PeopleFinder: Where are we and where are we going?

[Tim Schwartz]

Food for thought:

We're now starting day 6 after the earthquake and a lot of people are
starting to talk about "how much we've accomplished" -- i'm just not
quite sure that we can't accomplish more. I am concerned that without
the google product being visibly (ie on the website) shown to be
helpful, i'm not sure how much people will be using it.

Google: YOU NEED TO GET DATA ABOUT HOW MANY PEOPLE ARE IN THE SYSTEM
AND A LIVE FEED OF WHEN MISSING PEOPLE PAGES GET UPDATES!!!

Secondly, from taking to my haitian friend yesterday, it was very
clear that he had talked to his friends and family on the ground, and
was pretty much up to date on who was alive or missing, and probably
had better information than google ever could. Because of the fact
that in this situation SMS and cellphones are working (to an extent),
I think we should reassess about just what we're doing here. Who are
we helping? It is clear that this is a valuable application, I just
think with the timeframe here it is becoming less about survivors, and
maybe more about the deceased. This isn't to say that I believe we
should slow down work in any way, but I think we should be aware of
the facts on the ground and thinking how we can help the situation.

Finally, don't SLOW DOWN. There is going to be a next disaster and if
we can coordinate efforts this time around we will be much better off
for next time. When the next disaster strikes, google should already
have a team ready to deploy and integrate asap. Everyone has done a
tremendous job so far, and everyone should feel good about the time
and dedication they have been putting in. It's hard to drop everything
and just go go go, but it's how we can help, so keep it up.

Lastly, it is clear that the SMS tool is AWESOME! Great job everyone
on that front, below is the breakdown from Robert on incoming txt
messages. We do not currently have the capability to SMS back to
Haiti, but the phone numbers are attached to every incoming message
and going to volunteers on the ground.

Cheers,
Tim

As of Midnight (EST) Sunday 17

[20:59:46] Robert Munro: Here is the breakdown:

707 texts, and 402 with information deemed useful

28.06% Missing Persons
12.23% Water shortage
10.07% Food distribution
10.07% Asking to forward a message
9.35% Earthquake and aftershocks
5.04% Persons News
4.32% People trapped
3.60% Emergency
2.88% Health services
2.16% Medical Emergency
2.16% Died bodies management
2.16% Response
1.44% Looting
1.44% Shelter
5.04% Other

-All but 'Asking to forward a message' (and 'Other') correspond to the
WHO categories of the same name
-125 where the location was detailed enough that volunteers could plot
the latitude and longtitude

