TMSR work: interlude and plan for 2019 M7

Following recent discussion in the forum, as well as my own recent (spectacular) failure to report, let alone deliver, on yet ongoing work, in this post I'm reflecting on what's going on in spykedlands and proposing to take the bull by its horns from a different angle, with the aim of doing stuff more efficiently.

Long story short: on June the 22nd, Mircea Popescu observes a marked silence among the Lordship, including yours truly. This is followed by short reports and further discussion on the matter, including the problem of allocation priority for existing things on the to-do list. This is when it first becomes clear to me that my previous reporting style hasn't been working¹ and that I need to change something, although I'm not yet sure what or how.

Then the latest S.NSA report comes, and... well, read the whole discussion and draw your own comments, thoughts and conclusions from it.

As for my own thoughts: I happen to be in a situation somewhat similar to Stan's, so I understand the man's trouble. In theory it's easy to say "next week I'll allocate X hours out of a total of Y [of which Y-X must be allocated to work in the salt mines] for completing item I", until in practice, occasionally, surprisingly and in the last moment, your favourite saeculum overlord decides to sabotage you and drive Y-X up, so that X becomes close to zero. What's more, having both X and X-Y time there leads to context switching overhead and the reporting component on X is itself costly², which then spills into personal life and leads the more untrained ones to lack of sleep, and ultimately to a nasty burnout. All that being said, complaining about it won't be of any help and getting into that fucking hustle is not an option, iar vieața-i grea și nu toți poate, or how did the saying go.

So since properly managing my time is but the first necessary step in the great process of growing my own legs, let's see how I can tackle this issue. For one, I'm stating upfront that my to-do list as well as my priorities are consultative and not set in stone, and no, I'm not in a marriage with Hunchentoot and whatnot and I can move up and down the stream as the more pressing needs dictate. For now, as per previous discussions, I'll act from the cause that a working CL-based WWW stack is still a priority.

As for the other, my previous reporting style of giving approximate time intervals for large tasks is an utter failure, so I'll need to spend more time breaking everything into small pieces and report as often as possible on what I've found. More concretely, I'm aiming for monthly planning reports such as this one (minus the lengthy intro), plus weekly or at worst bi-monthly reports on the things that I'm working on. For now I'm resolving³ to do a planning report at the beginning of each month (though this may change depending on whether planning itself becomes a lengthy task, such as in the case of Hunchentoot this June) while the weekly reports are set out to be:

• Week 27: I will publish a cross-referenced version of the Hunchentoot coad, and the coad that makes cross-referencing possible, for subsequent use.
• Week 28: I will be away from the workbench, visiting yet unexplored meatspace, but will be following the logs closely and I'll be available for emergency situations, e.g. Feedbot maintenance.
• Week 29: I will publish an architecture of Hunchentoot, with comments for each component and the interworking between the parts.
• Weeks 30, 31 and onwards: I will publish detailed write-ups on Hunchentoot coad pieces, to be further discussed in week 29.

That aside, I have another three or so drafts of photoposts and miscellaneous writings lying about waiting to be posted, but publishing those depends on whether or not I reach the targets above early.