Note 57: R.I.P. 2021 Reboot Week
Welcome back to Normal (or what passes for that in a fading pandemic)
I hope everyone had a safe and happy New Year’s Eve celebration last night. My wife and I celebrated by introducing a friend to Netflix’s Cobra Kai, and after she left, we rang in 2022 around the middle of the third episode of the new series.
It was an ass-kicking good time.
Reboot Week, Redux
The past week - the period between Christmas and New Year’s Day - is typically my favorite time of the year. As I’ve laid out before, I tend to take my winter holidays very seriously, with Thanksgiving as the tailgating before the main event of Christmas. This year was no exception, so by the time I hit the sheets on the evening of December 25th, I had metaphorically left everything on the field. I didn’t get everything done (Danish Kringles substituted for my usual Helmut’s Austrian Strudel - it was never in stock when I went down to pick it up), and I’m still working on getting a few overdue holiday (née, Christmas) cards out this week. We still have some Christmas dinner leftovers and snacks around, and I’ve been working on those since the holiday.
That is the basic theme of Reboot Week, what I call the final period of the year where I get to play catch-up with all the stuff I neglected over the prior 51 weeks, in order to get myself on a good footing for the next year. While most of Chicago’s Merchandise Mart is shut down during the week, the menu for the in-office days becomes Egg McMuffin sandwiches for breakfast and Quarter Pounders with Cheese for lunch, with copious amounts of McDonald’s Diet Coke powering the day. Last year, I used the period to revisit The Lord of the Rings film trilogy (theatrical, not director, cuts) while working on random tasks, and this year, I used it to finish up my podcast backlog, leaving me in a position to begin 2022 with around 70 unlistened episodes (subject-matter shows, not current events).
While I was listening to those, I used the period to do a lot of infrastructure work for my business, which should make my Robot Army much more effective in the new year as I have quite a few more projects launching and I am now responsible for well over thirty ongoing studies and sites. I knocked out some overdue data exports and got everyone on a decent footing for this Monday.
I also spent some time on internal projects like Fresh Comics, which is the comic book site and apps I’ve been maintaining for over a decade now. One of the tasks that I kept punting in 2020 and 2021 was linking all the graphic novel records I have (over 40K) to their corresponding Amazon pages to capture some of that fithy affiliate marketing lucre. What was funny is that when I dove into it, I discovered that I had written code in 2019 to do exactly that, then Amazon swept my legs and replaced the version of their product API with a new one incompatible with all the code that’s been sitting on my server for years.
I dug into the new API and created a new script that would look up Amazon entries from Fresh Comics titles, and as soon as I ran it, I promptly ran into an HTTP 429 error, which typically means that I’m making too many requests in a short period of time. The problem was that I encountered that error on my very first request, which was odd - if technincally mathematically correct. (One request in an instant extrapolated over non-instant time would be me making on average an infinite number of requests per second.) Looking into the issue, I discovered that in order to use the new version of the API, I need to sell at least 3 items in 180 days before being granted access. Given that of my 40K+ records, I have an infintesmal chance of picking the right three that someone will use to purchase an item. (In the mid-Aughts, Amazon became very good at detecting accounts related to affiliate account, as many of us were using it for an automatic 5% off anything we purchased then. Good times.)
So, given my chicken and egg problem, I took one pass at the problem with a command-line driven approach that allowed me to associate a Fresh Comics record with a short-link. (For the sales- and tech-impaired, Amazon includes the ability to retrieve a product’s link manually when you’re visiting the item’s page while logged in with the affiliate account.) This was useful for validating that the affiliate code worked, but ended up being too slow. My second pass ended up being an interactive text-based program that prompts me for the Amazon ID (ASIN) and link associated with a Fresh Comics record that is missing that link.
The process here is that the script scours my database for records missing a corresponding Amazon link, then temporally sorts them based on how close the item’s release date is to the present. Items that have been released recently or will be released soon are prioritized over those a bit further off in the past or present. (I also have a switch that will change the sort to priortize the pages that historicallybeen visited more often.) The program shows me the Fresh Comics URL, the item’s name, publisher and release date. Over in another browser window, I search for the item, and when I find it (or a suitable substitute - books get re-released all the time), I pull out the ASIN identifier and get the short link for the Amazon listing. I enter this into the command line program, and it presents me the next item in the priority list.
Using (and refining) this approach, I’ve tagged approximately 1% of the targeted records, so I still have a ways to go. However, I’m focusing on the most recent items - on the theory that those will be of more interest than a graphic novel released seven years ago - and we’ll see if this will be sufficient to capture those initial three purchases, which will enable the API.
