On Friday 27 May 2011 22:03, Wes Groleau ([email protected]) opined: > On 05-27-2011 22:28, Bob Melson wrote: >> So, will SOMEbody please 'splain me this thing called a >> globally/universally unique ID and its place in the grand scheme of >> things? > Thanks. It's about what I expected - and took away from the previous exchange. > UUID / GUID were not created by genealogists to identify multiple > definitions of a person. They were created by computer types to > distinguish between two things that can't be told apart any other way. > Kind of like a serial number. > > A checksum on the other hand, is a way to be almost certain that > two items have or don't have identical contents without actually > comparing byte by byte. Or to verify that the item has or hasn't > changed since the checksum was generated. And here you have it. My mind insists that *IDs _should_ be like some sort of super checksum, with the same "record" resulting in the same *ID no matter _where_ the record might be found. Dunno if it'd be any more useful than a *ID, but it certainly couldn't be any _less_ useful. (And the pesky critter - my mind, that is - insists there's gotta be a reasonable use for *IDs beyond taking up space in a database.) > > If a record on your machine and one on my machine have identical > UUIDs, then either one of them was copied from the other (NOT > independently generated) or one of us (or our software) was naughty and > altered a UUID. If the UUIDs match and the items do not, then > either someone changed the UUID on another record, or changed the record > without giving it a new UUID. Well-behaved software will not > gratuitously change a UUID. But lots of programs will fail to create a > new UUID when the item ceases to be a copy of the other. That, in my > opinion makes them useless for genealogy. But they were never intended > to be some magic way of automatically identifying independently > generated records as being representations of the same entity. Then why bother with 'em? I won't further belabor the point. Kinda like the Kipling poem "East is East and West is West, and never the twain shall meet, 'til earth and sky meet presently at God's great Judgement Seat", except here we have GUID and UUID and ... > > -- > Wes Groleau > > There are two types of people in the world … > http://Ideas.Lang-Learn.us/barrett?itemid=1157 Sighin' Ol' Bob -- Robert G. Melson | Rio Grande MicroSolutions | El Paso, Texas ----- The greatest tyrannies are always perpetrated in the name of the noblest causes -- Thomas Paine

Bob Melson wrote: > On Friday 27 May 2011 22:03, Wes Groleau ([email protected]) > opined: > >> On 05-27-2011 22:28, Bob Melson wrote: >>> So, will SOMEbody please 'splain me this thing called a >>> globally/universally unique ID and its place in the grand scheme of >>> things? >> > > Thanks. It's about what I expected - and took away from the previous > exchange. > >> UUID / GUID were not created by genealogists to identify multiple >> definitions of a person. They were created by computer types to >> distinguish between two things that can't be told apart any other way. >> Kind of like a serial number. > >> >> A checksum on the other hand, is a way to be almost certain that >> two items have or don't have identical contents without actually >> comparing byte by byte. Or to verify that the item has or hasn't >> changed since the checksum was generated. > > And here you have it. My mind insists that *IDs _should_ be like some sort > of super checksum, with the same "record" resulting in the same *ID no > matter _where_ the record might be found. Dunno if it'd be any more > useful than a *ID, but it certainly couldn't be any _less_ useful. (And > the pesky critter - my mind, that is - insists there's gotta be a > reasonable use for *IDs beyond taking up space in a database.) > >> >> If a record on your machine and one on my machine have identical >> UUIDs, then either one of them was copied from the other (NOT >> independently generated) or one of us (or our software) was naughty and >> altered a UUID. If the UUIDs match and the items do not, then >> either someone changed the UUID on another record, or changed the record >> without giving it a new UUID. Well-behaved software will not >> gratuitously change a UUID. But lots of programs will fail to create a >> new UUID when the item ceases to be a copy of the other. That, in my >> opinion makes them useless for genealogy. But they were never intended >> to be some magic way of automatically identifying independently >> generated records as being representations of the same entity. > > Then why bother with 'em? > AIUI at the time -- they were intended to be used in a multi-person data-entry/research project which has a master/main/official database. The database "keeper" to use Legacy's term, shares the GED from that database with all researchers, who can modify or extend the data as they see fit before returning it to the "keeper". The "keeper" then uses the UIDs to find who changed or added what, without having to plod line by line through either the GED or the database. > I won't further belabor the point. Kinda like the Kipling poem "East is > East and West is West, and never the twain shall meet, 'til earth and sky > meet presently at God's great Judgement Seat", except here we have GUID > and UUID and ... (G) Cheryl