juan M

Weeks after Elon Musk’s finalization of the purchase of Twitter, a former colleague and friend asked me through whatsapp the following question

What do you think about Twitter’s proposed changes? (Paying for verification)

My first answer was

“verification” is a strong word, as we know working for fintech verifying that a “user” is a human being (or a given human being) it’s a tricky problem

Technically for one thing, but also complex from a jurisdiction perspective, etc

I am currently of the camp that it’s fine they try monetisation avenues, but not sure if “identity verification” and/or the bot/spam problem should be addressed in the same way

A few days later I expanded with the idea that it was hard to analyse Twitter with a single lense, as it used in a number of different ways, which I called “use cases”. Funnily, I ended that thread with

waiting for an application of GPT-3 (like Lex), that would that these messages above as input and create a nice blog post for me to publish 😝)

This was probably 2 weeks before chatGPT was unveiled. I had forgotten about this until today when I was checking my phone, so I gave it a try, so here you have below a chatGPT version of my messages, for which I used the following prompt:

“Hi, this is a transcription of a number of messages I wrote as part of a whatsapp conversation. Can you please convert this into a post I would put in my personal blog?

[+a crude copy&paste from my mac, with timestamps and other stuff whatsapp adds”]”

As with a previous example What will be the impact of chatGPT on society (a very slim reply), I felt it stripped several concepts I would have liked to keep, although I reckon that I might be too verbose in my writing, so I guess I’ll have to iterate to find a sweet spot. For now, please find below the GPT-ed post:

Twitter: A reflection:

Twitter has evolved into much more than just a social network, now it’s also a product and engineering company. The question is, what should Twitter be and how can we build a business around it? In this reflection, I’ll go over some of the unique aspects of Twitter that make it special and discuss some of the challenges that come with it.

1. Twitter as a source of direct voices from known or famous people: Twitter is a great platform for people to hear directly from famous people, without any third-party mediation. Twitter’s short text format allows for direct and unfiltered voices to be heard, a feature that other platforms like Instagram and TikTok can’t match. Twitter may explore licensing its content, but this could result in losing relevance and distribution.
2. Twitter as a source of learning and discussion among smart people and interesting topics: Twitter is also a great platform for people to learn and engage in discussions with smart and interesting individuals they would not normally have the chance to meet. Twitter needs to focus on discovery, curation, and filtering to keep this use case relevant. Twitter could capture some of the value created here through targeted ads or finding ways to get people who get value from it to pay.
3. Twitter for random snippets of brilliance, comedy, current affairs, and trending topics: This use case is where most people spend most of their time on Twitter, and the value created here is highly variable. Brands and organizations also have a presence on Twitter and use it as a way to voice their opinions.
4. Twitter as an alternative source of information and news: Twitter is also used as an alternative source of news and information, a function that was not initially intended but has since become a common use case.

In conclusion, it’s important to be explicit about which use case we are discussing when it comes to topics such as verification, moderation, and curation on Twitter. The categorization of Twitter’s use cases should be reassessed, but the overall goal should be to maintain Twitter’s unique features while also capturing value and being mindful of the challenges.