On Sun, Jan 17, 2010 at 12:49 AM, Tim Schwartz <tima...@gmail.com> wrote:
> +brianzimmer
>
> Chen,
>
> If you can concentrate on the CNN iReports that would be great. I
> ended up using the ICRC scraper I had so I could make it feed directly
> into google in realtime, it will continue to update google as things
> to into it. Brian Zimmer has created a system for volunteers to go
> through cnn ireport data parsed by chen's group and put the data into
> google by hand. This system will be up sometime in the morning PST.
> Chen, please pass him your updated scrapes of CNN, and remove all
> extraneous columns.
>
> 16,000 ICRC entries should be going into google within 30 min, that
> will confirm that the API is working and it will be opened tomorrow.
>
> good night and good luck,
> tim
>
>
> On Sat, Jan 16, 2010 at 11:56 PM, Chen Li <che...@ics.uci.edu> wrote:
>> We put three dump data files at http://fr.ics.uci.edu/ . Tim is using
>> the files to do the conversion for the Google repository.  We will
>> continue our crawling and scraping efforts.
>>
>> Chen Li
>>
>> On Sat, Jan 16, 2010 at 6:27 PM, Chen Li <che...@gmail.com> wrote:
>>> I want to confirm that our UC Irvine team of 7 people will deliver the
>>> data set Tim mentioned as different mysql dumps according to the web
>>> sites.
>>>
>>> Chen Li
>>> Associate Professor
>>> UC Irvine
>>> http://www.ics.uci.edu/~chenli/
>>>
>>> On Sat, Jan 16, 2010 at 6:12 PM, Andy Carvin <ACa...@npr.org> wrote:
>>>>
>>>> Don't forget that we now have a google group for this conversation; please
>>>> be sure to cc it in all correspondences so the broader group of people
>>>> working on the project has access to this information.
>>>>
>>>> person...@googlegroups.com
>>>>
>>>> (though we may want to quickly change the name to peoplefinder so it
>>>> corresponds with the name lots of peoplea are using.)
>>>>
>>>> thanks,
>>>> ac
>>>>
>>>> -----Original Message-----
>>>> From: Tim Schwartz [mailto:tima...@gmail.com]
>>>> Sent: Sat 1/16/2010 8:53 PM
>>>> To: che...@gmail.com
>>>> Cc: Mark Prutsalis; Pablo Mayrgundter; Ben Rigby; Luke Beckman; Stanton,
>>>> Katie; Andy Carvin; Tomas Krag; sundeep ahuja; Josh Nesbit; David Kobia;
>>>> Patrick Meier; dwade; Maria Mulcahy; csik; jrochelle; rasmussen; Cohen,
>>>> Jared A; Hall, Suzanne K; Dowd, Katie W; Erik Hersman; Brian Herbert;
>>>> jst...@rhok.org; Jacob Colker; da...@beextra.org; d...@google.com; Mike Yang;
>>>> eff Martin; Bobby Fishkin; Praneeth; Dominic König; pr...@google.com
>>>> Subject: Re: PeopleFinder Collaboration
>>>>
>>>> Major Update From The Front Lines:
>>>>
>>>> I just got off the phone with Chen Li from UC Irvine. His group has
>>>> been working all day at scraping and parsing content from:
>>>>
>>>> CNN iReports, ICRC, and other unstructured sources, they are currently
>>>> giving me a data dump of two things
>>>>
>>>> 1) totally structured data from ICRC that can go directly into google
>>>> tonight
>>>> 2) the unstructured data from all over the web
>>>>     - this data will come in a form of Title, Text Blurb, and Image
>>>>
>>>> I am about to start building an interface for volunteers to comb
>>>> through the unstructured data and input it into google. This will be
>>>> built on the code repository that ran haitianquake.com and is now
>>>> running the SMS volunteer interface by Brian Herbert. If anyone out
>>>> there is willing to work I could use help. I will be working for the
>>>> next 4 hours, so if anyone out there is in another continent and take
>>>> over in 4 hours that would be incredibly helpful. It would be great to
>>>> have this interface done by tomorrow and have volunteers ready to go
>>>> late morning.
>>>>
>>>> Keep it up!
>>>>
>>>> -tim
>>>>
>>>> 2010/1/16 Dominic König <dom...@nursix.org>:
>>>>> Oops,
>>>>>
>>>>> forgot to mention that though the document defines a namespace for pfif,
>>>>> it
>>>>> finally doesn't use it - meaning: all elements of that document except the
>>>>> <pfif:pfif> element would _not_ be PFIF in that layout.
>>>>>
>>>>> It could be easier to specify the namespace without any prefix and just
>>>>> use
>>>>> <pfif>. Otherwise the prefix must be used consistently.
>>>>>
>>>>> Dominic König
>>>>>
>>>>
>>>>
>>>>
>>>
>>
>

[TEL 4RENT]

Tim,

Thanks for the update.

Do you know if people on the ground have access to internet on their phone?

regards,
tel4rent

[Tim Schwartz]

I would assume not. SMS is the best your going to get for a while.

-tim

[Philip Ashlock]

Google: YOU NEED TO GET DATA ABOUT HOW MANY PEOPLE ARE IN THE SYSTEM
AND A LIVE FEED OF WHEN MISSING PEOPLE PAGES GET UPDATES!!!

  

Additionally maybe we can slightly rework the Google branding on the Person Finder app, Maybe it can be called Haiti Crisis Response instead of Google Crisis Response. If needed, then maybe we could add the Google logo to the bottom of the page along with the others who are actively working with it or pointing towards it like The White House, U.S. State Department, The New York Times, and hopefully CNN, Facebook et all. Sometimes Google accidentally mixes its brand in with open platforms (which is why the "G" in GTFS was just changed from Google to General.)

