a thoughtful web.
Good ideas and conversation. No ads, no tracking.   Login or Take a Tour!
comment
user-inactivated  ·  631 days ago  ·  link  ·    ·  parent  ·  post: Pubski: July 6, 2022

veen

Please excuse the delay. I've been in the middle of moving in with the S/O on top of a family visit that has thrown off my schedule without my full workstation... You've hit a loooooooooot of topics swirling on my mind for a few years now.

    As a whole the GIS world is...surprisingly shallow. There are some technical niches for sure, but compared to what I've seen in other domains of engineering, one can get incredibly fast to a point where one can do 80% of all GIS work. Really, a basic GIS course combined with a modicum of data-wrangling chops and Google skills can get you very far. To speak from personal experience; I had 2 mandatory GIS courses at uni, took one PythonGIS elective

Looking back on my certificate and MS in GIS. I've realized this profession could easily be taught in a trade school. A couple courses here (one on vector data and it's applications; followed by a course on raster), another there (types of data collection - in situ, remote sensing and its varieties). Maybe, just maybe some visualization via web dev (AGOL), storage/'big data' (SQL/PostGIS) or automation (ArcPy).

That right there could get a solid 2 years' worth of 'full-time' schooling and be done with it. This isn't to say I'm happy to have a couple extra letters after my name, but - call it imposter syndrome if you want - I don't think the value of what I learned in grad. school significantly impacted the type of job or salary I got coming out. Even looking at the market now, there is very little in between from technician to developer work. From what I've found, most job descriptions don't have that much variation between Tech - Analyst - Specialist. Unless, of course, you're starting out as an imagery analyst. Which is a job that has its day's numbered.

    and learned enough on the job the past 4 years (all of ArcGIS Online PostGIS + ArcPy) that I can prolly apply for most senior GIS jobs out there. A lot of GIS work is just about getting the right input into the right GIS tool(s) and ✨presenting✨the result. I know people who have done nothing more than "load data into GIS, apply pre-made tools, visualize" for decades.

    A shockingly small niche (over here at least) is the people who are good at writing queries and half-decent at GIS. PostGIS legitimately can replace 95% of the individual pre-made tools QGIS and ArcGIS has to offer. You can do much more complex things much faster. My largest project the past year ended up being 2300 lines of PostGIS/SQL code I wrote on my own. The first 30% is just data prep written in code - "make sure I properly join tables A thru G in the data type I want it to be without ever having to touch Field Mappings ever again". The rest is a bunch of clever geo-joins and a bunch of not clever regular joins of tables and features. Nothing special to anyone who already knows how to handle semi-long SQL queries; PostGIS is really just one new column type and a bunch of functions to do stuff with it.

The tipping point at my last workplace was a senior GIS dev that replicated my 2 months of census analysis (via ArcGIS) under 2 weeks with PostGIS. Now, I adore this dev, they were the only mentor in that floor. It made me realize how much more there was out there in GIS (and how little I could contribute to the field with my current skillset).

Most salient of the points made while watching said dev work was "You've done a great job grinding at this for the past two months, though QC'ing is not possible since ESRI writes fresh data every. step. of. the. way. Now scale that over the course of 2 months of troubleshooting, trial & error, etc."

It's a no-brainer.

Would you mind sharing the course you took by chance? My resources of yet are a couple textbooks by recommendation of the former GIS Dev co-worker, a 4 hour YT vid on PostgreSQL, and a coursera course on SQL.