Re: Who mapped it first with ref to forth coming deletions - implication

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Who mapped it first with ref to forth coming deletions - implication

john whelan-2
So essentially all data that existed on this date will need to be deleted since we can't be sure who entered or edited it or if they have agreed to the new license if the .odbl database is to be "clean".

Cheerio John

On 14 December 2011 11:31, 80n <[hidden email]> wrote:
On Wed, Dec 14, 2011 at 3:54 PM, Nick Whitelegg <[hidden email]> wrote:

Doesn't make any difference to the CTs, but I've noticed but I'm not the first named author of a few ways which I'm 99.99% sure that I created: the ways with the ID 2232-2235. I still remember the surveying/editing session in which I created the ways.

These were very early ways (spring 2006) so I'm guessing that recording the history of who created/edited ways only came in after that? However, ways with even lower IDs (e.g. 223) do have myself as original author. Curious as to why my involvement with the 223x ways appears to have been lost...


All history prior to 7th October 2007 was lost when the API was upgraded from 0.4 to 0.5.

An email from that date confirms this[1]:

4. History cleared. History will continue to be written as before,
but we have removed past history data from the database today. When
accessing existing objects you can still see the person who last  
modified them (even if that modification was before the switch), but
no details about any previous modifications.

Everything that was not version 1 on 7th October 2007 has an incomplete history and ought to be considered to be unsafe.

80n

[1] http://lists.openstreetmap.org/pipermail/talk/2007-October/018638.html

_______________________________________________
talk mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk



_______________________________________________
talk mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Who mapped it first with ref to forth coming deletions - implication

Richard Weait
On Wed, Dec 14, 2011 at 1:15 PM, john whelan <[hidden email]> wrote:
> So essentially all data that existed on this date will need to be deleted
> since we can't be sure who entered or edited it or if they have agreed to
> the new license if the .odbl database is to be "clean".

That's quite a conclusion that you are jumping to there, John.  Of
responding accounts registered by then, more than 98.5% have accepted
CT/ODbL.

_______________________________________________
talk mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Who mapped it first with ref to forth coming deletions - implication

David Groom
----- Original Message -----
From: "Richard Weait" <[hidden email]>
To: <[hidden email]>
Sent: Wednesday, December 14, 2011 6:32 PM
Subject: Re: [OSM-talk] Who mapped it first with ref to forth coming
deletions - implication


> On Wed, Dec 14, 2011 at 1:15 PM, john whelan <[hidden email]>
> wrote:
>> So essentially all data that existed on this date will need to be deleted
>> since we can't be sure who entered or edited it or if they have agreed to
>> the new license if the .odbl database is to be "clean".
>
> That's quite a conclusion that you are jumping to there, John.  Of
> responding accounts registered by then, more than 98.5% have accepted
> CT/ODbL.
>
Richard

if you are saying that that conclusion is incorrect then could you tell us
what will happen ?

Regards

David
> _______________________________________________
> talk mailing list
> [hidden email]
> http://lists.openstreetmap.org/listinfo/talk
>



_______________________________________________
talk mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Who mapped it first with ref to forth coming deletions - implication

Maarten Deen
In reply to this post by Richard Weait
On 14-12-2011 19:32, Richard Weait wrote:
> On Wed, Dec 14, 2011 at 1:15 PM, john whelan<[hidden email]>  wrote:
>> So essentially all data that existed on this date will need to be deleted
>> since we can't be sure who entered or edited it or if they have agreed to
>> the new license if the .odbl database is to be "clean".
>
> That's quite a conclusion that you are jumping to there, John.  Of
> responding accounts registered by then, more than 98.5% have accepted
> CT/ODbL.

Well, since all history of that data before API v0.5 is lost, and the
oldest history known is of the last person editing it, you don't know
who created it. Therefore you don't know if this data is created by
someone who agrees to the CT and/or the license move.

Isn't the conclusion then that that data should be deleted?