I've been wondering if the Google branding is making it difficult to fully bring on CNN and Facebook to point to or integrate with Person Finder.

CNN is definitely still pushing iReports as the tool for missing persons and reporting issues on the ground. That'd be fine if it plugged into the APIs and then it could just be like how Reuters worked to boost its image by association with 4636 - http://www.reuters.com/article/idUSTRE60G16420100117

Things work so much better when people don't care about getting credit for their work.

Phil

[Franklin Ho]

FYI, CNN's iReport system will be using the Person Finder API to push and pull entries from our database. We're working on that as we speak.

-Franklin
-------
Franklin Ho
Account Associate, Ad Serving Operations | Crisis Response Team
f...@google.com
P: (650)253-2699
C: (415)374-9008

[Prem Ramaswami]

in progress.

On Mon, Jan 18, 2010 at 5:17 PM, Jeff Martin <jeffm...@google.com> wrote:

Possible/desired to embed a counter in the gadget? Doesn't need to be exact, maybe in increments of 100 or 500.

+ elaine for comment

On Mon, Jan 18, 2010 at 1:10 PM, Pablo Mayrgundter <p...@google.com> wrote:

Hey Tim,

We have been overwhelmed and needed to prioritize data import/export
and quality in the first few days, and then stabilize our process over
the weekend to make sure nothing was falling through, and opened up
our process on our blog and published our tasks and data exchange
efforts to get a sanity check from others that we're doing it right.

All that said, you're right :)  Stats work has been underway and and
we expect to have it out this afternoon.  I would have liked to have
it done faster.

Thanks,
Pablo

---
http://sites.google.com/site/haitireliefwiki/

--
Jeffery Martin
Business Product Manager  |  Google Crisis Response Team Lead

--
=======================
Prem Ramaswami
Product Manager - Google
http://www.google.co.in/internetbus
=======================

[Ka-Ping Yee]

The counter is live.

— Ping

--
— Ping
Google PowerMeter team

[Tim Schwartz]

Franlin plz note that were converting th first 6000 reports with volunteers right now.

-tim

Sent via BlackBerry from T-Mobile

---

From: Franklin Ho <f...@google.com>

Date: Mon, 18 Jan 2010 14:01:33 -0800

To: <person...@googlegroups.com>

Subject: Re: PeopleFinder: Where are we and where are we going?

Tim Schwartz wrote:

Oops,

forgot to mention that though the document defines a namespace for
pfif,
it
finally doesn't use it - meaning: all elements of that document
except the
<pfif:pfif> element would_not_ be PFIF in that layout.

It could be easier to specify the namespace without any prefix and
just
use
<pfif>. Otherwise the prefix must be used consistently.

Dominic König

                

[Christina]

Hi all,

I'm Christina and I'm covering for Tim while he's gone for a couple of
hours. We have just finished setting up the "Streamlined Workflow"
version of the site for volunteers to quickly and efficiently copy
information from an initial batch of 300 ireports to the Google
database, and we are in the process of sending it out to the
volunteers right now. Please let me know if there's anything I should
be aware of that might be relevant to this process.

Thanks and respect,
Christina

On Jan 18, 5:25 pm, "Tim Schwartz" <timat...@gmail.com> wrote:
> Franlin plz note that were converting th first 6000 reports with volunteers right now.
>
> -tim
> Sent via BlackBerry from T-Mobile
>

> -----Original Message-----
> From: Franklin Ho <f...@google.com>
> Date: Mon, 18 Jan 2010 14:01:33
> To: <person...@googlegroups.com>
> Subject: Re: PeopleFinder: Where are we and where are we going?
>
> FYI, CNN's iReport system will be using the Person Finder API to push and
> pull entries from our database. We're working on that as we speak.
>
> -Franklin
> -------
> Franklin Ho
> Account Associate, Ad Serving Operations | Crisis Response Team
> f...@google.com
> P: (650)253-2699
> C: (415)374-9008
>

