-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Reza,
do not worry, I've been around uaprofs for some time now and db's even longer ;-) and there's always more than one way. But an approach from both sides and see where they meet can only add to the joy. I'm overjoyed you joined the ranks of 'us' who ask/beg/plead for log data : user-agent and NIC region [nothing identifiable]. Actually, I totally agree : ideally we should have a box, like I have here, and feed it 'logs' non-stop, as I try to here, and see what drops out, then go running to the uaprofile db and use the 'Knowledge' to navigate the maze. And yes, as expected current devices are shifting to non-device software determined 'devices'. However the OS a device was released with "stocked" is *important* : that is the difference between : method 2 [look up what OS it came with in DDR] and method 3 [see what OS it has now in the UA string]. So *not* a "misclassification" [as explained before]. That in itself is an *important* data-point and must be preserved ! The difference may be academic for some, but is significant for serious traffic analyses. *But* it is not related to the point I was raising which is : IF the difference in OS turns out to be significant MAYBE the DDR should contain that 'version' of the 'device' too [note 'method 2' *not* 'method 3']. That is what I'm trying to establish with the analyses. esjr -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (MingW32) iQEcBAEBAgAGBQJUAJ0MAAoJEOxywXcFLKYcMekIANSoUoaFXbOxFKd8EJmsAgHl BExBhRgVjslo4o+M5G+zrx0qddbSkxptl5qvGa4mepmkCdKT9M1OwAPbgfmUEsrf Px3SziZvGDz/dYT7eo8t0AdVouaP+Ld6fOYl9CaQ9f0mjJM4H/ilGDCQU8ZaR/Gf sJkSxHPyYqO6Fjakd+JXxq4XHBrPIalvyaNU+M4hOWakcXYCtPZ5xTJN7PrbfF7Q R55XW9TUbpN7LsURgnN9qHbvEA7KF+ZnTfG2/UzASgFsuD0ZFpKF2sx4vIBEmk+s vUEGM7eOZafXWyLm078mt7Lx2gPX7dpYSNUPX3hpoh1hskwtnjqT/lzeIPC9vlQ= =vmFp -----END PGP SIGNATURE-----
