Subscribe to this thread
Home - General / All posts - Great video from Jack Dangermond
Dimitri


7,413 post(s)
#17-Apr-20 05:46

Jack Dangermond, the founder and owner of ESRI, has just released a great video thanking GIS people for the work they are doing in the Covid-19 crises and setting forth how GIS is a key tool in the global response to the pandemic.

Jack, of course, makes an ESRI-centric presentation, but that's OK and still very inspiring, as what he discusses also applies to other GIS communities, such as ours, QGIS, MapInfo, and all the others. GIS people worldwide are making a great contribution, whatever tools they use.

Every one of us can contribute our skills and our knowledge of GIS to helping the people and the communities around us cope with the crisis, even if it is nothing more than clicking open a pandemic example map to keep an eye on real data, so we can discuss Covid-19 with family and friends using an evidence-based approach to the pandemic. There are so many wild rumors flying about, and the way to help people deal rationally and effectively with fear is through solid, accurate information. GIS is a great way to understand and present the real information that is there.

I differ with Jack on a few points. First, he implies the ESRI toolset is now free. It isn't, although in praise of ESRI they have greatly loosened up their licensing. But for all that, the ESRI toolset behind the slides in that video remains far too costly and beyond the reach of most of the world's population that needs GIS in a time of crisis.

But you can say that about Manifold as well. For all the better price/performance Manifold offers, the price of Release 9, what you have to pay to get 9's essential ability to handle the data sizes required to deal with Covid-19 in the detail a really strong epidemiological response requires, remains too high for most of the world. I very strongly believe Manifold has to change that. Now is the time for all of us to do all possible, not to hold anything back. Manifold has resources and reserves. Now is the time to commit them, to go all in.

I also respectfully differ with Jack on the dashboards. While 100% agreeing dashboards are a great idea for presenting data, I respectfully note the Johns Hopkins dashboard which has become ubiquitous ends up underlining the lack of data-handling backbone in the GIS technology powering it. In too many cases it is too slow and jerky, showing the results of building an end result based on too many layers of technology where each layer has inadequate performance to handle the data volumes in play.

That's a situation where 9 can help, sliding Radian technology into those layers to improve performance. For now, that's a technical challenge to do with 9, but the magnificent ideas Jack shows in his video set out a guide for how to make that easier, to help guide the web services technology Manifold is making ready. As is well known, Manifold is developing powerful web interfaces that will be built in, native, to 9. Those are the key to point and click Enterprise integration for presentations and sharing.

While that work goes on, the GIS community and we in Manifold already have very rich toolsets to help cope with the pandemic, using a mix of Manifold, ESRI, Q, MapInfo, FME, the various web servers, many varieties of enterprise DBMS, both FOSS and commercial, and a wide array of language tools and libraries. On the front lines of GIS work fighting the pandemic I think most of the work is routine, unglamorous but necessary slicing and dicing of data to whip it into shape for collaborative use. We certainly have the tools for that.

But most important of all, we have the intense skills and knowledge and genius of a world-wide community of very smart and very dedicated GIS people. Together, we can all make a difference in a time of great need. Thank you, Jack, for reminding us of that.

artlembo


3,400 post(s)
#19-Apr-20 02:57

I have been particularly impressed with the use of GIS in fighting this pandemic. what impresses me most is that nobody is using the term GIS. They are just providing answers to questions.

Case in point, Dr. Birx was showing coronavirus by state – no big deal. But then she pivoted and began to show coronavirus by metropolitan area. I was doing that a couple of weeks ago when trying to create a regression model to see if mass transit ridership has a correlation with CV. I had to use GIS to overlay the counties with the metropolitan areas. I’m guessing that’s what a lot of her analysts did as well.

One of the military men at the briefing mentioned that he asked his analysts to determine the number of people within 10 miles of all the testing centers. He said something like 90% of all Americans live within 10 miles of a testing center. Again, another GIS application, and yet the term GIS was not used.

To me that is the most encouraging thing. We are no longer trying to create a market for ourselves and prove that GIS has value. It just does. Nobody’s running around trying to tell the importance of spreadsheets. We just use them. I think GIS has now crossed that divide. It’s just part of the solution set to get people answers, and this virus has demonstrated that.

Because my university is so close to the DC area, many of my former students are working on this problem. In fact, one of my recent graduates is now working 12 hour days, seven days a week, and she tells me she reports directly to a four-star general! She is providing GIS answers to their questions.

My lab is starting to get involved in our State as well.