> On Mon, Jan 18, 2010 at 1:58 PM, Philip Ashlock <p...@ashlock.us> wrote:
>
> >  Google: YOU NEED TO GET DATA ABOUT HOW MANY PEOPLE ARE IN THE SYSTEM
> > AND A LIVE FEED OF WHEN MISSING PEOPLE PAGES GET UPDATES!!!
>
> > Additionally maybe we can slightly rework the Google branding on the Person
> > Finder app, Maybe it can be called Haiti Crisis Response instead of Google
> > Crisis Response. If needed, then maybe we could add the Google logo to the
> > bottom of the page along with the others who are actively working with it or
> > pointing towards it like The White House, U.S. State Department, The New
> > York Times, and hopefully CNN, Facebook et all. Sometimes Google
> > accidentally mixes its brand in with open platforms (which is why the "G" in

> > GTFS <http://code.google.com/transit/spec/transit_feed_specification.html>was just changed from Google to General.)

>
> > I've been wondering if the Google branding is making it difficult to fully
> > bring on CNN and Facebook to point to or integrate with Person Finder.
>
> > CNN is definitely still pushing iReports as the tool for missing persons
> > and reporting issues on the ground. That'd be fine if it plugged into the
> > APIs and then it could just be like how Reuters worked to boost its image by
> > association with 4636 -
> >http://www.reuters.com/article/idUSTRE60G16420100117
>
> > Things work so much better when people don't care about getting credit for
> > their work.
>
> > Phil
>
> > Tim Schwartz wrote:
>
> > I would assume not. SMS is the best your going to get for a while.
>
> > -tim
>

> > On Mon, Jan 18, 2010 at 12:47 PM, TEL 4RENT <tel4r...@gmail.com> <tel4r...@gmail.com> wrote:
>
> >  Tim,
>
> > Thanks for the update.
>
> > Do you know if people on the ground have access to internet on their phone?
>
> > regards,
> > tel4rent
>

> > On Sun, Jan 17, 2010 at 12:49 AM, Tim Schwartz <timat...@gmail.com> <timat...@gmail.com> wrote:
>
> >  +brianzimmer
>
> > Chen,
>
> > If you can concentrate on the CNN iReports that would be great. I
> > ended up using the ICRC scraper I had so I could make it feed directly
> > into google in realtime, it will continue to update google as things
> > to into it. Brian Zimmer has created a system for volunteers to go
> > through cnn ireport data parsed by chen's group and put the data into
> > google by hand. This system will be up sometime in the morning PST.
> > Chen, please pass him your updated scrapes of CNN, and remove all
> > extraneous columns.
>
> > 16,000 ICRC entries should be going into google within 30 min, that
> > will confirm that the API is working and it will be opened tomorrow.
>
> > good night and good luck,
> > tim
>

> > On Sat, Jan 16, 2010 at 11:56 PM, Chen Li <che...@ics.uci.edu> <che...@ics.uci.edu> wrote:
>
> >  We put three dump data files athttp://fr.ics.uci.edu/. Tim is using

> > the files to do the conversion for the Google repository.  We will
> > continue our crawling and scraping efforts.
>
> > Chen Li
>

> > On Sat, Jan 16, 2010 at 6:27 PM, Chen Li <che...@gmail.com> <che...@gmail.com> wrote:
>
> >  I want to confirm that our UC Irvine team of 7 people will deliver the
> > data set Tim mentioned as different mysql dumps according to the web
> > sites.
>
> > Chen Li
> > Associate Professor

