[Talk-us] Why does the USA currently lag in OSM map quality?

Tue Feb 17 21:17:09 UTC 2015

That's a very good question, and could be interesting in a number of ways.
  Oregon would be easier for a few reasons, namely that the state fair is
centralized in one reasonably well connected city with a lot of indoor
space with electricity (because it tends to be fairly predictably wet year
round), with folks from rural areas traveling great lengths from the desert
rangelands and wine regions to get their goods to fair to promote
themselves and learn what else is going on in the state.  Industry also
makes use of this time, you can rest assured every corner of agribusiness
has a booth up someplace to show off the latest, largest and greatest tech
in ag.  OSM could be a game changer if a tech minded farmer could leverage
it and contribute.

Oklahoma hits the opposite extreme, there's multiple state fairs differing
weeks, largely outdoors, and with the exception of the Oklahoma City State
Fair and the Tulsa State Fair, in fairly disconnected places.  And given
that these are predictably held during the hottest time of the year and
draw broadly from the public, it's hard to have a thought much more intense
than "it's hot, I'm hot, and maybe I shouldn't have gone to the cattle show
right after having three corndogs and riding the Tilt a Hurl".  Thinking at
least for this region, the Boat Show or the RV show (which are winter
indoor events) are more likely candidates (owing largely to Oklahoma
seeming to be a huge draw for RV touring and a number of lakes rivaling
Minnesota).  Could be interesting to see if it's possible to gamify this a
bit with something along the lines of walking papers and/or a scavenger
hunt for our state fair system.  Or just any of the tourist areas in
general, since they tend to be somewhat counterintuitive to navigate, under
dense tree cover rendering aerial imagery useless, and have very small
permanent population.

On Tue, Feb 17, 2015 at 2:57 PM, Eleanor Tutt <eleanor.tutt at gmail.com>
wrote:

> +1 to Bryce's comments about reaching out to existing communities with
> shared interests who may be using other tools/methods currently.
>
> This conversation reminds me a of a presentation I saw on Missouri's
> recent attempts to survey people about internet access and map broadband
> coverage, including in rural areas.  They had luck with outreach by setting
> up booths at State and County Fairs - that seemed to be where enough people
> gathered at once to make the outreach worth the time & effort (although
> they often used pins and paper maps to gather their data given spotty
> coverage).  I wonder if anyone has ever had an OSM State Fair mapping party?
>
> Eleanor
>
> On Tue, Feb 17, 2015 at 2:34 PM, Paul Johnson <baloo at ursamundi.org> wrote:
>
>> On Tue, Feb 17, 2015 at 1:15 PM, Ian Dees <ian.dees at gmail.com> wrote:
>>
>>> On Tue, Feb 17, 2015 at 2:06 PM, Paul Johnson <baloo at ursamundi.org>
>>> wrote:
>>>
>>>> I'm thinking if they wanted broader input, they'd use the mailing list
>>>> and not the forum.
>>>>
>>>> But I think a big part of it is the US is very large, and very empty.
>>>> Plot out a wall size map of the US, now pin the tail on the map.  Unless
>>>> you bumped a wall on the way there or have an acute sense of space with
>>>> your eyes closed and managed to stab somewhere in the Coruscant-like DFW
>>>> megaplex (seriously, drive US 75 south into DFW at night and you'll go over
>>>> a rise near Anna, TX from which DFW appears to roll from where you are all
>>>> the way to the horizon ahead of you; it's probably geographically larger
>>>> than several of the smallest states by area, possibly even combined), you
>>>> probably just pinned the tail to a part of the country that is just as
>>>> empty now as before Manifest Destiny.  Possibly even emptier given The
>>>> Removal and two waves of urbanization.
>>>>
>>>> People map where they know.  People know where they are.  Where are the
>>>> people in the US?  Well, if you take the top ten most populated
>>>> metropolitan statistical areas in the US,  you account for 97% of the US
>>>> population, and with the exception of Dallas-Fort Worth, Texas, they're all
>>>> within a day's bicycle ride or less of an ocean (I'm including places as
>>>> far in as Portland given I've made that ride to the ocean by bicycle
>>>> before, and I'm not even horribly fit or in great shape).  Extend it out to
>>>> the top 100 metropolitan statistical areas, and you leave a very small
>>>> fraction of 1% of the US population to account for the remaining 281
>>>> metropolitan statistical areas and 536 micropolitan statistical areas.
>>>>
>>>> TL;DR, Hitchhiker's Guide validated version:  When randomly sampled by
>>>> township and range, averaged out and rounded to the nearest integer, the
>>>> population of the US is 0.
>>>>
>>>
>>> It's easy and fun to hypothesize about why OSM is crummy in the US, but
>>> it's vastly more useful to think of ways to improve it.
>>>
>>
>> I don't think I'm speaking in the hypothetical here.
>>
>> Increasing awareness through mapping parties/events seems to help a lot
>>> in urbanized areas, but we still haven't figured out how to apply that to
>>> the rest of the country. Tools like MapRoulette and fixme can guide
>>> existing mappers to areas that are probably in need of help. Are there
>>> methods of remote sensing (street-level imagery, data from other places on
>>> the internet) that could help us with the locality problem?
>>>
>>
>> TL;DR for the next two paragraphs:  OSM tends to fall somewhere on or
>> between esteem and self-transcendence on Maslow's hierarchy.  The rural
>> extreme struggles with the physiological and can't take safety for granted,
>> and it's going to take something on the order of a New Deal that directly
>> benefits with improving access to sanitation, food, water, electricity, and
>> internet to less than 1% of the population that is going dozens of miles
>> for food and water (or collecting both in situ), generating what limited
>> electricity they have access to themselves, and whose trip to the toilet
>> still involves shoes and a shovel, to do much of anything to change this
>> within my lifetime.  Given the political climate of the country, I think it
>> goes without saying that this isn't going to happen.
>>
>> Speaking from experience, OSM is a bandwidth intensive project,
>> particularly when working with geography so freaking huge as the US.  And
>> for the sake of this conversation, I'm lumping in the likes of Kellyville,
>> Oklahoma, with all of it's 500 acres and 1100 people simply because that's
>> large enough to have electricity, indoor plumbing and store, and some hope
>> of getting anything viable in terms of internet access (even if only
>> through a limited bandwidth library/cafe wifi hotspot) as "urban."  That's
>> relatively easy to "spark" the same way as it is in larger places:  Just
>> find the like-minded individuals, spark the interest and you'll get crazy
>> detailed maps for their part of the world, and some interesting
>> applications come out of it.  I've seen it happen in Portland (where it
>> gripped the imagination of my hometown a bit more tightly than I expected),
>> I'm watching it start to happen again in Tulsa now.  Heck, mapping parties
>> might be more natural given this context simply because a cafe or the
>> library might be the only reasonably passable internet connection in town
>> that can fetch a mapping party weekend's worth of data and upload it back
>> again in under half a day (literally).
>>
>>
>>> Any other ideas for how to make the rural US better?
>>>
>>
>> Start looking for the modern day explorers and get ready to shell out
>> just like last time this country had a cartography problem.  I can't really
>> see this as even being something we could buy eleventybillion HITs on
>> Mechanical Turk to solve.
>>
>> _______________________________________________
>> Talk-us mailing list
>> Talk-us at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk-us
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-us/attachments/20150217/6dd9a235/attachment.html>