Let me get this straight: are the end users the ones using SQL to create
these derived/blended data sets or is the actual creation done by the
support or data library team, who are SQL specialists?
Ruven Brooks
On 9/25/2020 10:47 AM, Basques, Bob (CI-StPaul) wrote:
All,
I almost jumped over this thread from the beginning because I wasn’t
understanding the original question very well, mostly based on my own
labels for these types of services/datasets. I reference things like
this as “derived” data. The actual source data doesn’t exist (unless
it gets cached for performance reasons) but rather it a data blending
via a SQL call.
I agree with other comments here too, related to this type of product
usually being the sort of thing that can’t be easily done by most GIS
apps out of the box. Also provides for a pipelining of processes of
sorts for sudo processing on the fly.
I’ve got a few different examples of this, some fairly simple, some
very complicated that are treated as datasets by the end users,
because the SQL is embed into a config, like a Mapserver Mapfile for
example. More and more of our datasets are being created in this
fashion vs historically sourcing a “real” dataset directly.
In general the end users are starting to think and expect this type of
analysis approach to the data, especially related to time indexing and
looking at data over time. Consequently, this is pushing me (us) to
start thinking about time indexing of data and how to store datasets
accordingly.
Bobb
*From: *postgis-users <[email protected]> on
behalf of Shaozhong SHI <[email protected]>
*Reply-To: *PostGIS Discussion <[email protected]>
*Date: *Friday, September 25, 2020 at 3:11 AM
*To: *PostGIS Discussion <[email protected]>
*Subject: *Re: [postgis-users] Promoting PostgreSQL and PostGIS to
wider business intelligence community
*Think Before You Click: *This email originated *outside *our
organization.
All very interesting and useful points!
I am also thinking about data blending and new data production on the
data service platform for supporting the wider Business Intelligence
community.
Surely, we need excellent examples to show them that we make data
ready for their consumption, and are here to help them.
Regards,
Shao
On Fri, 25 Sep 2020 at 08:37, Andreas Neumann <[email protected]
<mailto:[email protected]>> wrote:
Hi,
In our GIS team (small team of 10, local government, province
level) we use a lot of SQL in collaboration with Gradle/GRETL and
Jenkins for our automated data flows and statistics. It is amazing
how much analysis and data aggregation you can do with SQL only -
without having to touch QGIS or ArcGIS or any other so called
"business intelligence" tools (that are often quite expensive).
Every new employee that wants to join our team has to have SQL
knowledge - that's a prerequisite - or they wouldn't get the job.
Most of our employees are not programmers though.
I also teach PostgreSQL/Postgis training courses (2-3 days
usually) - a lot of the participants are not programmers but still
manage to do analysis with SQL. Typical course participants are
scientists, people working at engineering companies or government.
So - I do think there is a significant number of people who use
SQL, but aren't programmers.
Andreas
On 2020-09-25 00:09, [email protected]
<mailto:[email protected]> wrote:
I doubt whether PostGIS has any direct value whatsoever for
desktop application users. At a very minimum, using PostGIS
directly requires a knowledge of SQL. In fact, the more
knowledge of SQL a user has, the more powerful PostGIS will
be. SQL is usually taught in a database course which, in
many computer science curriculums, is taught in the second or
third year, not to end users in another occupation.
Business intelligence systems such as Power BI and Tableau can
connect directly to PostGIS data bases and provide end user
commands and operations for querying and modifying those
databases. GIS systems such as ArcGIS and QGIS provide
similar capabilities. End users can get nearly all of the
power of PostGIS without having to learn anything outside of
the business intelligence system or the GIS system.
PostGIS is probably best reserved for people who have a
programming background and whose jobs or avocations involve
doing things that are difficult or impossible to do in
existing business intelligence or GIS applications. There seem
to be more than enough people like that to keep the PostGIS
developers quite busy.
Ruven Brooks
On 9/24/2020 3:58 PM, Shaozhong SHI wrote:
Though we have got some good examples of serving data
to wider business intelligence community, we are still
interested in finding excellent, compelling examples for
showing the value of PostgreSQL/PostGIS as a data service
to desktop application users.
I just wonder whether there are excellent examples, for
general users to appreciate?
_______________________________________________
postgis-users mailing list
[email protected]
<mailto:[email protected]>
https://lists.osgeo.org/mailman/listinfo/postgis-users
<https://lists.osgeo.org/mailman/listinfo/postgis-users>
_______________________________________________
postgis-users mailing list
[email protected]
<mailto:[email protected]>
https://lists.osgeo.org/mailman/listinfo/postgis-users
<https://lists.osgeo.org/mailman/listinfo/postgis-users>
_______________________________________________
postgis-users mailing list
[email protected] <mailto:[email protected]>
https://lists.osgeo.org/mailman/listinfo/postgis-users
<https://lists.osgeo.org/mailman/listinfo/postgis-users>
_______________________________________________
postgis-users mailing list
[email protected]
https://lists.osgeo.org/mailman/listinfo/postgis-users
_______________________________________________
postgis-users mailing list
[email protected]
https://lists.osgeo.org/mailman/listinfo/postgis-users
