FME

FME saves our group a lot of time by automating repetitive processes and preprocessing data for different types of tasks. I also really like the Automations and Flow Apps in FME flow. These are 2 great ways to quickly and easily deploy FME workspaces and allow non-FME users to reap the benefits of FME.

The standard libraries for conversion are good performance wise, the graphical interface is intuitive. For people with a GIS background it is a very good tool.

The best features of FME are it's integration capability, ease of use, and geospatial capability. In my own role as a geospatial analyst, there are a few software's that can be used to interrogate and transform spatial datasets - FME has a strong set of tools that allow users to view and manipulate geospatial data alongside non geospatial data as well. This combination makes it a great tool when needing to work with both geospatial and non geospatial data. I can see a lot of opportunity for FME to be used outside of my roles usual tasks due to it being an 'everything tool'. My personal learnings on how to use FME have largely been through practical work on the job rather than training. The tool itself is easy to get into and use and has a great online community and documentation to help its users. It is a tool that can easily be self taught to a certain level being a low code - no code solution to building various solutions. Especially for FME Form which is the desktop version. The standout function to me is FME's ability to integrate - pulling data from one place and loading it to another, connecting to APIs, webhooks etc have been very useful functions. FME Flow makes automation and scheduling of these integrations a very powerful tool - enabling many live processes to run and streamlining workflows. If someone comes to me and says 'can FME do it?' I normally say yes (maybe with a few buts!)

1.) I would love to see an option for dynamic parameters where you can pull drop-down values from a table or database. This would be huge for me so I do not have to constantly open workbenches, reimport list values, and republish the workspace because a new value was added. 2.) It would be nice to have more transformer-like actions for automations (e.g. tester, attribute manager, capability for loops, etc.) 3.) More robust customer support. I know and understand why Safe decided to step back from consulting and only focus on the technical aspects of the software, but there are some cases where there is some overlap or a quick question regarding formatting or something similar. It is a long, arduous process to get a new consultant added to our list of approved vendors and get a PO set up for a very small hurdle where I really just need a nudge in the right direction. I believe a small amount of technical consultation would go a long way in some of these types of cases. The last 2 are really just nitpicking because 3 dislikes were requested... It is hard to come up with 3 dislikes :)

Not a lot of people are familiar with the tool, it's difficult to find people in the market with FME as a skillset, this causes FME to be a liability at times as there's only a few people in the company that are familiar with it. This opposed to writing an ETL pipeline in Java/C# or Python which is much more generic and easier to recruit for.

The FME workbench interface can sometimes not be clear as to why something is failing, or why a field you are filling out is deemed as invalid. Often to tackle error messages and unknowns it is necessary to look at documentation on their site. Which is fine but I imagine tooltips or reminders in the UI would be helpful to improve ease of use. Perhaps the upcoming AI Assist in the newer version of FME will fill this gap. Using FME at a larger organisational level I would say does require someone in the business or a support partner to have a senior level understanding of FME to assist in best practice, debugging and maintenance. Particularly when using FME Flow.