<html><head><style type="text/css"><!-- DIV {margin:0px;} --></style></head><body><div style="font-family:arial,helvetica,sans-serif;font-size:10pt">Some musings on geocoding in Haiti ... any possible improvements to our tools?<br><div> </div>== Mikel Maron ==<br><span><a target="_blank" href="http://mapkibera.org/">http://mapkibera.org/</a></span><br>+254 (0) 724899738<br>mikel@osmfoundation.org<div><br></div><div style="font-family: arial,helvetica,sans-serif; font-size: 10pt;"><br><div style="font-family: times new roman,new york,times,serif; font-size: 12pt;"><font face="Tahoma" size="2">----- Forwarded Message ----<br><b><span style="font-weight: bold;">From:</span></b> Talbot Brooks <tbrooks@deltastate.edu><br><b><span style="font-weight: bold;">To:</span></b> crisismappers@googlegroups.com; crisismappers@googlegroups.com<br><b><span style="font-weight: bold;">Cc:</span></b> Rares Vernica <rares@ics.uci.edu><br><b><span
style="font-weight: bold;">Sent:</span></b> Thu, January 21, 2010 6:26:29 AM<br><b><span style="font-weight: bold;">Subject:</span></b> RE: [CrisisMappers] Search interface on the Person Finder data<br></font><br>
<title>Re: [CrisisMappers] Search interface on the Person Finder data</title>
<div dir="ltr" id="idOWAReplyText56784">
<div dir="ltr"><font face="Times New Roman" size="3" color="#000000">OK, so this requires two things, right? Solid underlying data to geocode to, and a solid geocoder. I think the latter might be a little easier to come by than the data. In talking to folks who've lived there, it sounds like addresses were doled out loosely, if at all? I can redirect some of my crew to help with this if someone can tell me a bit more how addresses work in Haiti (my gang spends a lot of time segmenting and assigning addresses to segments here in MS for 911 use).</font></div>
<div dir="ltr"> </div>
<div dir="ltr">This lends itself to another kinda related thought (sorry for my musings here) - there seems to be an emerging need for the same features in multiple geometries of like types. Eg., we would want intersection to intersection segments for routing or perhaps geocoding, but long unsegmented sections for linear referencing purposes where addresses might not exist and local knowledge edges in (eg., go this far when you see the barn - y'all know - MS style directions). I know there are dozens of data models and ways to handle this, but as OSM has become my epiphany this last week or two, I'm wondering if there's a way to start thinking of this sort of thing within that framework?</div>
<div dir="ltr"> </div>
<div dir="ltr">Sorry for the long musing - should probably have been in bed hours ago...</div>
<div dir="ltr"> </div>
<div dir="ltr">Talbot</div>
<div dir="ltr"><font face="Times New Roman" size="3" color="#000000"></font> </div></div>
<div dir="ltr" id="idSignature67518">
<div><font size="3" color="#000000">
<div class="Section1">
<div class="MsoNormal" align="left"><font face="Times New Roman"><span style="font-family: Garamond;">Talbot</span><span style="font-family: Garamond;"> Brooks</span><span style="font-family: Garamond;">, Director<br></span></font><font face="Times New Roman"><span style="font-family: Garamond;">Delta</span><span style="font-family: Garamond;"> State University GIT<br></span></font><font face="Times New Roman"><span style="font-family: Garamond;">Box</span><span style="font-family: Garamond;"> 3325<br></span></font><font face="Times New Roman"><span style="font-family: Garamond;">Cleveland</span><span style="font-family: Garamond;">, MS 38733<br></span></font><span style="font-family: Garamond;"><font face="Times New Roman">USNG (NAD 83): </font><a rel="nofollow" target="_blank"
