i want to see the world- the runtimes- run
there's a bunch of modules for instrumenting webservers, semi-automatically-ish.
i also think about, for example, a counter for each standard library call made & resolved & failed.
created a very simple instrumentation for node.js that records the number of modules that are loaded right now. i expect most programs reach a steady state really fast. some probably don't though! and those that do? hopefully they stay running only those modules & don't some day start running more.
in future iterations i'd like to measure lines of code loaded. :)
I wish for an "architecture cartographer" - someone whose sole job is to map out and explain the relations between teams, projects, and services. Curated technical writing and data viz to answer the "who should I talk to about X?" and "how does data get to Y?" problems.
there's so much super good information about what everything is, what types there are, what constraints it has.
the platform is incredibly well described.
but oh no, all my other good work, my hard work, is node centric! oh egads.
it'd be much wiser to bow to reality, & target node.js fs.
i haven't stared at things long enough to figure out whether native-fs for node is easier or harder than node's fs for native-fs, but my gut tells me, this path is pain, this path is horrible, but go for it.
as well as "this is a wicg spec that could vanish overnight", i'm also setting myself up for a a nightmare of re-purposing, re-using, re-making the wonderful isomorphic-git project.
git is tech i need & thusfar everyone in js does it atop the fs libraries that actually exist. which they've painstakingly copied to the browser.
putting my eggs in the "native-fs" basket & woo boy am i both nervous & excited.
always bet on the platform? hopefully the platform keeps betting on itself! if this spec up & vanishes, i'm really going to wish i'd used browserfs or lightning-fs to get a node-like fs in the browser.
i think i know which way the winds are headed, but this effort could all vanish.
wifi makes me so sad. affordable arm boards & off the shelf m.2 wifi should make building your own AP easy, fun, & high performing.
instead it's all proprietary products. there's few up to date, competitive wifi, and most are still the old mpcie form factor. very few low cost arm SBC have any pcie expandability.
meanwhile rpi4 actually has pcie, to power the usb3. but the rest of the world is still in the 200x's. c'mon! please! open the future, open possibilities again.
i am sad that #wayland world has so much anti-pathy towards dbus. dbus is amazing. it is a world that shows itself.
wayland has many amazing powers but it's still so far from the discoverable universes of the desktop bus. it's tragic that there is so much resistance to further building atop the good, righteous base.
Early cybernetic lifeform, going beyond re-inventing to make virtualforms we can endlessly settle into.
Cybrespace is an instance of Mastodon, a social network based on open web protocols and free, open-source software. It is decentralized like e-mail.