This series is aimed towards bloggers or researchers interested in creating databases for their work. So the challenge is to generate an external database for my blog posts. Needless to say, WordPress does an excellent job of creating a well-structured database in order to ‘serve’ the posts in the cloud. There is a real benefit though from having a separate database for analysing the content and i’m sure there’s some really valuable uses for this that I haven’t quite figured out yet.
So after some initial ‘discussion’, the conversation quickly turned to JSON, SQL and Python code. The issue is, I just want to keep it simple at the beginning and then iterate. I’m hoping along the journey, I get to learn some useful skills that I can apply for the management of references. So it turns out, from the conversation that the simplest approach is just to have a text file with some structured headings and just to test this with a few posts.
Now the problem with this, is that even with a few posts, a large number of different headings are possible. Then it becomes a matter of ‘editing’ the headings to produce a ‘sensible’ table within the database. So even in databases, just as in the posts, there is an editorial decision to be made.
So the next thing is that there is a terminology to learn. These three posts are assembled into a database which contains all of the information. Within the database there is a table which collects together similar things. In this case there will be a table of posts. This table has fields. The fields are the different headings which require some editorial decisions. Each post is then a record.
The next issue then is data entry, because having generated potentially useful headings, it becomes time consuming to enter the data in the fields. That’s where Chat GPT comes in useful because I can just scoop up the post, give it to Chat GPT and get it converted into a text database.
This doesn’t look sophisticated and it isn’t. A text file with some structured text, but it is a database. What it is not, is a database management system but as I said i’m keeping it really simple.
Chat GPT 5.4 produced a 21.4 Kb text document with a number of headings including:
POST_ID
TITLE
URL
DATE_PUBLISHED
AUTHOR
TOPIC
TAGS
POST_BODY:
REFERENCES:
PREPARATION_NOTES:
DISCLAIMER:
As expected, the textfile doesn’t look human-friendly but that’s not the point of it. I’ve achieved the first objective of getting a simple database of three posts and i’m hoping the next steps should get more interesting in terms of working out how to manage the database and review the value of the fields.
Notes on the Preparation of this Post
This is a technical database post, so the preparation was a bit different from usual. For these kinds of tasks, AI is incredibly useful. So I started with a chat with Chat GPT 5.4 Thinking. So there’s a lot of back and forth and Chat GPT 5.4 as expected from leading large language models is very good at generating high quality text, with expert level answers. The blog on the other hand is from my perspective about the human experience of all of this and i’m sure it must be the same as for many others. Needless to say, the ‘conversation’ quickly generates large volumes of high quality text akin to problem solving about the specific database challenges in this case. I could have included that here but given the ubiquity of these conversations and reproducibility by others, I took an editorial decision not to include it. Reflecting on this, the reader may be interested in the prompting and i’m always mindful of how I can improve on my approach in this regards. Another point is about the quality of responses in a multi-turn conversation based on recent research which favours a well structured initial prompt. Maybe that’s something for another time.
I’m more mindful these days of trying to promote human artists where I can, so i’m including another artist from the Pexels collection.
Disclaimer
The comments made here represent the opinions of the author and do not represent the profession or any body/organisation. The comments made here are not meant as a source of medical advice and those seeking medical advice are advised to consult with their own doctor. The author is not responsible for the contents of any external sites that are linked to in this blog.