> > UC Irvinehttp://www.ics.uci.edu/~chenli/

>
> > On Sat, Jan 16, 2010 at 6:12 PM, Andy Carvin <ACar...@npr.org> <ACar...@npr.org> wrote:
>
> >  Don't forget that we now have a google group for this conversation;
> > please
> > be sure to cc it in all correspondences so the broader group of
> > people
> > working on the project has access to this information.
> > person...@googlegroups.com
>
> > (though we may want to quickly change the name to peoplefinder so it
> > corresponds with the name lots of peoplea are using.)
>
> > thanks,
> > ac
>
> > -----Original Message-----
> > From: Tim Schwartz [mailto:timat...@gmail.com <timat...@gmail.com>]
> > Sent: Sat 1/16/2010 8:53 PM
> > To: che...@gmail.com
> > Cc: Mark Prutsalis; Pablo Mayrgundter; Ben Rigby; Luke Beckman;
> > Stanton,
> > Katie; Andy Carvin; Tomas Krag; sundeep ahuja; Josh Nesbit; David
> > Kobia;
> > Patrick Meier; dwade; Maria Mulcahy; csik; jrochelle; rasmussen;
> > Cohen,

> > Jared A; Hall, Suzanne K; Dowd, Katie W; Erik Hersman; Brian Herbert;jst...@rhok.org; Jacob Colker; d...@beextra.org; d...@google.com; Mike

> > 2010/1/16 Dominic König <domi...@nursix.org> <domi...@nursix.org>:

[Tim Schwartz]

What happen to the 6000 entries?

[Christina Xu]

Apologies, everyone, my mistake! The initial batch IS of 6000
reports...due to something weird in the interface that is probably not
worth worrying about right at the moment, it only displays 300 of them
at a time. However, we've verified that there are other reports in the
system because the number of "to do"s sits at 300 even though
volunteers are completing them.

Christina

[Kevin Price]

Just a quick FYI, been working today on bringing this data in/out of WebEOC instances in use specific to Haiti. Though I am a .NET developer, if there is any additional assistance I can bring, please let me know.

Thanks,
Kevin

[Chen Li]

Two quick updates from the UCI team.

- We put at http://fr.ics.uci.edu/#data some of our daily crawled and
scraped data sets from several web sites. Some of them have been sent
to "Tim Schwartz" <tima...@gmail.com> and <brian...@berkeley.edu>
to be put into the "Person Finder" data server.

- We are currently working on search-related issues. As a
demonstration, we have built a type-ahead, error-tolerant search
interface at http://fr.ics.uci.edu/haiti/ on some of the data. We are
exploring the possibility of developing a similar interface for the
"Person Finder" data. Meanwhile, we are also building a similar
interface for the Red Cross web site
(http://www.familylinks.icrc.org/haiti/people) in case they find it
useful.

Chen

[Pablo Mayrgundter]

Hey Phil,

We changed the branding.  Just have a "powered by Google" at the bottom now.

The original branding wasn't well thought out.. I just grabbed the logo from the landing page linked from google.com and used it (http://www.google.com/relief/haitiearthquake/ at the top), and we didn't have enough time to do a proper UI review for a few days.

Hope that's looking better now.  Should fit more naturally into others' sites now.

Thanks,
Pablo

[Philip Ashlock]

Nice, I think that's a good move.

You guys are absolutely amazing. This isn't just a killer code-sprint you guys are doing, it's an incredibly impressive collaboration sprint.

Thank you!
Phil

[Tim Schwartz]

Anyone have a contact at CNN. They changed the structure of all of the
images uploaded to iReporter. I need these images to work, so we can
put them into google, i have 7000 ready. Can someone get this to where
it needs to go?

I need image urls like these to work:

http://www.ireport.com/Untitledjpg-1189076_sm.jpg
http://www.ireport.com/SDC11710jpg-1192675_sm.jpg

We can't launch the new volunteer converter tool till this is working.

-tim

[Andy Carvin]

Ccing Lila King from iReport.

Andy

Sent via smoke signals

>>> eff Martin; Bobby Fishkin; Praneeth; Dominic König; premr@google
>>> .com

[Prem Ramaswami]

Tim-

I added you to a cnn thread yesterday.  (will re-add)

[Mark Lugo]

Yes most cell phones are working and there is high speed internet WiFi
around the airport.