href="http://nmviewogc.cr.usgs.gov/Run.htm?BBOX=-90.74485,33.74149,-90.7262,33.7513&CLASSIDSON=6%7C20&LAYERSON=11914%7C9942%7C10362%7C9880%7C10361%7C10360%7C23115%7C1695&LAYERSOFF=17260%7C16837%7C1652%7C11777%7C16788%7C16787%7C16962%7C16964%7C22599"><font face="Times New Roman">15SYT09823640</font></a><br></span><span style="font-family: Garamond;"><font face="Times New Roman">O: 662-846-4520<br></font></span><span style="font-family: Garamond;"><font face="Times New Roman">C: 662-588-8649<br></font></span><span style="font-family: Garamond;"><font face="Times New Roman">F: 662-846-4099<br></font></span><span style="font-family: Garamond;"><a rel="nofollow" target="_blank" href="http://gis.deltastate.edu/"><font face="Times New Roman">http://gis.deltastate.edu</font></a><br></span><span style="font-family: Garamond;"><a rel="nofollow" target="_blank" href="http://mississippi.deltastate.eduand/"><br></a></span></div>
<div class="MsoNormal"><span lang="EN">CONFIDENTIALITY. This e-mail (including any attachments) may contain confidential, proprietary and privileged information, and unauthorized disclosure or use is prohibited. If you received this e-mail in error, please notify the sender and delete this e-mail from your system.</span></div>
<div>(Please note Delta State University's email system blocks all email with .zip attachments)</div></div></font></div></div>
<div dir="ltr"><br>
<hr tabindex="-1">
<font face="Tahoma" size="2"><b>From:</b> crisismappers@googlegroups.com on behalf of Chen Li<br><b>Sent:</b> Thu 1/21/2010 12:15 AM<br><b>To:</b> crisismappers@googlegroups.com<br><b>Cc:</b> Rares Vernica<br><b>Subject:</b> Re: [CrisisMappers] Search interface on the Person Finder data<br></font><br></div>
<div>
<p><font size="2">Talbot,</font> </p>
<p><font size="2">Currently we treat addresses simply as keywords and allow search on</font> <br><font size="2">them. We would like to do location-based "fuzzy" search, but noticed</font> <br><font size="2">that most addresses are very inaccurate. If people can provide a good</font> <br><font size="2">geocoder for these addresses, we can show the information on a map.</font> </p>
<p><font size="2">Any suggestion or geocoder helper?</font> </p>
<p><font size="2">Chen Li</font> </p><br>
<p><font size="2">On Wed, Jan 20, 2010 at 9:33 PM, Talbot Brooks <tbrooks@deltastate.edu> wrote:</font> <br><font size="2">> Checked it out - very cool work. A couple of thoughts:</font> <br><font size="2">></font> <br><font size="2">> Does the fuzzy search apply to the locations where folks might be located in</font> <br><font size="2">> case an entry is fat fingered?</font> <br><font size="2">></font> <br><font size="2">> Is it possible to geocode the location of the most recent report or provide</font> <br><font size="2">> a last know location that can be mapped as part of the results?</font> <br><font size="2">></font> <br><font size="2">> Talbot Brooks, Director</font> <br><font size="2">> Delta State University GIT</font> <br><font size="2">> Box 3325</font> <br><font size="2">> Cleveland, MS 38733</font> <br><font size="2">> USNG (NAD 83): 15SYT09823640</font> <br><font size="2">> O:
662-846-4520</font> <br><font size="2">> C: 662-588-8649</font> <br><font size="2">> F: 662-846-4099</font> <br><font size="2">> <a rel="nofollow" target="_blank" href="http://gis.deltastate.edu/">http://gis.deltastate.edu</a></font> <br><font size="2">></font> <br><font size="2">> CONFIDENTIALITY. This e-mail (including any attachments) may contain</font> <br><font size="2">> confidential, proprietary and privileged information, and unauthorized</font> <br><font size="2">> disclosure or use is prohibited. If you received this e-mail in error,</font> <br><font size="2">> please notify the sender and delete this e-mail from your system.</font> <br><font size="2">> (Please note Delta State University's email system blocks all email with</font> <br><font size="2">> .zip attachments)</font> <br><font size="2">> ________________________________</font> <br><font size="2">> From: crisismappers@googlegroups.com on behalf of
