Marc Sibert wrote:
Le 18/09/2012 23:24, Lester Caine a écrit :
Marc Sibert wrote:
Again, I'm not a vandal : I do not detroy any work (and nobody complains about
that), I just "update" data (replace), that is not the point why my account was
blocked !
So, what have you done in my case ?
It is your opinion that you have not destroyed any work, but this is one of
the major complaints about this type of import process. The EXISTING data has
been destroyed, and that is historic data that is now lost. When a new import
comes out will you again destroy this one and upload the new one? If some one
has gone through and added all the missing address and other details how will
you link that to the new import? Are you sure that no one has added some extra
data which has been delete this time?
I respond the point in my previous message.
What is missing with this type of import is any mechanism to link to the past
history and THAT is my complaint and one of the points of the guide line - you
have destroyed data - just as you have with other edits you have done where
you have deleted objects with several years history and replaced them with a
new object.
I remember a few months before, I use to destroy way and *replace* them with
brand new nodes & data in order to pass thru the "redaction bot", and you are
saying history is important ? LOLOLOLOLOL !
That was a problem for material in the UK as well, but the replacement data is
of a much higher quality then before.
What we need in order to PROPERLY import this data is a unique ID for each
element in the source data that is maintained by the originator of the data,
so that when an 'update' arrives, the new data can be correctly matched to
that already contained in the OSM database.
Uniq ID ? LOLOLOL again ! Tell me what appends when I cut a way in 2 peaces : a
part keep the old ID and the other get a new one without *any* link with the
previous one. In fact, do you ever contribute ? Do you realy know how OSM (and
primary keys) works ? (just kidding). By the way history is still in diff files.
Well I took the trouble to look you up ... so you can see my history and look me
up in the lists.
All your arguments are sensless !
Why?
This is an area where there HAS NOT been any agreement other than the history
WILL currently be maintained in the database. My question still remains - what
happens when the data is next updated - will you delete everything again?
So with regards the 'import guidelines' do you still think you have complied
with them? In some peoples eyes you probably have, but in others some useful
historic data has been lost. I'm in the second camp ...
Please explain me what the guidelines are protecting from : in *my* (and no
other) case I have still no answer. So I still do not consider "guidelines".
"Same player, try again..."
A mass delete of data without a proper comment triggered a response that
something bad was happening. An explanation that this was a replacement for an
import would have been flagged if the action was from an account that was
FLAGGED as doing imports, but since it was not ... any more changes need to be
stopped until the matter is sorted. PERSONALLY I would like to see the buildings
that were present in 2010 still flagged as such in the database and that is my
gripe with the lacks way the guide lines are written, but we DO still need to
properly manage every import that is going to be updated at regular intervals so
that YOU do not have to manually check every building every time!
--
Lester Caine - G8HFL
-----------------------------
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk
_______________________________________________
talk mailing list
talk@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk
