Prioritising 50 Projects I

The Weasel suggested, wisely, that I need to prioritise my projects. There's a lot of them, so this might take several posts.

    1. cataloguing all my books including ebooks.

Most of my physical books are still in my storage unit in Brentford with some in Preston.. I am hoping to arrange for them all to be brought to Seattle when I am in the UK in May, so perhaps they will get here in by late June or July. Getting them shelved and then organised will be no small tasks and there's an interesting debate to be had about how one should organise a private library. Of course, it might be possible to catalogue them without organising them. I believe there are apps that will scan books. There are various possibilities for library databases, so this will be something to thought about and looked into. Ebooks are both easier and harder. I have my Amazon UK and US plus lots of things on my computers and tablets. The old UK Kindle Fire is probably lost.     

    2. securely backing up ebooks and various electronic documents.

Needs some kind of media server as well as offsite backup. Probably 2026.

    3. digitising tapes and, possibly, CDs.

Who knows what condition the tapes are in. There's just a few really that need digitising. A fair number of CDs in the storage unit. DVDs too? I would need to get a tape and disc deck to this. Probably 2026. Would also require some kind of media server (see above). Needs to be thought through.

    4. finalising MSs of the first volume of The Mundanomicon and Eleven Short Stories About Love.

This can be done now and TM is the nearer to finalising of the two. A few weeks would be enough, but there's all the other things to do too... But what excuses do I have?

    5. attend Hugo House Year-Long in Speculative Fiction course.  

They did a separate science fiction course one year. Expensive as an online course, but online is probably handier for me these days. Might be a possibility for autumn 2025, money allowing.

    6. read Anna Karenina.

My copy (Oxford World Classics edition) is in storage unit, so, I hope, it will be revealed when I unpack them finally on their arrival. Ideally, one would read it in the original Russian. I know, I know. Probably late 2025. 

    7. build radio telescope.

The kit is not that expensive, but I will need to have sorted out my dry lab in the basement and to have gained some kind of experience with electronics, so I think this is nt going to be until 2026.

    8. get optical telescope and observe bright objects in an urban environment.

I understand that a halfway decent beginner's Celestron refractor costs a few hundred dollars (up, of course). Will probably have to wait until I have a job and have built up some reserves, so 2026. 

    9. complete Udacity nanodegree in Deep Reinforcement Learning.

Got to keep plugging away. I am currently ~50% through, but the Jupyter Notebook environment for Windows does not play well, which is irksome. Might have to wait until I have a Linux box to complete the project.

    10. write AlphaZero/MuZero-style DRL system that plays Avalon Hill Baseball Strategy at a superhuman level.

I need to understand how to play BS to start with! Then a computer version that can self-play, which should be reasonably straightforward. With a fast enough computer, it should be able to train itself. Of course, it's just the time to do this as it's a non-trivial task, but would be fun.

Comments

[Celestial M Weasel]

How does one write software to play the game? Is there an API or a UI conducive to automation? My experience (for money) is that using software to automate UIs is a collosal pain in the arse

[Dr Paul M. Cray]

It's a boardgame. There's no UI until I write one and it would be APIised, to allow either two humans, two computer or one human and one computer to play. An example is given here: https://www.manning.com/books/deep-learning-and-the-game-of-go