[Openstreetmap] Re: Openstreetmap Digest, Vol 6, Issue 3

wabanstar waban_star at yahoo.com
Mon Feb 7 16:39:40 GMT 2005


Just a few thoughts on you latest post. If you are
using commercial resources use at least two that
verify the same feature or you face copyright
infringment, make sure they are by different publisher
and or owners. You need a GPS that is highly accurate
12 meters to 2 cm and can download tracks and points
and can interface with your computer. The best is
Trimble, my Garmin GPSII+ works well (12 channel
reciever and dataported), a great freeware is
TrackMaker. don't buy the cheapest. Another topic
regarding quality controll is that to be highly
accurate you need a good protocol or rule and
standards going in, what is good data, who cleans up
the overlaps, what are your definitions of major vs
minor roads, what constitutes a road, list the
different types of roads you are aware of and how you
might classify them, do you want to show all routes of
traffic (which ones?), what is your addressing sceme,
what is your water, parks, infrastructure shemes as
well. what types of polgons (water, etc.)You need
verification for all features, You need projection
accuracy accross all datasets, you need ways of
tileing information so that it can be seperated and
rejoined as you will find
that you dataset grows to amounts to even load on a
PC.
I know how one of the largest closed source providers
operates and as an example you could spend hundreds of
thousands of hours and only cover a small portion of
one continent, the valuable part. You could also spend
millions aquiring datasets and millions more storing
the data. Just a bit of a reality check. Your
advantage is that you have free data collectors, and
programmers. 
A lot of boring work, ie. administration, data storage
and quality checks and enforcement have to be done as
well to make any challenge to the closed source model.

The closed source would pay to be able to use your
information, They might provide free base mapping for
updates of new geometry from your group. 
     You also need a scientifically rigorous data
structure that is not corruptable,only provide, I
believe, by commercial unix type providers such as
oracle.
PS: Another thought on sat imagery is that it doesn't
always provide a good view of city streets, if it
isn't just overhead at 90 degrees the buildings block
the grid.     
--- openstreetmap-request at vr.ucl.ac.uk wrote:

