My initial thought was wow, this is going to truely be a game changer. I could imagine that I could test hypotheses in near real-time and get amazing insights using the largest body of knowledge available that is powered by probability.
Being curious I want to try to work out the workings.
I assumed there was some prompt engineering using the inputs, but was interested in the GSS training model and how that would come into play.
So I was like lets see first how good or bad ChatGPT is, it was poor, though I put zero effort into optimising the prompt.
```Using your knowledge of business owners, calculate a plausible distribution of responses to the question: What is the biggest pain point for your business that a SaaS platform could potentially solve?
A. Inefficient project management
B. Poor data management and analytics
C. Ineffective customer relationship management
D. Challenges in managing remote teams
E. Lack of robust communication and collaboration tools```
ChatGPT (GPT4)
```
A. Inefficient project management - 20%
B. Poor data management and analytics - 25%
C. Ineffective customer relationship management - 15%
D. Challenges in managing remote teams - 20%
E. Lack of robust communication and collaboration tools - 20%
```
The results were not that exciting and very neat (20%, 25% etc.)
I was ready to see how much more accurate RoundTable was...it was exactly the same.
Makes sense. The further away the target question is from the GSS training distribution, the more it relies on the ChatGPT prior. I assume if you click 'Investigate Results', the confidence is 'low' and the most similar questions in the dataset are pretty far off
Please visit this group in FB which was created to register all the concerns from hosts and guests from all over the world. We can collectively work together to fix this situation.
https://m.facebook.com/groups/1316864141779135/
WHAT??? Okay I am going to bite, state is managed by the browser and then SPAs came along and tried to hack the state management ever since. I remember creating SPAs in 2004 when it was called AJAX and there was the debate of HTML or JSON over the wire. The biggest issue was back and forward buttons, even today that issue is not solved very well by SPAs.
The issue I have is we had a great idea to refresh parts of a page with AJAX and then we went overboard and tried to move the entire application to complex frameworks which add more problems then they solve. We became so scared of screen rendering we avoided at all costs. This is bad UX as users actually get good confirmation that something has happened when there is a screen refresh.
I think it is more to do with your maturity allowing you to see more of your weaknesses.
When you are young, myopia makes you think you know it all and can solve everything. Young devs run gun ho into things and they focus on happy paths and ignore a lot of the complexity that you learn from experience, noting there is some exceptions.
The second problem is scope, when you are a junior you normally have a single project and a lot of guardrails to support your learning, and typically just need to think about a small bit of code. As you become more senior you are suddenly responsible for a P&L, other people, lifecycle management, stability and scalability.
Last year I was feeling overwhelmed and couldn't learn Quarkus and Java. Turns out I just needed a week of focus. I was able to take my years of .Net, PHP and JavaScript experience and produce amazing code in Java.
My biggest blocker to learning was managing other people, projects, scope, budgets, requirements, approving leave requests, preparing status reports, managing vendors, interviewing candidates, dealing with shareholders, backlog grooming, workshops, committee meetings, CAB preparation, ARB negotiations, and the desire to be instantly good.
This is also something many forget. Learning takes time. I learn one simple concept in a day or so. A framework? That is going to take a couple of weeks at a minimum. Staying motivated during that time can be difficult if you have nothing to directly apply what you just learned on. The old 'teacher when will I ever use this' question. For many if it is not right away it feels silly to do it.
Being curious I want to try to work out the workings.
I assumed there was some prompt engineering using the inputs, but was interested in the GSS training model and how that would come into play.
So I was like lets see first how good or bad ChatGPT is, it was poor, though I put zero effort into optimising the prompt.
```Using your knowledge of business owners, calculate a plausible distribution of responses to the question: What is the biggest pain point for your business that a SaaS platform could potentially solve? A. Inefficient project management B. Poor data management and analytics C. Ineffective customer relationship management D. Challenges in managing remote teams E. Lack of robust communication and collaboration tools```
ChatGPT (GPT4) ``` A. Inefficient project management - 20% B. Poor data management and analytics - 25% C. Ineffective customer relationship management - 15% D. Challenges in managing remote teams - 20% E. Lack of robust communication and collaboration tools - 20% ```
The results were not that exciting and very neat (20%, 25% etc.)
I was ready to see how much more accurate RoundTable was...it was exactly the same.