· Mar 12

[Water Cooler Talk] Is ChatGPT effective in providing ObjectScript guide?

Hi Community!

As an AI language model, ChatGPT is capable of performing a variety of tasks like language translation, writing songs, answering research questions, and even generating computer code. With its impressive abilities, ChatGPT has quickly become a popular tool for various applications, from chatbots to content creation.
But despite its advanced capabilities, ChatGPT is not able to access your personal data. So we need to build a custom ChatGPT AI by using LangChain Framework:

Below are the steps to build a custom ChatGPT:

  • Step 1: Load the document 

  • Step 2: Splitting the document into chunks

  • Step 3: Use Embedding against Chunks Data and convert to vectors

  • Step 4: Save data to the Vector database

  • Step 5: Take data (question) from the user and get the embedding

  • Step 6: Connect to VectorDB and do a semantic search

  • Step 7: Retrieve relevant responses based on user queries and send them to LLM(ChatGPT)

  • Step 8: Get an answer from LLM and send it back to the user


  For more details, please Read this article

My personal conclusion

In my personal opinion, ChatGPT is effective in providing ObjectScript code and examples, especially for simpler tasks or basic programming concepts. It can generate code snippets, explain programming concepts, and even provide examples or solutions to specific coding problems. However, the effectiveness may vary depending on the complexity of the task and the specific programming language involved. ChatGPT can be quite helpful in providing code examples and explanations.

Is ChatGPT effective in providing ObjectScript guide?
Discussion (6)5
Log in or sign up to continue

We've been looking into custom GPTs -- thank you for the diagram of your process! 

With our team, we found that if you are already solid in ObjectScript, quite effective. Beginner -- outright disruptive. Intermediate, somewhat effective.  

The skills that are in play are code evaluation, analysis, and troubleshooting to make things work. Those are higher order skills than code generation. Trick is how to train for the intermediate to advanced levels to start coding with ChatGPT? I fully expect the upcoming devs to warp speed ahead with these tools.  

I will definitely agree with @Chi Nguyen-Rettig - it may be pretty helpful for some basic stuff you forgot when you're middle or senior developer and when you know exactly what you're looking for. So it's basically a bit like doing the usual search but without the need to wade through the search results.

But when you're talking about beginners - I see all kinds of things from my students (who for some reason just don't want to open the lecture and prefer to get answers themselves from other sources) that have nothing to do with the reality and don't work most of the time.

All in all, I wouldn't use ChatGPT to write code that does something complicated. But it may or may not be a good source answers for simple questions. And in this case, you still need to understand what's going on to be able to assess the correctness of what you see. Case in point, some spam posts on the Community - people try to cheat and write "articles" using ChatGPT which generates class/package names that don't even exist.

And if you think about it, where does ChatGPT get its answers? From the knowledge base somewhere. So if this data is not there, it can extrapolate and come up with something that may give you general direction but may not work.

I tried out GPT 3.5 for ObjectScript a while back just for fun, and it came up with some really plausible-looking hallucinations with class names that don't exist, but sound like they could. GPT4 is probably better, but I still find it likely that a beginner to intermediate developer would end up getting confused about why the output doesn't work.

I have a lot of opinions about good programming patterns in ObjectScript (especially in intermediate-to-advanced areas), and I doubt there's enough publicly-available good ObjectScript code in existence to serve as training input.