51 +0000

Given that the survey’s methodology was seriously flawed, it’s hardly surprising that the results are seriously dubious.

With regard to slide 29, I suppose my vote is counted towards the 0% of respondents who selected the option I did. So as it turns out, 0% ≠ 0%.

By: Bob Roth

Bob Roth — Thu, 13 Sep 2007 16:04:40 +0000

For load reasons. The more secure version was very server load intensive, while the less secure version could handle a large number of requests. In order to get the invitations sent in a timely manner to all the members, we split them between the secure and non-secure versions. The study itself was exactly the same. The highly secure version locked a member’s study so that it could not be accessed again. The less secure version could be accessed again, but we only recorded the first completion of the study.

The sampling was random across the secure versions of the study.

Remember, we are pushing the limits of existing technologies while also building things that will be entirely new. This is only the first vote and we’ve already done a lot of work behind the scenes to increase our ability to accept more connections in secure mode for the next version.

By: marc

marc — Thu, 13 Sep 2007 14:50:05 +0000

Good question. I can understand culling away the outliers but that much massaging seems invalid.

As one of the people who completed the survey I’d prefer to know that my vote counted.

I’d also prefer to have an unfiltered view of the other views represented by Unity participants in order to determine the reliability of the people I’m associating with.

By: Tully

Tully — Thu, 13 Sep 2007 14:38:11 +0000

The study was completed by more than 20,000 members, but we only reported a sampling of those results ~3000.

Why?