On Jan 18, 2:47 pm, TEL 4RENT <tel4r...@gmail.com> wrote:
> Tim,
>
> Thanks for the update.
>
> Do you know if people on the ground have access to internet on their phone?
>
> regards,
> tel4rent
>

> > On Sun, Jan 17, 2010 at 12:49 AM, Tim Schwartz <timat...@gmail.com> wrote:
> > > +brianzimmer
>
> > > Chen,
>
> > > If you can concentrate on the CNN iReports that would be great. I
> > > ended up using the ICRC scraper I had so I could make it feed directly
> > > into google in realtime, it will continue to update google as things
> > > to into it. Brian Zimmer has created a system for volunteers to go
> > > through cnn ireport data parsed by chen's group and put the data into
> > > google by hand. This system will be up sometime in the morning PST.
> > > Chen, please pass him your updated scrapes of CNN, and remove all
> > > extraneous columns.
>
> > > 16,000 ICRC entries should be going into google within 30 min, that
> > > will confirm that the API is working and it will be opened tomorrow.
>
> > > good night and good luck,
> > > tim
>
> > > On Sat, Jan 16, 2010 at 11:56 PM, Chen Li <che...@ics.uci.edu> wrote:

> > >> We put three dump data files athttp://fr.ics.uci.edu/. Tim is using

> > >> the files to do the conversion for the Google repository.  We will
> > >> continue our crawling and scraping efforts.
>
> > >> Chen Li
>
> > >> On Sat, Jan 16, 2010 at 6:27 PM, Chen Li <che...@gmail.com> wrote:
> > >>> I want to confirm that our UC Irvine team of 7 people will deliver the
> > >>> data set Tim mentioned as different mysql dumps according to the web
> > >>> sites.
>
> > >>> Chen Li
> > >>> Associate Professor
> > >>> UC Irvine

> > >>>http://www.ics.uci.edu/~chenli/<http://www.ics.uci.edu/%7Echenli/>

>
> > >>> On Sat, Jan 16, 2010 at 6:12 PM, Andy Carvin <ACar...@npr.org> wrote:
>
> > >>>> Don't forget that we now have a google group for this conversation;
> > please
> > >>>> be sure to cc it in all correspondences so the broader group of people
> > >>>> working on the project has access to this information.
>
> > >>>> person...@googlegroups.com
>
> > >>>> (though we may want to quickly change the name to peoplefinder so it
> > >>>> corresponds with the name lots of peoplea are using.)
>
> > >>>> thanks,
> > >>>> ac
>
> > >>>> -----Original Message-----
> > >>>> From: Tim Schwartz [mailto:timat...@gmail.com]
> > >>>> Sent: Sat 1/16/2010 8:53 PM
> > >>>> To: che...@gmail.com
> > >>>> Cc: Mark Prutsalis; Pablo Mayrgundter; Ben Rigby; Luke Beckman;
> > Stanton,
> > >>>> Katie; Andy Carvin; Tomas Krag; sundeep ahuja; Josh Nesbit; David
> > Kobia;
> > >>>> Patrick Meier; dwade; Maria Mulcahy; csik; jrochelle; rasmussen;
> > Cohen,
> > >>>> Jared A; Hall, Suzanne K; Dowd, Katie W; Erik Hersman; Brian Herbert;

> > >>>> jst...@rhok.org; Jacob Colker; d...@beextra.org; d...@google.com; Mike

> > Yang;
> > >>>> eff Martin; Bobby Fishkin; Praneeth; Dominic König; pr...@google.com
> > >>>> Subject: Re: PeopleFinder Collaboration
>
> > >>>> Major Update From The Front Lines:
>
> > >>>> I just got off the phone with Chen Li from UC Irvine. His group has
> > >>>> been working all day at scraping and parsing content from:
>
> > >>>> CNN iReports, ICRC, and other unstructured sources, they are currently
> > >>>> giving me a data dump of two things
>
> > >>>> 1) totally structured data from ICRC that can go directly into google
> > >>>> tonight
> > >>>> 2) the unstructured data from all over the web
> > >>>>     - this data will come in a form of Title, Text Blurb, and Image
>
> > >>>> I am about to start building an interface for volunteers to comb
> > >>>> through the unstructured data and input it into google. This will be
> > >>>> built on the code repository that ran haitianquake.com and is now
> > >>>> running the SMS volunteer interface by Brian Herbert. If anyone out
> > >>>> there is willing to work I could use help. I will be working for the
> > >>>> next 4 hours, so if anyone out there is in another continent and take
> > >>>> over in 4 hours that would be incredibly helpful. It would be great to
> > >>>> have this interface done by tomorrow and have volunteers ready to go
> > >>>> late morning.
>
> > >>>> Keep it up!
>
> > >>>> -tim
>

> > >>>> 2010/1/16 Dominic König <domi...@nursix.org>:

[Prem Ramaswami]

From our latest report, 1:20 calls are going through right now ...