And when the API is enabled, I think I’m going to keep a human in the loop regardless - the human will be responsible for verifying the link - since a lot of the Amazon metadata is pretty dirty. When I was doing this manually, often finding the right Amazon listing was tricky. When the search engine wasn’t recommending shaving razors to me (WTF?), it often recommended the in-house Kindle or Comixology versions of a volume instead of the physical one that I want to link to.
Overall, this was a weird task that took some actual time and labor - about an afternoon from beginning to end. It wasn’t the kind of thing that I would have space for in my normal hectic schedule, but getting to this point also allowed me to finally answer the question that’s been plaguing me for the past two years - why am I not attempting to monetize Fresh Comics via Amazon? (I do monetize with affiliate links to Things From Another World, which has a much smaller catalog.) The answer that I came to wasn’t the one I expected, but it was an answer and I was able to implement enough usable augmentation that will serve me well in my routine data updates of the Fresh Comics properties. We’ll have to see to what extent this compares with the TFAW affiliate links, which may be the topic of a future Note. However, given all the headaches and time spent, this was the perfect kind of task for Reboot Week.
Booting into 2022
I’m not a big fan of New Year’s resolutions, but I also spent my Reboot Week preparing a few things for 2022. I have three major new things that I want to do this year:
First of all, I want to resurrect The Pnakotic Atlas before August’s NecronomiCon Providence conference. NecronomiCon is a summer conference and convention in Providence, Rhode Island centered on the works and cultural phenemenon of H.P. Lovecraft. I’ve been attending for the better part of a decade now, and it’s an ideal place and time to re-launch an app that focuses on the geography of Lovecraft’s stories. In addition to revamping the UI and recuiting new artists to contribute illustrations, I’m also using this project to build out some blockchain infrastructure that will be the basis of the distributed offline-available database that will power the app. (I have zero plans or interest in doing anything related to cryptocurrency or NFTs.) If I can get some blockchain implementations and infrastructure under my belt, that will open up some interesting developments and may also generate some new consulting opportunities.
Secondly, 2022 is the year where I plan to Learn Music. I’ve been a music fan my entire life, but there’s still something very mysterious and gnostic about the entire topic to me. I’ve picked up some theory and instruction books and will be attempting to see if I can get to a reasonably competent place in being able to both talk about it and perform some songs (on the piano) by the end of the year.
This will largely consume the same mental space as my “read 100 books on even-numbered years” project (completed), and I’m planning on spending significant time on this. On the performance side, I set up a Yousician account which seemed to be a decent way to get started developing the physical aspects and muscle memory that I’ll need to play effectively.
Next, I’m working on getting back into the gym under a decent workout plan. Unlike prior years that had only nebulous goals with respect to losing weight or “getting stronger”, I’m approaching this year’s training as preparation for the 2022 sailing season. I have a full list of tasks and things that I wished I could have done better on the boat (need to work on those pulling and scrambling around deck muscles). My goal for the 2022 sailing season is to build on the foundation laid last year, and I’ll need some decent physical condition to make that happen. (It also won’t hurt a bit at all to shrink the gut a bit for scrambling in and out of hatches!)
Finally, I’m planning on getting back on a regular posting schedule here. I fell off the Substack wagon during the second half of 2021 and I want to be back on a weekly posting schedule this year. I’m planning on cutting back on my “commentary” over at Facebook to post more here, and I’m also planning on doing much less “hate reading” over at the sites of the Catholic Integralists and populist conservatives. I figure that taking that energy I’m using to tell folks that what they’re doing isn’t going to work is largely wasted breath, and I could use more attention over here on my own projects. I don’t know if I’ll resume with the original Wednesday posts, or move this newsletter to the weekend. If you have thoughts or preferences, let me know in the comments below.
One of my favorite discoveries of 2021 was The TRY Channel, which mostly consists of videos of “Irish People Try X”. Yesterday’s entry had them tackling American sodas:
While browsing, I also discovered that the Irish also enjoy alcoholic advent calendars, and spend over an hour and half going through each of 24 days of rum, vodka, and Scotch calendars. I haven’t watched it in full, but it’s definately on my to-finish list:
I can’t fully explain why I’m addicted to these videos, but every new one is a treat. And their ongoing Durian Fruit Saga also picqued my interest in what will be a foul and nasty taste experience.
If you’re looking for something to watch while nursing that hangover from last night, you could do far worse.
On that Note (see what I did there?), I hope everyone’s 2022 is off to a decent start. I haven’t been spending much time in Elite Dangerous, but I may give it a spin this weekend to see how VR CMDRs in the future are faring.
Until next week, o7!