Hi all. I am asking this open-ended question (bottom of this post) because I am considering making contributions to an open-source project that would directly benefit me and other users.

Some background:

I have worked with an engineering simulation software called Ansys MAPDL basically everyday for the last 4 years, in both an academic and a professional capacity. It's not necessarily relevant whether you are familiar to that program to participate in this discussion. The relevant thing is that the GUI for MAPDL is written in Tcl/Tk and I don’t imagine it is going to be modernized (because of more modern, but distinctly different, replacements). This is a screenshot of the GUI for reference.

Why do people put up with such an old interface?

The power of the program is not its GUI, but the scripting language that can be run to setup and solve simulations. The program name is really the scripting language name, Ansys Parametric Design Language (APDL). It's somewhat like Matlab. The program also offers an enormous amount of control when compared to the more modern GUI that's been released, since the modern GUI holds a totally different philosophy.

The older GUI is really helpful in certain circumstances because it will spit out a file containing commands that were used in the session. This is a great demonstration of how to run a command or use a setting/config command, but a lot of newer features are buried in the documentation and aren't available in the older GUI.

My coding experience

I know the MAPDL language very intimately, but my experience beyond it is limited to some Perl scripting, and a bit of Python exposure.

Motivation

Open-Source Ansys API

Recently, Ansys started supporting an open-source Python project called PyAnsys. MAPDL is otherwise fully closed source, and this is really the only public-facing API. PyAnsys has basically converted a lot of MAPDL script commands to a pythonic format, hence Python can now be used to interact with MAPDL. This is great for several reasons, but is limited regarding interactivity. Interacting with MAPDL via Python is basically happening in a fancy console via Jupyter notebook or IDE like Spyder. Certain commands will bring up Python-based graphics displays of solid models and results plots, but there isn't a dedicated GUI open all the time.

The Question(s)

My question is whether it is feasible to write a frontend GUI to a bunch of python commands. If you were going to do it, how would you do it? What might you write it with? Would you even do it? Is this a stupid endeavor?

My fiance is an artist who was laid off because he worked for a major employer in the entertainment industry that required him to be in contact with people.

So he's been unemployed for a long time now, but he has been trying to find work, but there isn't really anything available that uses his particular skills. So we invested in a digital art workstation (basically just bought a monitor with a digitizer built in to use a pen with) so that he could work on becoming an independant artist. But he's so bad at marketing and he spends so much time worrying about fine details that it takes him over a week to finish a single piece, so he hasn't had any success.

So if you combine this with a broken tooth that he hasn't been able to get taken care of because of a lapse in his dental insurance, he's not been in a good place.

I just bought a Mac and was looking for mac-native graphics programs (I sometimes work on marketing, so I need to do photo editing from time to time. Also I used to do photography as a hobby and want to get back into it), when I came across Serif's Affinity Designer Workbook. And I thought to myself that getting my fiance a training course in graphic design would allow him to shift gears into a segment where there is more work. It'll also give him a bit of a kick in the pants to get him moving and feeling better about himself.

I'm sure everyone knows there are millions of online training courses available right now, so I'm hoping someone might have any recommendations. I have some money saved up, so I don't mind paying a little bit extra if it'll result in better results. I'd prefer if it were a class that didn't rely on Adobe Illustrator if possible, but I know that it's the 'standard' and he'll probably have better luck if he has experience in it.

There was a recent discussion about the Windows App Store where people said they don't like using it. I'm in the Apple world, and people here constantly complain about the Mac and iOS app stores. I grudgingly use Steam to download games that are only available there. Everyone seems to hate using app stores, but most agree that having them is better than having to find stuff on the web or in bricks and mortar stores.

I don't tend to "shop." When I decide I need a product, I do research. I try to find unbiased sources, though that's problematic in itself. But I don't go browsing for anything because it's largely pointless and tends to drive you towards what the company that's best at selling wants you to buy rather than what's best for your needs. So for the most part, my interaction with app stores is searching for a specific program and either finding it or not.

What do people think would make app stores better? Complaints I've heard include:

• Too hard to find a product you want when you don't know the specific name of a particular one (like you want a photo editor that can make a photo mosaic, but don't know the name of a specific photo mosaic app)
• You search for a specific product and the top hit is a paid placement for a competitor
• Stores are full of crapware with similar names and similar keywords
• Top apps are all games

How could app store makers improve the situation? What would make using an app store a joy for you?