William Ikenna-Nwosu's Blog

Hello to My Readers 👋🏽 I know it's been a long wait but I want to thank you all once again for your patience. I can tell you the delay for this issue was due to the complexity and size of the topics I intend to address in the next few issues. I want to do them justice so it requires deeper research than usual. Recap of Previous Issue ⬅️📰 In the previous issue of NPT, I gave my take on GenAI chatbots. It was a more practical issue of NPT but I hope it helped because it helped me reflect on GenAI chatbots and I use them for and what they are good at so far in my opinion. Foreword 🗞️ In this issue of NPT, I will be talking about the ActivityPub Protocol created by the Web3 consortium. It's a protocol and open standard for decentralized social networking. Hence, the aptly named title of this issue of NPT 😁. Why should social networks be connected and interlinked? Find out below as I conduct a sort of cost-benefit analysis comparing some pros and cons as well as dropping some i...

It has come to my attention that art is not being appreciated for what it is and craft is not being respected for what it enables artists to achieve.  This is a great disappointment in our society and it is being exacerbated by artificial intelligence (AI) practitioners who feel because their models can achieve similar results then the work that's gone into creating those wonderful works of art was not so strenuous and of low effort. They are mistaken. It is only something that has been written well that can be read with ease. A good teacher amplifies a good student. A clean and diverse dataset makes for models that have learned from the data in an optimal and flexible way without overfitting or underfitting. We are almost looking at a sort of P vs NP problem where we say "Because the results are easily reproducible at scale, the dataset is cheapened and the value it provides is not recognized and the effort to create it is not appreciated". This is the wrong way to look ...