[TEL 4RENT]

Thanks for the update!

The reason being is they could access www.personfinder.haitiearthquake.tel to check the missing person feed. We gathered as much information as possible (phone, web sites, charity agencies etc....) under www.haitiearthquake.tel.

It offers the following functions and save time on searching.

phone number to dial,
sms number to dia,
email address to start typing

If there is anything else they need on the ground, please let us know.

God bless,
tel4rent
haitiearthquake.tel team

[Ajay]

The URLs hosted on http://i.cdn.turner.com are alive. Of course I have
to do html scaping to get them.

ex: http://i.cdn.turner.com/ireport/sm/prod/2010/01/16/WE00385762/1198841/...

-Ajay

On Jan 19, 2:02 pm, Tim Schwartz <timat...@gmail.com> wrote:
> Anyone have a contact at CNN. They changed the structure of all of the
> images uploaded to iReporter. I need these images to work, so we can
> put them into google, i have 7000 ready. Can someone get this to where
> it needs to go?
>
> I need image urls like these to work:
>

> http://www.ireport.com/Untitledjpg-1189076_sm.jpghttp://www.ireport.com/SDC11710jpg-1192675_sm.jpg

>
> We can't launch the new volunteer converter tool till this is working.
>
> -tim
>

> On Tue, Jan 19, 2010 at 12:12 AM, Philip Ashlock <p...@ashlock.us> wrote:
> > Nice, I think that's a good move.
>
> > You guys are absolutely amazing. This isn't just a killer code-sprint you
> > guys are doing, it's an incredibly impressive collaboration sprint.
>
> > Thank you!
> > Phil
>
> > Pablo Mayrgundter wrote:
>
> > Hey Phil,
> > We changed the branding.  Just have a "powered by Google" at the bottom now.
> > The original branding wasn't well thought out.. I just grabbed the logo from
> > the landing page linked from google.com and used it

> > (http://www.google.com/relief/haitiearthquake/at the top), and we didn't

> > have enough time to do a proper UI review for a few days.
> > Hope that's looking better now.  Should fit more naturally into others'
> > sites now.
> > Thanks,
> > Pablo

> > On Mon, Jan 18, 2010 at 4:58 PM, Philip Ashlock <p...@ashlock.us> wrote:
>
> >> Google: YOU NEED TO GET DATA ABOUT HOW MANY PEOPLE ARE IN THE SYSTEM
> >> AND A LIVE FEED OF WHEN MISSING PEOPLE PAGES GET UPDATES!!!
>
> >> Additionally maybe we can slightly rework the Google branding on the
> >> Person Finder app, Maybe it can be called Haiti Crisis Response instead of
> >> Google Crisis Response. If needed, then maybe we could add the Google logo
> >> to the bottom of the page along with the others who are actively working
> >> with it or pointing towards it like The White House, U.S. State Department,
> >> The New York Times, and hopefully CNN, Facebook et all. Sometimes Google
> >> accidentally mixes its brand in with open platforms (which is why the "G" in
> >> GTFS was just changed from Google to General.)
>
> >> I've been wondering if the Google branding is making it difficult to fully
> >> bring on CNN and Facebook to point to or integrate with Person Finder.
>
> >> CNN is definitely still pushing iReports as the tool for missing persons
> >> and reporting issues on the ground. That'd be fine if it plugged into the
> >> APIs and then it could just be like how Reuters worked to boost its image by

> >> association with 4636 -http://www.reuters.com/article/idUSTRE60G16420100117

>
> >> Things work so much better when people don't care about getting credit for
> >> their work.
>
> >> Phil
>
> >> Tim Schwartz wrote:
>
> >> I would assume not. SMS is the best your going to get for a while.
>
> >> -tim
>

