Needless to say. I'd love to compare to LibreOffice download numbers. They used to quote them, back when they started. But for some unknown reason they stopped publishing such numbers as soon as Apache OpenOffice started publishing their numbers,

You either did not read the survey results or did not understand them. Survey participants were asked about "the software application called OpenOffice". They were asked whether:

1) They had heard of it

2) They had tried it

3) They use it occasionally

4) They use it regularly.

The "continued to use" percentage is the sum of "they use it occasionally" and "they use it regularly". It excludes those who just tried it.

I'm not sure you read or understood what was in my blog post. In particular, one of the survey choices was "I tried it once". That was around 6% of survey participants. Absolutely no where is it assumed that a person is a regular user just because they installed OpenOffice.

Actually, the survey was repeated, three times over 18 months, with similar results. And Google Consumer Survey's does post -stratification weighting to ensure the survey participants match the target demographic by age, sex, geography and income. The approach has been validated. So I have a good data set.

I seem to have all the facts here, while you seem to have all the opinions.

Actually not. It was a random survey, conducted via Google Consumer Surveys. I had absolutely no input on the surveyed participants.

Actually, we do have those stats, via surveys. 78% of users who try OpenOffice continue using it.

I can say with authority that it was 100 million of full installs of Apache OpenOffice specifically, not counting OpenOffice.org release, not including beta releases, not including language packs.

According to our survey data, 78% of those who try OpenOffice continue to use it. And I suspect (though I have not wasted time on that specific survey question) that far more than 30% of those who download it install it. In fact, it would be a common occurrence to download once and install on multiple machines.

Your theory is sound, but your numbers are not. For example, Apache OpenOffice has only had 4 releases in two years.

There are many other factors to consider: Users can take the same download and install on multiple machines, they might share with friends or family members (I do that). A corporate installation might have a single download sitting on a network file server shared with many. There are also many 3rd party sites that themselves have seen millions of OpenOffice downloads, e.g., download.com. And of course, not all users upgrade, or upgrade quickly.

In any case it is a fair point that you cannot simply equate downloads with users.

I actually have been looking into that question and tracking it via surveys. Of those who tried OpenOffice, 78% continued to use it "sometimes" or "regularly":

See: http://www.robweir.com/blog/20...

Unless you are a business user you are unlikely to use any office application daily.

If that were true then there would not have been 100 million downloads of Apache OpenOffice, would there? Therefore...

Microsoft claims 1 billion MS Office users. No doubt some/many are pirated, but that gives a sense for the scale of the potential user base for OpenOffice. And from what I've seen, Apache OpenOffice gets around 1 million downloads per week, a steady rate that can certainly continue for quite a while. So even if Apache did nothing, we would get to another 100 million downloads in another two years.

The question is whether we want to glide or really take off?

To really advance among mainstream end-users, people like your mother, this will only happen as average people, not just the techies, learn about open source and are comfortable with it. This means better documentation, especially geared toward newbies.

To advance among corporate users OpenOffice needs better interop with Microsoft Office. Yes, I hate to say that as much as you probably hate to hear it, but it is the reality we (some of us at least) live with.

Finally, we should find a way to extend the OpenOffice brand to the web and tablet editing experience, since traditional desktop PC use is a diminishing proposition.

It sounds like SUSE, the largest contributor to LibreOffice, is ending their investment in LibreOffice, and their engineers are looking for new employment. This echos the way they got out of the Mono business a few years ago. But taking the same people and putting them in a much smaller company, with far less enterprise sales experience, is not something that will cause Microsoft to lose any sleep.

It sounds like SUSE, the largest contributor to LibreOffice, is ending their investment in LibreOffice, and their engineers are looking for new employment. This echos the way they got out of the Mono business a few years ago. But taking the same people and putting them in a much smaller company, with far less enterprise sales experience, is not something that will cause Microsoft to lose any sleep.