2022-W02-W37, 2022-01-09/09-15

almost three quarter of the year 2022 is gone without a full log written during it.

from memory i remember two things: dae.wiki and a job application. [^work-org-tools] buycloud also took place, and twla. these projects have been completed, and the last two are now in either maintenance mode or archival process.

smaller things happen in between, primarily some existing projects in maintenance mode: some have been converted to a major version of the underlying software, others are in constant patch-up flow.

[^work-org-tools]: curiosly, i went to use a small script i put together to check when i updated the work file of each project, and while i got back some info, i also got reminded that during summer i changed how i use those work files (the system has transformed) and am not sure how much is accidentally being left out of it.

writing this entry feels difficult, as many processes have ended. i ended them. at some point this action began to feel alien to my normal array of operations, as more projects took on a long, stretched out forever trend. this year i put an end time on them, and it's a great feeling. as it is re-discovering the "taste" of this action: is this an example of manifesting and experiencing agency?

dae.wiki

newest final release was planned to be for mid February 2022. a mix of problems pushed it further of a few weeks into early March. testing began from the school side and bugs began to be reported. a possible problem (up to that point only a big consideration) at last took shape and while it made me die i already looked into it at the very beginning of the project 3 years ago and i knew how it could be solved. due to this last point, i asked more time (3 weeks) after the agreed upon delivery was finished. the list of bug report grew and began to mix with wishes, last minute disagreement and overall misunderstandings of how things should work on the website. after 3 years.

i fixed all the bugs in the list and discussed those points which were new request (see disagreements and misunderstandings). spent the summer to implement many of those new points and integrated everything into a Docker environment. wrote setup scripts and documentation, added more code to make the initial setup smoother.

in mid August i uploaded both the dae.wiki repo as well as the dae-infra Docker code to a new code hosting account handled by the school. i told my delivery was finished, ready with all info necessary to run it. sent the last big invoice for the work done in the previous 8 months — as the delivery stretched from March to August.

as of the time of writing (2022-09-20–22), the same endless cycle of acting completely lost, with no idea of how to run the website, how to manage "the last part", etc from the side of the school has taken place again. i made it clear my time is up at every of their attempts to flip around the story of whom responsibility is for this and that. not moving from my position.

job application

in early spring i applied for a job application at a US company. it was for a front-end position at a p2p company building a new computer-stack from scratch. they gave me a bounty to work on, eg a real Github Issue, for the span of 3 days and got paid for it. did an interview before and after it.

ultimately did not get the job because i am not a good fit as "i never worked in a medium-big team and have no direct experience with particular code frameworks". both reasons reminded me why i like freelancing!

buycloud

this year i finished to work on the urcloud part of Buycloud, which is the archive of the project. the main Buycloud website lets you buy a cloud in the context of an exhibition. all the bought clouds are then displayed in the archive website.

this website also enables the second part of the buying process, which is to put the clouds in a pacman-like game environment and make the bigger clouds eat the smaller ones. while at last this feature was not enabled in the current online website, it's there. besides, clouds roam on the sky and exit / enter the "sky" space as in pacman.

an index list of all bought clouds — with info of whom, when, and cloud's features — is available as well.

the project took a longer time due to the initial challenging collaboration process. not having a shared language to talk about things and express ideas made communication sometimes confusing or produced misunderstanding. all in all a very nice project! i would have enjoyed exploring even more the game element of the website, but it was a good intro nonetheless.

on the other side, it was a python-based project after a few years of doing nothing like that, and i learned that python slows down small VPSs with few resources a lot. while using it was a good experience, i began to dislike that it requires so much computer resources to not slow down the whole machine. well!

TWLA

another stretched out project that i completed was TWLA. roughly, it was concluded around February-March, and then smaller patches of work were performed in the months following.

i liked i had the chance to look up and use the git bundle function to share the project repo, and upload it online at a public URL (on my VPS). this was necessary due to another institution mis-management of departments. originally we agreed i would have uploaded the git repo to a hosting service, but 3 months into the project a new ITC team took over the previous one and all of a sudden "no one knew anything". ah!

smaller bits

i upgraded two websites i built almost five years ago to a major version of the CMS software (kirby 2 to kirby 3). had a problem in upgrading one plugin, and it took quite some time to find a way around it.

another 3 year old project received a few small updates, as it did every now and then over the years.

began to read and learn Common Lisp. been wanting to do that since i moved to Berlin 3 years ago, but during the first attempt things collapse under over-complications circumstances.

now i want to try it out as a possible, more fitting, programming language to work with as a (mostly) one person freelance biz (at least programmer in the team). a constant "live" environment, debuggable, fast, compilable to one binary file (the whole website code), documentable. and... less dependency problems? less anxiety induced from a new web framework, that a version of something changed or is not supported anymore, etc. and at the core, a different approach to work and build with code.