Here”s the next important step (something I’m starting to get involved in): as you say, dashboards are cool looking, but there are much deeper questions that need to be asked and they have spatial components to them. The problem is, many of our leaders are unaware that the questions can even be asked. That is what impressed me so much about the question the military leader asked about the number of people within 10 miles of the testing center - perfectly suited for GIS. this is where we come in. This is an opportunity to reach out and help people think about the kinds of questions they might want to ask. this is also what manifold GIS does very well – quickly allows you to bang out answers to questions (hopefully using SQL).

So, my challenge to all of us is to reach out to the people we know who are trying to provide answers, and maybe give them some ideas about the kinds of questions they could ask. And finally, offer to help them find those answers.

Stay safe, everyone.

wvayens108 post(s)
#19-Apr-20 16:27

Just a quick note to say thanks for posting the pandemic example map....examples like these are great for people like me who are slowly transitioning from M8 to M9....I see things in these and ask myself "now how are they doing that" and then dig into it and always learn a new thing or two. Thanks!

lionel

995 post(s)
#15-Jul-20 02:32

ESRI website service for share data

https://experience.arcgis.com/page/landing

RKI that use ERSI service for covid

https://experience.arcgis.com/experience/478220a4c454480e823b17327b2bf1d4/page/page_0/

ESRI + Microsoft ( Bing ..)

https://www.esri.com/library/brochures/pdfs/microsoft-plus-esri.pdf

Microsoft covid

https://www.microsoft.com/en-us/p/bing-covid-19-tracker/9mttj8zkc1wb?activetab=pivot:overviewtab

Attachments:
microsoft_covid.png
RKI_ESRI_covid.png


Book about Science , cosmological model , Interstellar travels

Boyle surface fr ,en

tjhb
10,094 post(s)
#15-Jul-20 05:37

Yes, the Robert Koch dashboard looks clearly wrong.

A day later, the totals for cases and deaths have increased (slightly--they still have not caught up with Microsoft's figures), but the daily change figures remain identical with your example above.

Possibly there is a bug in the ESRI toolset, or perhaps the toolset is being misused. (Who knows what day it is anymore, honestly?)

Dimitri


7,413 post(s)
#15-Jul-20 05:55

It makes a noticeable difference what source of data you use. Might be that.

For example, most US dashboards use the Johns Hopkins data, while some European dashboards, as well as the Manifold pandemic.map example, use ECDC (European Centre for Disease Prevention and Control) numbers. The two sources publish data aligned to different dates, and with the rate of growth of both infections and fatalities there are noticeable differences from one day to the next.

Manifold's example uses the ECDC data because it is published as a CSV server that is easy to harvest and to immediately use. The Johns Hopkins data is published in a form that takes more digestion.

Ah... another idea that may affect some dashboards. What figures you get when normalizing to population depends on the population data you use for each country.

I forget what Manifold's example uses, but that example is intended as a demo of how to do it and to produce a result that allows relative comparisons. In that case, if populations haven't changed all that much in the last few years for relative comparison purposes it doesn't matter for relative positions if you use population data from 5 years ago or 10 years ago.

But if somebody is trying to look at specific numbers (believing there is anywhere near the accuracy required in the data for that...), then it will matter what the population numbers are that are used for normalization, if normalized figures are reported.

tjhb
10,094 post(s)
#15-Jul-20 06:01

Sure, the two sources could refer to different dates.

But the ESRI feed seems to be referring to different dates within itself, since the daily change figures have not moved from the day before (not impossible but highly implausible).

That would indicate a problem.

Dimitri


7,413 post(s)
#15-Jul-20 06:14

Ah, I missed that. Yes, that looks like a problem. Hard to say what causes it.

It's easy to forget that some tools which get handed around or copied need to be re-configured or updated. Manifold's own pandemic.map is an example of that in the way it uses Style.

Thematic formatting in Manifold right now is calculated based on the data as it was when the style was applied, because the style is a fixed, specific JSON string. So what were reasonable intervals when worldwide cases were few becomes totally out of date when the numbers have increased tenfold or a hundredfold (everything is pushed into the last interval).

The temporary solution is to reapply the thematic format from time to time, to allow intervals to be recalculated to match data. A real solution is to ask Manifold to provide a thematic formatting method that is a relative method that updates on the fly: "give me ten equal intervals based on this field using these colors."

The key is to have reasonable refresh logic, since recalculating intervals on big data from remote databases (imagine a billion records sitting in some distant server connected over a slow link...) is not something you want to do on every update to every record. So maybe a recalculation every time the project is first opened, or whenever a manual refresh is commanded. [In the future it would be nice to have options for such triggers, such as perhaps every day or every week...]

I wrote that up and sent it in as a suggestion. That would make it possible to create a pandemic map that used the same relative color intervals but which automatically updated.

lionel

995 post(s)
#08-Sep-20 22:19

In france during very hot summer ( 14 jours en aout 2003 see TV); ined study ( 4 years after november 2007) the number of death using parameter like french departements and age .