Regards,
Maarten

_______________________________________________
talk mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Who mapped it first with ref to forth coming deletions - implication

Richard Fairhurst
Maarten Deen wrote:
> Well, since all history of that data before API v0.5 is lost

Hey hey hey. Slow down.

Data before API 0.5 is _not_ lost. It is archived.

cheers
Richard

Reply | Threaded
Open this post in threaded view
|

Re: Who mapped it first with ref to forth coming deletions - implication

andrzej zaborowski
In reply to this post by Maarten Deen
On 14 December 2011 20:14, Maarten Deen <[hidden email]> wrote:

> On 14-12-2011 19:32, Richard Weait wrote:
>>
>> On Wed, Dec 14, 2011 at 1:15 PM, john whelan<[hidden email]>
>>  wrote:
>>>
>>> So essentially all data that existed on this date will need to be deleted
>>> since we can't be sure who entered or edited it or if they have agreed to
>>> the new license if the .odbl database is to be "clean".
>>
>>
>> That's quite a conclusion that you are jumping to there, John.  Of
>> responding accounts registered by then, more than 98.5% have accepted
>> CT/ODbL.
>
> Well, since all history of that data before API v0.5 is lost, and the oldest
> history known is of the last person editing it, you don't know who created
> it. Therefore you don't know if this data is created by someone who agrees
> to the CT and/or the license move.
>
> Isn't the conclusion then that that data should be deleted?

The conclusion should probably be that it needs to be treated as if v1
was not ODbL compatible.  With a clever enough algorithm this may not
always be the same thing.

However http://planet.osm.org/history/ contains daily diffs starting
from 2004-07-01.  But, it looks like only the last edit of each day is
kept which means that there may be edits by other users in between
these edits (?).

Cheers

_______________________________________________
talk mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Who mapped it first with ref to forth coming deletions - implication

Maarten Deen
In reply to this post by Richard Fairhurst
On 14-12-2011 20:32, Richard Fairhurst wrote:
> Maarten Deen wrote:
>> Well, since all history of that data before API v0.5 is lost
>
> Hey hey hey. Slow down.
>
> Data before API 0.5 is _not_ lost. It is archived.

That is something different than what 80n said earlier, quoting a
message from Frederik Ramm:

http://lists.openstreetmap.org/pipermail/talk/2011-December/061087.html

 > An email from that date confirms this[1]:
 >
 > 4. History cleared. History will continue to be written as before,
 > but we have removed past history data from the database today. When
 > accessing existing objects you can still see the person who last
 > modified them (even if that modification was before the switch), but
 > no details about any previous modifications.
 >

[1] http://lists.openstreetmap.org/pipermail/talk/2007-October/018638.html

So, which is it? Cleared and no details about any previous modifications
or archived and earlier details available?
You have to excuse the confusion, because these are conflicting messages.

Maarten

_______________________________________________
talk mailing list
[hidden email]
http://lists.openstreetmap.org/listinfo/talk
Reply | Threaded
Open this post in threaded view
|

Re: Who mapped it first with ref to forth coming deletions - implication

Richard Fairhurst
Maarten Deen wrote:
> That is something different than what 80n said earlier

80n, not for the first time, is wrong.

> So, which is it? Cleared and no details about any previous
> modifications or archived and earlier details available?
> You have to excuse the confusion, because these are conflicting messages.

We have the full history of every OSM object ever created.

Post-0.5, the history is stored in the main database, as you know.

The API 0.5 changeover involved a very significant change to the data model (we dropped the segment datatype, and changed the definition of a way to be an ordered list of nodes rather than an unordered collection of segments). It therefore wasn't possible to directly keep an equivalent object history _in_ _the_ _database_. Hence Frederik's message: "we have removed past history data from the database today".

However, it was, of course, archived, and is sitting on a server somewhere.

I don't believe this data has yet been analysed or republished; and I'm sure that analysing it will not be a simple task; but no doubt finer minds than mine are thinking about it. :)

cheers
Richard