Love.Law.Robots. — 27 August 2021
August is almost over! With this, I have completed 3 months on Ghost. I am still feeling my way around this newsletter thing, so bear with me as I continue working on it. One of my new experiments is to use little title cards to feature quotes from my article for social media marketing. Like this little gem from my last post:
Hmm. Needs a tad more work, back to Inkscape.
What I am reading now
- The last newsletter featured Tech.Law.Fest in some detail, and this week featured a bigger conference on the other side of the globe — ILTACON. You can follow some of the reporting done by Artificial Lawyer here. It's interesting how different these conferences are. In Singapore, the discussion easily conflates the technology of law and the law of technology. See this Exhibit A. Innovation doesn't happen just because you know a little bit about the legal issues surrounding artificial intelligence or blockchains.
- Here's something about coding. SQLModel was dropped suddenly, and if it is anything like FastAPI, I will love it. For folks who don't know SQL (pronounced “S Q L” or “sequel” depending on how old you are), it is the database that runs the world. The language that runs queries on SQL might look like an Excel formula on steroids, but I have been putting off learning it (including ORMs like SQLAlchemy and tortoise). If the interface were more like Python, I would really welcome learning one less thing. Databases are essential to any application, so this would open up new possibilities.
- IAPP's Asia Privacy Forum for 2021 is online only with a significantly scaled-down program. Now I am having serious doubts about whether Tech.Law.Fest will feature any in-person content. Furthermore... where am I going to get all these CPE points for my certification? 😵
It's great to hear good things about your posts from time to time, and this week features “Mining PDFs to obtain better text from Decisions”, which was described by a reader as helpful and inspiring. The post talks about reverse-engineering PDFs, which is the prevailing format in court decisions. I've always wondered why I didn't think about using the formatting of a document to improve parsing its information. It's literally right in front of me! Inspiration is such an odd journey, which is why it can be exciting. I am thinking of employing the technique to lift text from all sorts of decisions, which will make them more accessible.
Mining PDFs to obtain better text from DecisionsAfter several attempts at wrangling with PDFs, I managed to extract more text information from complicated documents using PDFMiner.Love.Law.Robots.Houfu
I am still experimenting with the right balance between free, subscriber and paid content so that you might notice some A/B testing or modifications to my approach.
I also just learnt that I am supposed to prune my subscriber list for non-active readers. If you are reading this, you aren't likely to be affected. In any case, I would send a reminder email before making any drastic moves.
Thanks for reading, and feel free to reach out!
Love.Law.Robots. – A blog by Ang Hou Fu