Flawed reasoning
by Volker Weber
Just read something from a very smart person, that I find deeply flawed:
The source of the data is the Radacati Group, and shows that Exchange has more downtime than Domino. I don't recall Radicati ever identifying a positive trait of IBM/Lotus Domino, so I'm willing to accept these numbers.
The consensus is that Radicati has no credibility at all. In that case the numbers are always bogus. You can't just pick the ones that are in your favor. If you do accept some numbers, you will have to accept them all.
Comments
66% of all statistics are made up...
@Boudewijn - I heard that it was 80% of all statistics that are made up. What's your source? ;-)
As far as Sarah's Sewing Circle of paid pseudo independent reporting, I would agree with the general consensus -- She's been caught in outright misinformation and astroturfing enough times to have zero credibility.
Unfortunately, that doesn't help when customers who don't follow that kind of nonsense read her reports.
excellent statement
I think there are some people who think she's biased against Lotus. Personally, I think the explanation is either incompetence or lack of professional ethics. Or both. heh...
Fair enough, Vowe. I'll address the comment in my blog. I do agree that the stats are stats, providing the sampling population is a reasonable source.
The only statistics you can trust are the ones you falsified yourself
About reliability of statistics, Wikipedia says: "Statistics is a mathematical science pertaining to the collection, analysis, interpretation or explanation, and presentation of data". It is usually the interpretation and presentation where things go wrong.
In our DNA service, the Domino statistics from customer Notes networks (presently we have appr. 500K users/10TB) are collected and analyzed in our data warehouse. Presenting results back to the customer only works after having reached consensus over the methods we use.
So I tend to agree with Volker. I wonder which statistics this company used or refers to, and how they were collected and analyzed. Asking 2.5K for that report could be to justify their immense effort in collecting or analyzing, or just a smart way of marketing their reports.
From the Table of Contents, it appears they have surveyed people within organizations.
Personally, I prefer hard facts over people's opinions...
In the past, I think a direct customer payment relationship was shown between Sarah's coffee club and Microsoft. She made some predictions a few years ago, and something like 75% of them were released as statements less than a few days later by Microsoft with almost the same language.
