On Mon, Jun 24, 2013 at 11:35 AM, janI <j...@apache.org> wrote: > On 24 June 2013 16:16, Rob Weir <robw...@apache.org> wrote: > >> On Mon, Jun 24, 2013 at 9:26 AM, RGB ES <rgb.m...@gmail.com> wrote: >> > 2013/6/24 Rob Weir <robw...@apache.org> >> > >> >> A quick update. As of last night we're now at 55,155,204 downloads of >> AOO >> >> 3.4. >> >> >> > >> > Good! >> > >> > >> >> >> >> To put it in perspective Instagram reports that they have 100 million >> >> users, and Facebook thought they were worth $1 billion (US). >> >> >> >> Of course, we are free and always will be. As Warren Buffet said, >> >> "Cost is what you pay; value is what you receive". We're providing a >> >> lot of value. >> >> >> >> I don't think it is worth the time to do a blog post, etc. It is not >> >> news any more. But maybe at 10 million download intervals, or 75 >> >> million or something like that, we announce more broadly? >> >> >> > >> > Related though: How 4.0 downloads will be counted? Starting from zero? >> > Adding them to the 3.4 numbers? Maybe both? Something like "X million >> > downloads of which Y millions from our latest mayor release alone" would >> be >> > interesting. >> > >> >> The data is tracked at the most fine grained level, at the level of >> individual files. For example, I can tell exactly how many times the >> Swedish Linux 64-bit RPM AOO 3.4.1 was downloaded. For the chart, >> however, I combine all 3.4.x downloads together, for full downloads >> (not including language packs): >> >> http://www.openoffice.org/stats/downloads.html >> >> So when 4.0 is released we have the ability, at the data level, to >> distinguish 4.0 downloads from 3.4.x downloads. >> >> The widget I use in the chart is the Timeplot widget from the SMILE >> project. It has the ability to show multiple data series on the same >> plot, e.g.,: >> >> http://www.simile-widgets.org/timeplot/examples/housing/index.html >> >> So one idea might be to show the 3.4 data and the 4.0 data as >> individual lines on one plot, and then show the cumulative downloads >> (of all versions) on the other plot. Or some variation on that. >> >> But it is a fair point that the equivalence of downloads and users >> breaks down once you have a major update. The estimate for numbers of >> users then will need to be aligned with the greatest number of >> downloads of any specific version. >> > > Super work rob, without having dived into the details myself I hope you can > answer one question. > > Considering the download structure of 3.4.1, do you see any big backlogs of > 4.0 (point where we could assume users would download, but not supported by > 4.0) ? >
I'm not sure I understand the question. -Rob > rgds > jan I. > > >> >> Regards, >> >> -Rob >> >> > Regards >> > Ricardo >> > >> > >> > >> >> >> >> Regards, >> >> >> >> -Rob >> >> >> >> --------------------------------------------------------------------- >> >> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org >> >> For additional commands, e-mail: dev-h...@openoffice.apache.org >> >> >> >> >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org >> For additional commands, e-mail: dev-h...@openoffice.apache.org >> >> --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
