[Imports] Import guidelines - official or not? was: [Talk-us] parcel data in OSM

[Frederik Ramm]

Hi,

> Can you provide a link to the guidelines the DWG uses?

...

> Please note that this guidelines on this page
> (http://wiki.openstreetmap.org/wiki/Import/Guidelines) are the ones
> cited by DWG members when they request sysadmins to place blocks on
> accounts that are importing

I'm kind of unsure what exactly all this is about.

There are no hard and fast rules; there is no rule that says: "If you do 
this, then you will definitely get blocked", and there is no rule that 
says "if you do this, then you will definitely not get blocked".

(For example, DWG may occasionally put a short time block on someone 
just for not answering questions sent to him by other community members, 
but of course not everyone who chooses not to reply to a personal 
message has to expect being blocked.)

DWG members can themselves place short-time blocks on users without the 
involvement of sysadmins, and will do so at their discretion. This 
measure is normally used to stop someone who is doing something 
detrimental, and not as some form of punishment. A DWG member does not 
need to cite a guideline violation to place a block on someone; it 
could, for example, simply be a "wait, let's talk about this before you 
continue" block.

There was some ruckus recently when the French community was upset about 
DWG requiring a separate import account; they demanded to see the 
written guideline, or board resolution, or any other "law" that 
described this requirement. When pointed to the page you quote above, 
which says:

"You must not use your standard OSM user account."

the French community pointed out that this was changed on 22 March 2012 
by Richard Fairhurst (who is a member of the OSMF board but not of DWG) 
and before that it had read "Create a new user for the import. Do not 
use your standard OSM user account.", without the "must not".

Some members of the French community felt that the old wording meant 
that the separate account was a recommendation only and that they should 
have had a say, or at least be informed, about that change of policy 
that came into effect on 22 March 2012.

In reality, there was never a change of policy, and even before that 
date DWG would occasionally ask people to create a separate account for 
their import.

That's why I'm wary when someone asks me for an applicable policy - I am 
in absolutely no mood to set up some kind of "law" where "lawyers" can 
then discuss whether the wording "Do not use your standardard OSM user 
account" means the same as "you must not use your standard OSM user 
account" or not.

The page http://wiki.openstreetmap.org/wiki/Import/Guidelines is 
currently not in a form that I would like to see DWG stamping with a 
"DWG official guidelines" seal because it doesn't have the big 
"community is more important than importing" message on top that it used 
to have half a year ago. As you know, I think that it is dangerous if 
people get the idea that they could substitute people with imports. But 
on the technical stuff, I think it is relatively sound, and anyway I am 
only one of 8 in the DWG.

There's also the very old document

http://wiki.openstreetmap.org/wiki/Automated_Edits_code_of_conduct

and the slightly younger

http://wiki.openstreetmap.org/wiki/Mechanical_Edit_Policy

all of which should perhaps be merged into a better document, then 
discussed by DWG and signed off by OSMF in some form - either by the 
OSMF board or by the Management Team (which is a 
currently-not-very-visible group that the OSMF board hopes will some day 
take over the responsibility of implementing and publishing all sorts of 
guidelines that working groups think are necessary). After being made 
"official" in this way, the document would go onto the OSMF site where 
it would remain unchanged until "official" updates are decided upon.

Making such a consolidated guideline is part of DWG's plan for the near 
future. As you can see, these documents come from slightly different 
backgrounds and their focus is different, and if you read them with a 
lawyer's eyes you might also find contradictions (of the kind where one 
document says you must do something and the other only says it would be 
a good idea to do something, or so).

There's also the question of subsidiarity - or, as I cynically put it in 
another post, in how far the project must allow a partial community to 
shoot itself in the foot with an import or mechanical edit). This is 
something that was also brought up by the French community but it might 
apply to the import-hungry US too: Do import guidelines have to be the 
same the world over?

To come back to your original question: Someone who wants to make an 
import should certainly have read the three pages mentioned in this 
post; they together form the canon of guidelines that will apply to most 
imports. They are not however official, nor exhaustive; they cannot be 
official because anyone can edit them, and they are not exhaustive 
because it is always possible that there's an unwritten rule you are 
expected to follow - that's why we always ask people to discuss their 
plans, expecting that anything out of the ordinary would come to light 
at that stage. It is also possible that for a certain import, not all of 
the guidelines actually apply.

Now before someone pulls out the constitutional heavy guns ("how can I 
do anything in a project where I don't even know what the rules are"), 
let me say that the consequence of violating a rule that you didn't know 
in OSM is, at worst, a short-time block and a revert of some edits. You 
need to actively and knowingly play foul to get into more trouble than 
that. Not having hard-and-fast official rules also means there's no rule 
lawyering - no way for someone to find a loophole and claim immunity 
because they follwed the letter and not the spirit of a rule. Once we 
devise "official" rules they will certainly contain some paragraph to 
avoid such lawyering.

I hope this explanation was helpful.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"