No, it is not possible for the next ten years. Reason for that because ESRI corporation is one of the oldest in the world, and it covers more of 65% of markets. But QQIS is awesome and inserting software with a lot of possibilities. The main difference between QGIS and ArcGIS in architecture. QGIS is built by the support of Python code and a little bit in C++. Arc GIS is under C++ code. This question is very interesting and I suppose that it deserves one conference for example.

Thank you @ Aleksandar Dordje Valjarevic. My observation is that QGIS seems to be a trajectory of regularly updating their software. Probably in the next 10 years the narrative could be different

In my opinion, Qgis (actually, the whole OSgeo ecosystem) will overtake ArcMap (actually, the whole ESRI-related softwares) in less than 10 years.

There is only 1 last (huge) step to achieve for OSgeo community: getting into education.

ESRI has pioneered R&D in geospatial analysis, mapping, cartography and geodata management since the '70s and from back then to today, they are regarded as the leader in this sector.

Universities who teach geospatial, GIS or cartography courses teach ESRI products because of their easy to understand GUI, wide array of fully documented toolboxes and an amazing support. If you want to do geostatistical analysis, with ArcMap you can do it even if you are not an ace in geostatistics, because the Help panels will guide you thru and actually teach you geostatistics.

So, if most if not all universities around the world teach ArcMap since early days, what do you expect that most of the business will adopt for their GIS department...ESRI of course.

However, things are changing quite rapidly nowadays and OSgeo community driven R&D is getting boosted not only from user donations but also by fundings from private companies and the outcomes are amazing.

Qgis3 now is developing 3D capabilities (equivalent to ArcScene) and the adoption of Geopackages (.gpkg files) is extremely handy as it is only one file (SQlite-type) basically containing all the functionalities of the typical 3 files-structure of ArcMap.

Also, Qgis is free and this means that a lot of people around the world use it and post questions that eventually get answered in forums (like StackOverflow or Geographic Information Systems), and this forms the corpus of the support.

Another super cool point for Qgis is that the processing tools (even if coded in C++, for efficiency) are exploitable with the Python based API.

This means that (almost) all the tools there are scripts written in Python, an easy to learn scripting/programming language that it is extremely important in automation of geospatial analysis.

Also, Qgis implements very well Postgres databases with Postgis extensions...the ability to link a spatial database and run spatial functions and SQL queries in this age of Big Data analysis, then display the results directly into Qgis as a dynamic views....well, this is magic. And it is simple. Extremely powerful, and simple.

Not mentioning the easy link with Geoserver and webmapping applications..oh yeah, the plugins!

Qgis it is plugin based, meaning that community of python (or C++) developers can transform their scripts in plugins for Qgis and then publish it in the public repository, accessible thru Qgis. Basically, it means that the world will help you, not only ESRI developers.

Getting back to education, in my opinion Qgis is way better in learning GIS analysis for these points:

• it forces you to learn how to find help online
• it is free, meaning that you can install it for free and play with it home forever, without student licence, meaning lifetime learning by doing
• it forces you to learn how to update your software and cope with version-related issues
• it makes you think about learning Python to expand your freedom in analysis or automate processes
• It makes you understand how a GIS GUI (like ArcMap) works behind the curtains
• eventually, you might also switch to Jupyter notebooks and python with all the magical Geopandas, Rasterio and co. packages for your data crunching and Qgis as a cartographic tool.

CONLCUSION:

Thus, I believe that soon the education sector (if not politically biased) will adapt to Qgis, and as soon as this will happen, OSgeo will spread and become the new standard, just like Google became the new standard even if Yahoo was there first web search engine.

Note about me:

I started my GIS career with ESRI, then I moved to Qgis lately but I am using Jupyter Notebooks and Python for geodata analysis.

All I written is my personal view, related to my own experience as a learner and geospatial analyst.

Cheers!

Nick

Thank you @ Nicolas Pucino for that comprehensive response. What about QGIS tutorials for OBIA, where can I get it?

Wilfred Ochieng , it has been a pleasure.

For OBIA in Qgis, I know a plugin called Semi Automatic Classification (SMC) or something like that. I am not sure if deals with segmentation and object based classification, but that is what comes to my mind right now. I am sure if you google it you will find some helpful hints fromthe web!

Cheers,

Nick

Dear Wilfred

I am a person who has been working with ESRI software since 1990 and free & open GIS software since 2005. For that reason I want to clarify that the following opinions where only taken only from my experience of working with those softwares every day.

I think QGIS have already overtook Arc-GIS Map. There are severeral reason for that:

ArcGIS consume too much resources (memory, CPU, etc) without necessity.

ArcGIS has a lot of code errors and it will be noted when automation process has to be implemented (Could be broken in a middle of a process).

ArcGIS doesn't respect properly international standards (like OGC) or standards are difficul to implement.

Arc Python in ArcGIS Map is wrongly implemented. ArcPy doesn't looks as Python and doesn't implement correctly object oriented premises. This fact affect the lenght of python code. For example on my experience ten lines of ArcPy can be implemented in one line of PyQGIS.

Professional ArcPy is much more pythonic and OO but still not convice me because to many important APIs are out of the main package and new extensions has to be contracted ($$$$).

You can have a new version of ArcGIS nearly earch two years. You have a new QGIS version each four months. QGIS code is continuosly chequed and corrected.

ArcGIS do not have a good database integration. With QGIS you can access SQL natively and new powerfull formats like SpatiaLite and Geopackage.

In my personal opinion, when you work with ArcGIS you learn about tricks that help you to avoid software problems. With QGIS you learn GIS.

The reason of those problems of ESRI software I think are:

A private company cannot share easily a code in order to be checked by many persons. Free & Open software is checked constantly by thousands of persons.

At present GIS has too many components that cannot be managed only by one company like: standards, database systems, modelling systems, geoprocessing algorithms, programming APIs, documentation, graphics, digital printing, etc. etc. etc. The Free & Open Software reuse multiple componets generated freely by multiple independient projects and this kind of innovation ecosystem promote the development of complex software in a controlated environment.

At last if you do not like QGIS you can move to others like Grass, GVSIG, 52 North, Spring, etc. etc. without problems because those softwares use same data standards.

I hope these comments will help you. Cheers Gabriel

Thanks @Gabriel Asato. I can now confidently introduce QGIS to my students

Gabriel Asato says " In my personal opinion, when you work with ArcGIS you learn about tricks that help you to avoid software problems. With QGIS you learn GIS. "... I totally agree with him and I think that this is why QGIS should be taught instead of ArcMap.

Cheers

Nick