Is there already same study for covid using same parameters ( location age) ? in winter flu also occur and I need to know is there a signal a différence that could be already see detect ?

Study data give more information than media social network !!

Regard's


Book about Science , cosmological model , Interstellar travels

Boyle surface fr ,en

rk
621 post(s)
#09-Sep-20 05:25

I'd star at https://www.euromomo.eu/graphs-and-maps

The agency that provides french data may have more detailed info about France.

lionel

995 post(s)
#01-Nov-20 19:52

infography and gis are usefull for have a photo of covid (death, ill,+) even many ( not all) essay for find a treatment fail for prove efficiency of treatment.

-----------------------------

Here a French Web service that appear related to COVID for know wich test ( RT PCR , antign, blood) is avaliable in france with different view ( in a software way).

the table view: https://covid-19.sante.gouv.fr/tests

the camenbert view : https://covid-19.sante.gouv.fr/tests/dataviz

-------------------------------

Here also a gis map from Sané Publique France that use Geoclip Technology

https://geodes.santepubliquefrance.fr/#c=indicator&f=0&i=covid_hospit.rea&s=2020-10-31&t=a01&view=map2

------------------------------

there is new restriction in france for limit the covid spread. The new Manifold UI disturb me since idon't use it a lot since 4 months ( select column use for start and for end =result ) .

Howto create a 1 km circle around each home location ? ( to export kml )

Is there a way to copy/strore the projection of a Google map Layer( create using : create -> new data source)

When create a Drawing layer , i have not the choice to choose for projection of the google map layer available in map file ! in old versino of manifold i could select a drawing and manifold take the projection of this drawing here i have in more 3 tab : standard, EPSG, custom that don't help the newby i am !!

Does the projection use by google, osm , bing should be available by default in manifold ?

Strange we can't save the the current location when google map is open inside manifold !!

in some transform tool tere was sometimes column or label about unit not for compose circle here !

Attachments:
manifold_covid-home-distance-limit1km.png


Book about Science , cosmological model , Interstellar travels

Boyle surface fr ,en

lionel

995 post(s)
#01-Nov-20 20:25

B) Here some link that compute circle area online ( even android Locus Map is better with kml)

https://www.calcmaps.com/map-radius/

https://1kilometre.fr/

https://www.madeinsante.fr/1km/

https://covidradius.info/

B) the best ll be to have geofencing tool on android that accept kml fie to alert us when we cross the external Circle limit !! ( it is like a pet tracker or GSM wrist GPS that prevent people to go outside or inside an area.)

If people know such software for android that work locally ( not as a service) and can alert us when we cross the limit .. I am interesting


Book about Science , cosmological model , Interstellar travels

Boyle surface fr ,en

Dimitri


7,413 post(s)
#02-Nov-20 07:27

Howto create a 1 km circle around each home location ? ( to export kml )

Use buffers.

Is there a way to copy/strore the projection of a Google map Layer( create using : create -> new data source)

The Info pane is the gateway to coordinate system dialogs. The Properties dialog of the google layer's table also shows the projection, which means it is in the mfd_meta table as well.

i have not the choice to choose for projection of the google map layer available in map file !

Yes, you do: the Pseudo-Mercator projection used by Google is both the default, and also one of the two default Favorites.

Does the projection use by google, osm , bing should be available by default in manifold ?

It is available, and it is even the default. See the prior answer above.

Strange we can't save the the current location when google map is open inside manifold !!

Yes you can. You just can't save it into Google's servers. The Location button tries to save the location into the storage context of the active window. That is very convenient when working with linked projects and data sources, for example, nested projects. But if that window is coming from a server, you must (of course) have write access to save to that server.

If you want to save a location to the map file, create a map in the project containing the google layer and open the map and work with the google layer in the map. Then the context will be in the map and locations can be created there.

See also the discussion in the Project pane topic.

in some transform tool tere was sometimes column or label about unit not for compose circle here !

Correct. Read the topic for the tool being used. There you will find useful information like...

The Radiusbox takes whatever are the native units used by the coordinate system used by the geometry field we are using. Since our drawing uses Pseudo-Mercator projection, which uses meters, the unit of measure is meters.

I note that these questions have nothing at all to do with a thread about Jack Dangermond's video. Start a new thread if you want to raise new questions / discussion.

lionel

995 post(s)
#02-Nov-20 09:59

thank's i ll try again

thread continue here


Book about Science , cosmological model , Interstellar travels

Boyle surface fr ,en

Manifold User Community Use Agreement Copyright (C) 2007-2021 Manifold Software Limited. All rights reserved.