On Fri, 2015-02-13 at 03:35 -0700, musikbear wrote: > ..but dl ( and not including linux) 77k ! since dec ! > * 120 peeps did the survey > * about 20-30 peeps are 'regulars' on forum > * forum has 1347 members > * most logged in at a day : 60 > * FB has 2772 'likes' > > imo these 'social' numbers does not 'match' with the dl numbers.. > it would be extremely interesting to know Why people eiter do not like to > follow lmms on either platforms or actually does dl, but either never > install (<-because its free?) or (worse) give up on lmms.. >
And they never will. You're only going to see a tiny subset of users engaged enough to want to keep up to date with the project. Still, it's a useful sampling to get feedback from the most engaged users who happened to see and respond to the survey. > The number discrepency could also emerge from dl's who does not use > social-media at all.. Yes. > One such reason could be that a huge bulk of dl comes from china, and other > non-english speaking areas? > Also yes. And don't forget the people who download LMMS, play with it for a few minutes and either have problems or otherwise decide it isn't for them. And the users who download multiple times (either they have multiple machines and re-downloading is easiest for them or they have problems with the download etc.) Downloads != active users or even active installs. I'll take a wild guess that the active users are <10% of the download number. > Several interesting focus point here: > > * Are there any demographic in the dl statistic? > * Could a 'Would you like to give the project feedback' form be part of the > installation process? > * Could the un-installation process have a 'Would you like to tell why you > choose not to use lmms' (upgraders would be annoyed here, alas) > You could do that, but don't expect data that's much better than what the survey produced. There's a reason (having nothing to do with advertising) that analytics have taken over: most users generally can't or won't provide the feedback you really need to know. If what you really want to know are active users by platform, what sound system they're using, is feature X really even being used, what are the most common bugs, etc. it would be far more useful to implement anonymous analytics with an easy opt-out mechanism and the data would be orders of magnitude better. > Finally, i smiled a bit on the x32 / x64 conversation, and two somewhat > related storries came to my mind: > > "If you have a fire breaking out in your apartment, What ever you do, do NOT > call the fire-deperment! > All Statistics shows, that the more fire-fighters that arives at a fire, the > worst is the damage" > > "There are three types of lies: > 1) normal lying > 2) bloody lying > 3) statistics" > https://en.wikipedia.org/wiki/Lies,_damned_lies,_and_statistics ------------------------------------------------------------------------------ Dive into the World of Parallel Programming. The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net/ _______________________________________________ LMMS-devel mailing list LMMS-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lmms-devel
