pbjerkeseth

I hopped into this hoping it would land where it did! I recently built in basic observations about cyclomatic complexity, churn, and authorship into an 'analysis' view of codebases in ouijit[1].

Some benefits can be reaped immediately (cyclomatic complexity), but others emerge over time (churn). A good example of this is say something like a 500 line file that has experienced 2500 lines of churn, and if that rate of churn is trending down or up. Surprisingly useful for understanding you've got a hotspot with an opportunity to pay down debt by spending more time on API design, or just breaking out whatever subset of the code is experiencing thrash.

The funny thing about complexity is that assuming you're lintting/formatting well, you can do a poor mans check by just looking at something like average indentation per line, deepest line, etc.

[1]: https://ouijit.com

I read it as a pretty obvious jab at him being inhuman.

Desire is probably the wrong word, fixation is more accurate. Like the skiing advice that you shouldn't look at things (trees) that you want to avoid.

As for Utopias/Dystopias, I don't know how to answer that without waxing poetic about ideas like chaos being the natural order, or beauty being in the eye of the beholder, <insert other platitude>.

Yeah, this whole area of UX scares me. Silent speech techniques could strike a balance, but it seems obvious it wont stop there - consider what a state actor would do if they could skip interrogation altogether?

Unfortunately I think this is inevitable because as a species we perform a steady march toward collective desires for items of science fiction, so start your adversarial double-think practice now.

> Some tools have started making licenses that are free for individuals and startups but cost money for profitable companies.

Ive seen this as well, but at a certain point it just seems like if your code is online, someones either taking it wholesale or recreating core pieces from the spec regardless of any licensing.

Its a tough time to be building OSS if you don't have distribution solved AKA some pedigree from the before times.

Website not even needed! Janky terminal screenshot in the repo would have been good enough for me lol.

On another note, the reason I was perusing the blog was because I was curious about more of how exe handles review/quality/testing because you mentioned there being no code review in another post about stripe billings (maybe a different author). So consider this a casual request for more content on reducing delivery bottlenecks :)

So, I'm a subscriber/user of exe.dev but even so I was a bit disappointed when the meat.dev tool linked in the article had no screenshots/meaningful docs. So I installed it and was bummed to see it only supported openAI and exe.dev llm integration by default. "You can just fork" - yeah I know, so take it with a grain of salt.

That aside aside, I agree. Since reading this article yesterday I've probably been overthinking an MIT from AGPL license switch for my own project Ouijit (shill time: https://ouijit.com). Its feels a little counterintuitive since AGPL encourages more open source downstream, but at the same time if I have solved some problem other agent harness devs are curious about, I just want them to take the solution without worrying about paying it back/forward.

At my work we found lots of these types of memories to be tightly coupled to individual workflows and preferences, and shared things tended to be enforceable by other means (tests/types/linting/review gates). For anything that had team wide alignment, it's hard for me to imagine putting in some 3rd party platform as opposed to committed somewhere or a part of internal mcp tooling.

Not familiar with this space, is 'screening' some domain specific concept?

You don't need funding or teams to make these types of tools anymore, they are all basically engineering as marketing for running agents in the cloud, or some other value add/funnel into something that could make money.

codex(or whatever they're calling the bundled desktop experience) = selling inference

claude code = selling inference

conductor = selling hosted agents

superlogical = selling... something

I recommend everyone build their own. It's fun, you'll learn a bit, and assuming you work as a software engineer the dogfooding is built in. I built ouijit as my own example of this: https://ouijit.com & https://github.com/ouijit/ouijit (note that I am not trying to make money, just free as in libre AND beer)

Since you seem curious about these types of tools, might I interest you in ouijit https://ouijit.com/ https://github.com/ouijit/ouijit

Its intended as a terminal agent orchestrator aimed at power users who want a little more design than what I like to call 'hackerman' UXes.

a "username.tngl.sh" schema, which is that I meant by the unfamiliar protocol in the next bullet

First hearing of tangled, tried signing up and this first time user experience needs to be tightened up. Currently unwilling to sign in because of the friction I ran into using a password manager. From what it looks like they:

- ask you for an email

- send you an email

- ask you for a username

- except you cant actually log in with this username directly

- im being forced to learn some new social url protocol

- why does the auth flow pass me through a new ui/url that seems owned by the project but visually disconnected (eg, different branding/colors for the form)

- my password manager couldnt bridge the gap

I'm notoriously fickle about dealing with signup/login friction, but the project sounds cool so hopefully my feedback is more actionable than curmudgeony.