Chen Li</font> <br><font size="2">> Sent: Wed 1/20/2010 11:21 PM</font> <br><font size="2">> To: crisismappers@googlegroups.com</font> <br><font size="2">> Subject: [CrisisMappers] Search interface on the Person Finder data</font> <br><font size="2">></font> <br><font size="2">> Dear all,</font> <br><font size="2">></font> <br><font size="2">> In addition to working with Tim to crawl and scrape more data, the UCI</font> <br><font size="2">> team has been studying how to develop a powerful search interface on</font> <br><font size="2">> the integrated Google data set. We have built a search prototype at</font> <br><font size="2">></font> <br><font size="2"><span>> <a target="_blank" href="http://fr.ics.uci.edu/haiticrisis">http://fr.ics.uci.edu/haiticrisis</a></span></font> <br><font size="2">></font> <br><font size="2">> which uses a similar style as the current UI at</font>
<br><font size="2">></font> <br><font size="2"><span>> <a target="_blank" href="http://haiticrisis.appspot.com/query">http://haiticrisis.appspot.com/query</a></span></font> <br><font size="2">></font> <br><font size="2">> A few features of our search interface:</font> <br><font size="2">></font> <br><font size="2">> - It supports full-text search as you type. Example: "philpe camene".</font> <br><font size="2">> - It supports fuzzy search. Example: A query "bernedeute" can find</font> <br><font size="2">> records with "BERNEDETTE". This feature is important especially due to</font> <br><font size="2">> the noisy data possibly in different languages.</font> <br><font size="2">></font> <br><font size="2">> We are sending this email for two purposes:</font> <br><font size="2">></font> <br><font size="2">> - Get your feedback on this search interface;</font> <br><font size="2">> - In case
some other people (e.g., those from Google) are working on a</font> <br><font size="2">> search interface, we want to see whether we can contribute our</font> <br><font size="2">> strengths into the search interface.</font> <br><font size="2">></font> <br><font size="2">> Thanks.</font> <br><font size="2">></font> <br><font size="2">> Chen Li@UCI</font> <br><font size="2">></font> <br><font size="2">></font> <br><font size="2">> __________ Information from ESET Smart Security, version of virus signature</font> <br><font size="2">> database 4791 (20100120) __________</font> <br><font size="2">></font> <br><font size="2">> The message was checked by ESET Smart Security.</font> <br><font size="2">></font> <br><font size="2"><span>> <a target="_blank" href="http://www.eset.com/">http://www.eset.com/</a></span></font> <br><font size="2">></font> <br><font size="2">></font> <br><font size="2">> --</font>
<br><font size="2">> You received this message because you are subscribed to the Google Groups</font> <br><font size="2">> "CrisisMappers" group.</font> <br><font size="2">> To post to this group, send email to crisismappers@googlegroups.com.</font> <br><font size="2">> To unsubscribe from this group, send email to</font> <br><font size="2">> crisismappers+unsubscribe@googlegroups.com.</font> <br><font size="2">> For more options, visit this group at</font> <br><font size="2"><span>> <a target="_blank" href="http://groups.google.com/group/crisismappers?hl=en">http://groups.google.com/group/crisismappers?hl=en</a>.</span></font> <br><font size="2">></font> <br><font size="2">></font> <br><font size="2"> </font> </p>
<p><font size="2">__________ Information from ESET Smart Security, version of virus signature database 4791 (20100120) __________</font> </p>
<p><font size="2">The message was checked by ESET Smart Security.</font> </p>
<p><font size="2"><a rel="nofollow" target="_blank" href="http://www.eset.com/">http://www.eset.com</a></font> <br><font size="2"></font> </p></div>-- <br>
You received this message because you are subscribed to the Google Groups "CrisisMappers" group.<br>
To post to this group, send email to crisismappers@googlegroups.com.<br>
To unsubscribe from this group, send email to crisismappers+unsubscribe@googlegroups.com.<br><span>
For more options, visit this group at <a target="_blank" href="http://groups.google.com/group/crisismappers?hl=en">http://groups.google.com/group/crisismappers?hl=en</a>.</span><br>
</div></div>
<!-- cg3.c2.mail.re1.yahoo.com compressed/chunked Wed Jan 20 02:11:10 PST 2010 -->
</div></body></html>