[Osmf-talk] DWG survey on organized editing

Allan Mustard allan at mustard.net
Sun Oct 8 13:13:11 UTC 2017


Simon, et al, mon cheris, dostlarım, meine Freunde, мои друзья,

The bias of over-participation by any particular group can potentially
be controlled since the survey specifically asks if participants are
employed by an organization that performs organized mapping (as an
employee of the U.S. Department of State, I had to answer that question,
"yes", since the Humanitarian Information Unit organizes map-a-thons,
even if I do not engage in it myself).  The statistics generated by the
survey can be normalized to reflect the population as a whole if the DWG
has a sense of what proportion of the OSM mapping population consists of
organized mappers and paid mappers.

As for the bias of including only English speakers, as a U.S. diplomat
who has functioned in the multilingual world of diplomacy for 30+ years,
the easiest workaround for us when we survey our local workforce is to
sit a non-English speaker in front of the computer with a fluent speaker
of English plus the non-English speaker's language nearby and have at
it.  It works pretty well for our opinion surveys ("Is Ambassador
Mustard a) bad b) rotten c) horrible d) dysfunctional e) terrible" :-),
semantic concerns notwithstanding.  I suspect a fair number of the
mappers out there who don't know English fluently have a friend who
does, and could take the survey with a little locally available help.

cheers,
apm-wa


On 10/8/2017 2:28 PM, Simon Poole wrote:
>
> The survey is going to be skewed in many ways for example by higher
> participation from people associated with organisations with a vested
> interest in paid/organized editing. I don't really think anybody was
> even remotely assuming that this was going to be representative (doing
> a representative survey would face many many challenges in any case),
> but more simply a rough gathering of sentiment on the issue. Given
> that, I think we can live with the slight advantage we've given to
> English speakers. Just as with the suggestion to have the questions
> vetted by the community in advance, there is just so much bureaucracy
> thatВ  you can apply to such a simple action without it becoming too
> heavyweight to ever happen.
>
> Simon
>
>
> Am 08.10.2017 um 10:28 schrieb Manfred A. Reiter:
>> Hi Allan, *,
>>
>> One of my biggest reservations is the fact that the survey is being
>> conducted in English only.
>>
>> From my work for weeklyOSM I know for sure that many contributors
>> from French-speaking cultural circles or speakers from Latin America
>> will certainly be overwhelmed by the survey in English.В The same
>> applies to many of the German-speaking contributors I have spoken to.
>>
>> In this way, the result will certainly not be representative.
>>
>> cheersВ 
>>
>> 2017-10-08 9:37 GMT+02:00 Allan Mustard <allan at mustard.net
>> <mailto:allan at mustard.net>>:
>>
>>     Dear Friends,
>>
>>     I just took the survey, and have to say that my graduate degree
>>     in agricultural economics, which included a course in survey
>>     sampling and statistical methods, informed me that there was not
>>     a lot of bias in the survey, with all due respect to my OSM
>>     volunteer colleagues who have raised objections to it.В  IMHO the
>>     survey addresses a very real problem of correcting problems
>>     inserted by paid organized mappers (with which I have had to
>>     contend more than once in mapping the distant, largely unheard of
>>     country of Turkmenistan), who are thousands of miles away, and
>>     base their edits on five-year-old Bing imagery coupled with gross
>>     assumptions of what the ground truth would reveal.
>>
>>     Could the survey have been designed better?В  Yes, probably it
>>     could have been, had we engaged some social scientists with
>>     Ph.D.s in statistical survey techniques, but for the OSM
>>     Foundation's purposes, this survey will likely prove adequate to
>>     gauge the opinions of the mapping community.В  I'm glad I had the
>>     opportunity to voice my opinion about something that irritates
>>     me--having to fix errors inserted by people who aren't within a
>>     thousand miles of here, but think they know what they are doing.
>>
>>     Warm regards to all, best regards from sunny Ashgabat--and be
>>     sure to look up my Mapillary images from this weekend of a trip
>>     to see the longest dinosaur tracks in the world, once my
>>     smartphone finishes uploading them!
>>
>>     cheers,
>>     apm-wa (Allan Mustard)
>>
>>
>>     On 10/6/2017 1:01 PM, Frederik Ramm wrote:
>>>     Everyone,
>>>
>>>     On 20.09.2017 06:50, Paul Norman wrote:
>>>>     The Data Working Group is conducting a survey as part of its work on a
>>>>     policy covering paid mapping.
>>>     We plan to close the survey on Sunday night (at 23:59:59 UTC), that's 2
>>>     days and 16 hours from now. We'll then take some time to analyze the
>>>     results and share them with you, then draft a policy informed by the
>>>     results and the discussion here and elsewhere, and present that for
>>>     further discussion.
>>>
>>>     Bye
>>>     Frederik
>>
>> -- 
>> ## Manfred Reiter - -
>> ## N 49В° 25' 11.028" E 06В° 50' 47.328"В 
>> ## www.weeklyOSM.eu <http://www.weeklyOSM.eu>
>>
>>
>> _______________________________________________
>> osmf-talk mailing list
>> osmf-talk at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/osmf-talk
>
>
>
> _______________________________________________
> osmf-talk mailing list
> osmf-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/osmf-talk

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/osmf-talk/attachments/20171008/b0ec5027/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0xB5116796.asc
Type: application/pgp-keys
Size: 1714 bytes
Desc: not available
URL: <http://lists.openstreetmap.org/pipermail/osmf-talk/attachments/20171008/b0ec5027/attachment.key>


More information about the osmf-talk mailing list