Discovered data-star a little around that time, maybe a little after HTMX. I'll probably dabble with personal projects, but no stomach for the business to try something in that vein again. "Nobody ever got fired for picking ~SAP~ React" and all that.

HTMX is excellent. We made it a long way at Convictional[1] with HTMX + AlpineJS, but the eventual transition of our product into lots of live collaborative surfaces had us feeling like we had pushed the envelope as far as we could under modern startup constraints. Unfortunately, frontier models have really hurt development with budding tech that doesn't have the training data presence of things like React.

[1] https://get.convictional.com/

Thank you!

"Prolix is an adjective describing speech or writing that is excessively wordy, drawn out, and tediously long"

Prolix affected me at least once today, I appreciate this.

This is a neat idea. Definitely needs tuning.

- I tried it on a repo that has very obvious boosting tells, like 99% of stars from week old or less accounts and it had higher trust signal than my own very much so real project. The issue to star ratio gave 5/5 despite the issues being core contributor created.

- In same vein as above, it seems to prefer open issues over closed ones in the scoring.

Have you considered scoring contributors/authors/issue creators separately to use as a proxy, instead of measuring the core things that are being gamed?

Hard to trust a project that is clearly github star boosting out the gate, consider that you're putting yourself at risk of account suspension for reports.

Ive been surprised at how effective (for personal projects) this type of rote promotion by link sharing can be. It feels fruitless initially but i guess whatever-engine-optimization keeps paying to be consistent.

awesome-lists like you see on github seem to be missing, which would be a good inclusion. Although I think many of those curators are suffering a wave of slop right now so maybe not.

Funny seeing that post, looking into it, asking the same question, and arriving here. Grifters gonna grift.

Every task automatically gets git worktree isolation to support the parallel work. The VM sandboxing is project scoped and more about isolating agent workloads from user-space for those who want to --dangerously-skip-permissions for example.

No problem, I actually went through a similar path of trying milkdown/tiptap/a few others as the core for my own editor needs but kept running into into issues where the abstractions got in the way eventually. I was thinking about using ProseMirror for a custom 'one big text file' concept so I guess Ill close the door on that idea and give this a try.

Nice work! It seems like selection highlighting(?) doesn't work but the interaction feels good otherwise.

I'm curious when I see things more geared toward prose using CodeMirror instead of ProseMirror. Any comment on that decision?

Blood in my case, and it would probably be obvious. In general though I think a lot of these questions are answerable by paying attention to the changes in your body, how you're feeling, researching, and raising thoughts/concerns/etc with your primary cary provider when that comes up short. No need to go searching for issues if there isn't some leading indicator that sticks out to you as something to be curious about, or there isn't some related family medical history.

I recommend getting a colonoscopy if you have any symptoms. There is a lot of stigma that prevents people from being proactive about this type of issue.

My anecdote (M, 35) is that I got one after experiencing symptoms that turned out to be unrelated, but they did find pre-cancerous polyps so now I will be getting them more regularly. I received received meaningful early detection and peace of mind. Also aside from the prep, its a very convenient procedure. You get put under anesthesia and do a quick time travel.

Ha, it obviously picked up on the ouija board concept with the colors and copy. I think the detail expansion shown is a great illustration of another thing they tend to do - uncanny valley of 'sky is blue' type product descriptive copy that reminds me of dribbble shots or landing page template demos from a few years ago. Technically correct language that doesn't communicate much emotionally.

I've built/am building something similar, but I spent the first half of my tech career as a UI/UX designer before becoming a software engineer and I'd _like_ to think it shows, but there is something about designing-in-code with agents that leads to homogenous outputs if you don't spend equal time on visual design as on the technical parts.

I'm a bit anxious about putting myself out there, but I'd be curious if my efforts cross that bar for you or not? https://ouijit.com/ (and the repo is at https://github.com/ouijit/ouijit)

This has unfortunately been going on for years at this point, for as long as there has been an OSS-to-profitability pipeline gamed for startups I'd guess. I wouldn't be surprised if it has progressed to fake contributors/discussions/issues/forks as well. Seems like an inevitable outcome for any platform with social signals.

I have a terminal manager project[0] I'm currently using xterm for, but very curious to try libghostty. Have mainly been hesitant because it hasn't been promoted from an internal ghostty dependency (only awareness of the place was from this article by the creator[1]), but from the sounds of it here people are finding it stable enough. Gonna give it a whirl today.

[0]: https://github.com/ouijit/ouijit [1]: https://mitchellh.com/writing/libghostty-is-coming