> >> On Mon, Jan 18, 2010 at 12:47 PM, TEL 4RENT <tel4r...@gmail.com> wrote:
>
> >> Tim,
>
> >> Thanks for the update.
>
> >> Do you know if people on the ground have access to internet on their
> >> phone?
>
> >> regards,
> >> tel4rent
>

> >> On Sun, Jan 17, 2010 at 12:49 AM, Tim Schwartz <timat...@gmail.com> wrote:
>
> >> +brianzimmer
>
> >> Chen,
>
> >> If you can concentrate on the CNN iReports that would be great. I
> >> ended up using the ICRC scraper I had so I could make it feed directly
> >> into google in realtime, it will continue to update google as things
> >> to into it. Brian Zimmer has created a system for volunteers to go
> >> through cnn ireport data parsed by chen's group and put the data into
> >> google by hand. This system will be up sometime in the morning PST.
> >> Chen, please pass him your updated scrapes of CNN, and remove all
> >> extraneous columns.
>
> >> 16,000 ICRC entries should be going into google within 30 min, that
> >> will confirm that the API is working and it will be opened tomorrow.
>
> >> good night and good luck,
> >> tim
>
> >> On Sat, Jan 16, 2010 at 11:56 PM, Chen Li <che...@ics.uci.edu> wrote:
>

> >> We put three dump data files athttp://fr.ics.uci.edu/. Tim is using

> >> the files to do the conversion for the Google repository.  We will
> >> continue our crawling and scraping efforts.
>
> >> Chen Li
>
> >> On Sat, Jan 16, 2010 at 6:27 PM, Chen Li <che...@gmail.com> wrote:
>
> >> I want to confirm that our UC Irvine team of 7 people will deliver the
> >> data set Tim mentioned as different mysql dumps according to the web
> >> sites.
>
> >> Chen Li
> >> Associate Professor
> >> UC Irvine
> >>http://www.ics.uci.edu/~chenli/
>

> >> On Sat, Jan 16, 2010 at 6:12 PM, Andy Carvin <ACar...@npr.org> wrote:
>
> >> Don't forget that we now have a google group for this conversation;
> >> please
> >> be sure to cc it in all correspondences so the broader group of
> >> people
> >> working on the project has access to this information.
>
> >> person...@googlegroups.com
>
> >> (though we may want to quickly change the name to peoplefinder so it
> >> corresponds with the name lots of peoplea are using.)
>
> >> thanks,
> >> ac
>
> >> -----Original Message-----
> >> From: Tim Schwartz [mailto:timat...@gmail.com]
> >> Sent: Sat 1/16/2010 8:53 PM
> >> To: che...@gmail.com
> >> Cc: Mark Prutsalis; Pablo Mayrgundter; Ben Rigby; Luke Beckman;
> >> Stanton,
> >> Katie; Andy Carvin; Tomas Krag; sundeep ahuja; Josh Nesbit; David
> >> Kobia;
> >> Patrick Meier; dwade; Maria Mulcahy; csik; jrochelle; rasmussen;
> >> Cohen,
> >> Jared A; Hall, Suzanne K; Dowd, Katie W; Erik Hersman; Brian Herbert;

> >> jst...@rhok.org; Jacob Colker; d...@beextra.org; d...@google.com; Mike

> >> 2010/1/16 Dominic König <domi...@nursix.org>:

>
> >> Oops,
>
> >> forgot to mention that though the document defines a namespace for
> >> pfif,
> >> it
> >> finally doesn't use it - meaning: all elements of that document
> >> except the
> >> <pfif:pfif> element would _not_ be PFIF in that layout.
>
> >> It could be easier to specify the namespace without any prefix
>

> ...
>
> read more »

[Tim Schwartz]

ajay, dont worry about CNN any more, they are taking care of their
data and importing it directly into google. if you can continue to
work on CBC that would be great, it is currently being converted into
google here:

http:/www.haitianquake.com/converter/

-tim