Any Interest in a Coding Camp Type Thread?

[Banderson (Brandon Anderson)]

@Evan_Purdy

But it hasn’t actually worked out that way in my experience, for both Widgets and code, the reason they have broken for me is the way the software works has changed, or the BO signature has changed, which broke the widgets and code equally.

[klincecum (Kevin Lincecum)]

@NateS

Isn’t that what epicor ‘wants’ us to use?

To be honest, I will use what I am comfortable with.

@NateS

Aren’t they also less prone to breaking during upgrades?

They say so, but I’ve never had a bpm or function break.
I have had the opposite.

[JoeEpicor (Joe Haw)]

Definite interest, I have a coding background but when it comes to Epicor there are so many business objects, assemblies, and etc. that I’m unfamiliar with. It’s difficult finding how to properly use their methods and while there are amazing solutions to lots of questions, they’re scattered in the many posts on this forum and can be hard to find sometimes. There’s some documentation Epicor provides, but when things don’t work the documentation isn’t always enough to solve the problem.

[Ernie (Ernie Lowell)]

I feel your pain daily.

It’s hard to remember that Epicor (and indeed all ERP systems) are somewhat generic packages that can work in almost any manufacturing environment, and that each of us is an individual business that does stuff that isn’t cut straight from the “generic manufacturing environment” cloth. Having the wealth of tools to make it dance to our own unique tune is a blessing… and also a curse.

[klincecum (Kevin Lincecum)]

@Ernie

blessing… and also a curse.

[klincecum (Kevin Lincecum)]

Here is the thread:

Epicor Code Camp - Bring your skills, or lack thereof

Welcome to Code Camp! Please bear with me while I reserve a few posts and get some content in here. Continuing from here: Any Interest in a Coding Camp Type Thread? There seems to be enough interest in this, so I am going to get it started. In this thread we will both lay out some pre-defined topics, as well as answer questions. I encourage you to still ask questions in the main forums if your question is very specific and needs more help. This thread will probably tend to be a bit more gene…

Certainly welcome to continue discussion here about how it works.

[Evan_Purdy]

Might be as simple as curating some posts but I know it was a big breakthrough for me when I learned how to do a trace and how to turn on field help to see a binding.

[Banderson (Brandon Anderson)]

Step one of the class, how to do a trace, and how to read it.

[klincecum (Kevin Lincecum)]

@Banderson

Step one

Too late, I chose one I could easily do in my time available!

[Banderson (Brandon Anderson)]

Just make the bullet point for the place holder.

[klincecum (Kevin Lincecum)]

@Banderson

bullet point for the place holder.

@klincecum

Topics: (Bear with me, I’m no professor… No particular order!)

• HowTo: Create a tracelog / Read a tracelog
• What is a Func, or an Action?

[jkane (John Kane)]

I don’t think this has been established yet, so excuse me if it has.

Is it Kinetic Code Camp or Epicor Code Camp??

[Evan_Purdy]

Pretty sure its calvin ball

[klincecum (Kevin Lincecum)]

BTW, if anyone uses ChatGPT to post tutorials in my thread, you are dead to me 4 eva.

[TDray (Todd Dray)]

What about Blackbox?

[Banderson (Brandon Anderson)]

That’s actually a really good idea. As long as they read the answer and make sure it’s correct, it would save a lot of time!

—runs off to create my insights presentation with chat GPT

[klincecum (Kevin Lincecum)]

@TDray

Blackbox?

FCS, I don’t even know what that is.
The world is moving too fast.

When is the Mars mission? I’m leaving.

[klincecum (Kevin Lincecum)]

@askulte

e.g. what’s a variable

[TDray (Todd Dray)]

Chat Blackbox: AI Code Generation, Code Chat, Code Search - Blackbox

It’s basically ChatGPT specifically for coding.

I am not saying the results are any better.

[Ernie (Ernie Lowell)]

There was a great article in The Economist recently that reminded me of something important…

All the LLM-trained AI’s are basically giving you back some form of “what most of their training data would be the best next word or phrase based on some algorithm of the number of times it followed what you gave it”.

Nothing in there about accuracy or factualness or what we, the great unwashed, would call “sensible”.

Just because much (even most) of the time it IS accurate or factual or sensible doesn’t mean it was designed to be so.

Verify first. That way trust is a lot easier.