> Send Openstreetmap mailing list submissions to
> 	openstreetmap at vr.ucl.ac.uk
> 
> To subscribe or unsubscribe via the World Wide Web,
> visit
> 
>
http://bat.vr.ucl.ac.uk/cgi-bin/mailman/listinfo/openstreetmap
> or, via email, send a message with subject or body
> 'help' to
> 	openstreetmap-request at vr.ucl.ac.uk
> 
> You can reach the person managing the list at
> 	openstreetmap-owner at vr.ucl.ac.uk
> 
> When replying, please edit your Subject line so it
> is more specific
> than "Re: Contents of Openstreetmap digest..."
> 
> 
> Today's Topics:
> 
>    1. Buying Ikonos data (Hugh Barnard)
>    2. Re: Buying Ikonos data (Saul Albert)
>    3. Re: Buying Ikonos data (Saul Albert)
>    4. Re: Buying Ikonos data (Hugh Barnard)
>    5. Re: Buying Ikonos data (Andy Armstrong)
>    6. Re: Buying Ikonos data (Lester Caine)
>    7. Re: Buying Ikonos data (Andy Armstrong)
>    8. Re: Buying Ikonos data (Lester Caine)
> 
> 
>
----------------------------------------------------------------------
> 
> Message: 1
> Date: Sun, 6 Feb 2005 14:19:56 -0000
> From: "Hugh Barnard" <hugh.barnard at hughbarnard.org>
> Subject: [Openstreetmap] Buying Ikonos data
> To: <openstreetmap at vr.ucl.ac.uk>
> Message-ID: <008701c50c56$ef587640$0300000a at max>
> Content-Type: text/plain;	charset="iso-8859-1"
> 
> Hi Folks
> 
> I'd also be interested in banding together to buy
> some Ikonos data. One
> suggestion would be
> to take a small bit off them and (all) see what we
> could do with it. On the
> basis of some kind
> of progress we could refine what we want from
> them/buy another bit..
> 
> Just to declare interest, I'm really only interested
> in bits of London. I
> think most of it (except maybe
> the parks, not really relevant) will represent the
> same level of practical
> problems...
> 
> Best regards Hugh
> 
> http://hughbarnard.org
> 
> 
> 
> 
> ------------------------------
> 
> Message: 2
> Date: Mon, 7 Feb 2005 01:45:32 +0000
> From: Saul Albert <saul at twenteenthcentury.com>
> Subject: Re: [Openstreetmap] Buying Ikonos data
> To: Hugh Barnard <hugh.barnard at hughbarnard.org>
> Cc: openstreetmap at vr.ucl.ac.uk
> Message-ID:
> <20050207014532.GC74491 at chinabone.lth.bclub.org.uk>
> Content-Type: text/plain; charset=us-ascii
> 
> hi hugh, folks,
> 
> Great!
> 
> Jo and I went to cambridge today and had a really
> good chat with Rufus Pollock
> from okfn/ffiuk and he gave us a whole lot of
> motivation and spanked our brains
> about getting a bit more organised... both to get
> our open geodata projects
> further underway, and to start the necessary
> political lobbying with a suitable
> groundswell of use cases, zealots and code.
> 
> We thought about a few different ways to take things
> forward. Here are some of
> the ideas that came out of this list/people and lots
> and lots of conversations:
> 
> - We could do an Open Knowledge Forum (the
> okfn.net's meet up format) in London
>   in April '05, a kind of launch meeting to get lots
> of people on board
> 
> - We would need to prepare for this. We could do a
> serious getting things done
>   hack session in the first week of march (physical
> or remote, but we'll
>   provide a well hackbase in Bristol if you want to
> stay and hack for a few
>   days -> a week)
> 
> - There was an outline of a plan of action / needs..
> 
> It went something like:
> 
> First order:
> 
> * A data repository that can be mirrored /
> syndicated
> * Lots of methods of adding to that data store
> * Many routes to openstreetmap - inc. GPS traces and
> Ikonos/other derivatives
> * Some compelling use cases or at least feedback
> that makes adding data
>   worthwhile seeming.
> 
> Second order:
> 
> * Line simplification / segmentation stuff
> * Annotation / Metadata adding and sharing stuff
> 
> Many bits of this are already in place, but maybe we
> can get down to some
> serious discussions and tick them off one by one,
> then find the shortest path
> to having them all, aim then fire.
> 
> THEN we can start to build a significant political
> attack on closed geodata,
> the inspire directive etc.. becase we'll have an ace
> in the hole (is that the
> phrase..?) basically, we'll be saying to the OS &c.
> sort us out, or we'll sort
> ourselves out, and you'll be sorted out of business
> in the process.
> 
> Is this realistic? Or even desirable? I think so,
> but then it is late.
> 
> X
> 
> s.
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> On Sun, Feb 06, 2005 at 02:19:56PM -0000, Hugh
> Barnard wrote:
> > Hi Folks
> > 
> > I'd also be interested in banding together to buy
> some Ikonos data. One
> > suggestion would be
> > to take a small bit off them and (all) see what we
> could do with it. On the
> > basis of some kind
> > of progress we could refine what we want from
> them/buy another bit..
> > 
> > Just to declare interest, I'm really only
> interested in bits of London. I
> > think most of it (except maybe
> > the parks, not really relevant) will represent the
> same level of practical
> > problems...
> > 
> > Best regards Hugh
> > 
> > http://hughbarnard.org
> > 
> > 
> > _______________________________________________
> > Openstreetmap mailing list
> > Openstreetmap at vr.ucl.ac.uk
> >
>
http://bat.vr.ucl.ac.uk/cgi-bin/mailman/listinfo/openstreetmap
> > 
> 
> -- 
> -- http://chinabone.lth.bclub.org.uk/~saul/
> 
> 
> 
> 
=== message truncated ===



		
__________________________________ 
Do you Yahoo!? 
Yahoo! Mail - 250MB free storage. Do more. Manage less. 
http://info.mail.yahoo.com/mail_250




More information about the talk mailing list