Pardon My Dust

My recent efforts on this site didn't exactly go to plan. Apologies, this post is going to be a bit stream of consciousness writing.

I will say some of what I had done recently worked really well. I think going to a markdown driven site was a good call. Additionally, using GitHub to drive everything was also a good move. Being able to work on the site from virtually any machine has been really good (I'm currently working on this through my chromebook, but the site is now to the point that virtually anything with a keyboard will work).

Some things didn't work so well. I never really developed a good workflow to capture information hasn't gone so well. Thankfully, I haven't spent a lot of time reading lately. The professional knowledge I've been working on has been captured through other systems. I guess that is good in that it isn't lost, but I'm never going to benefit from a synergy by keeping multiple systems. Second, Jekyll and I just don't get along very well. I don't know Ruby, so I struggle with that, but also there's just not as many examples of other Jekyll sites from recent times that I can grab ideas from to implement. While I don't mind some customization, my goal in this effort is not to showcase front end development, rather to deploy a system with as little friction as possible.

So I've been trying to work out solutions on both fronts. For this website, I've decided to switch to Hugo from Jekyll. I wanted something a little newer and simpler, which so far has seemed to be true. This new site has a bunch of new features I wanted to implement already baked in which has been great. Second, the engine behind it is extremely quick which has been great early on to preview changes. It's far from perfect, but I can quickly implement tags, categories, thumbnails, sections, navigation, etc fairly quickly and withough digging into an unfamiliar templating system.

I'm not entirely sure about my workflow yet. I've made some initial advances that I think are going to work. I've started using GitHub issues to capture quick thoughts. This has proven to be outstanding. I can easily attach links and notes, it works on pretty much any device so I can capture ideas from anywhere. My next step to implement is to set a timer on closing issues. I'm probably going to use a kanban approach to slowly ratchet up the throughput of the system. Many of my current systems suffer from collection overload. I end up sending more stuff in than I can ever consume, which made each system useless. At some point I thought about trying to implement a smart system that could help me process that information, but recently I've decided that building in something that was forgetful isn't such a bad idea. In my work, there's lots of information to process of varying degrees of utility. I should be a lot more selective, but in the short term this should at least keep the system from breaking.

Step 2 has been a bit harder to come up with. My current website iteration has split out a curated reference I'm going to make visible publically with a separate section that is more of a working memory. I'm probably going to set the latter up in the spirit of a zettelkasten, but I've been slow to get started because my limited attempts thus far have resulted in things too much like a wiki, rather than the resource I had imagined them to be. I suspect that's just me not trusting the system, but I have a history of failure to consider.

I'm going to continue to work through my workflow issues, but I wanted to go ahead and get this new site deployed and start thinking about other things. I hope that I can leave this one alone for more than a year.