MS Office on Linux 14817

Wait, whoa whoa whoa. Where do you see *either* cite being "inconsistent with itself"? Two *different* companies with access to both open and closed-source code using *different* methodologies come up with different bug counts. This does point to varying methodologies, but I don't see either one specifying different numbers at different times.

The two companies may not agree on how to clbuttify a defect, perhaps, but by either definition Linux does better than the commercial code they've seen.

You have access to your commercial code, right, and Microsoft tools to evaluate it? Well, do your own test. Go download the Linux kernel and run these tools on it. See how they compare. I'd *love* to hear even a summary. I've asked you before, but you somehow keep missing it. I'm sure there's a good reason...

Please elaborate on these axes.

Do you *have* a "National Standards Lab kind of reference"? It appears to me that you were suprised that there was any kind of numerical data in this area, and unpleasantly surprised to discover that what there was argues against your position. So, you get to work trying to discredit the numbers rather than trying to think about what they might mean.

But hey, I'm only the guy who produces actual cites to back up what I say, what do I know?

-- Sincerely,

Ray Ingles (313) 227-2317

"You actually do more to support Al Qaeda by driving when you go out to pick up your drugs! Hell, all of us can help by driving less but, unfortunately, not everyone has a smack habit to give up. (And among those who do, is 'George Bush really wants you to!' going to be the thing that makes a junkie quit?) ... Heroin is the only drug that really benefits persons." - Bill Maher, "When You Ride Alone You Ride With bin Laden"