I’ve been tinkering with an LLM-based assistant for small-scale web development team's knowledge base.
Picking a tool and setting up a knowledge base to have technical and design decisions, notes, and maintenance logs in a single place is usually straightforward. It's the consistently feeding the system with high-quality data part that we get stuck at.
Eventually, I have to input information when I lack the necessary mental bandwidth, I end up creating incomplete entries that lack essential details.
The initial idea is to develop a chat interface that allows us to share random pieces of information, either through text or voice, and have it scanned for actionable data that can be ingested. Then it should start probing the user, looking to expand and refine the new information, connecting it with existing data, and general maintenance of the repository. All with no active guidance.
For instance, I could launch it and say: “Today the team and I discussed how frequently the databases for project A and B should be synced. We established the user can tolerate up to 10 minutes of tardiness", at which point I would expect it to:
- Create a note of this discussion happening
- Check and fix the existence of entries for projects A and B
- Link the note to the projects
- Create database entries for both projects
- Look for previous entries that could be linked to the projects
- Ask for further explanation of how we came to that conclusion
I consume most of my video/audio informational content at 1.5x or faster. It wasn't FOMO or that I felt unproductive. I just felt that some of the content was going too slow for my thought process and leaving me kind of hanging, waiting for the next bit of information so I could continue coming up with my conclusions about the topic. But if I'm slower that day or feel that the topic is too dense, I don't mind at all slowing it down or even pausing it.
To me, it just feels like when I scan some text with the eyes looking for the juicier parts or hooks that I most care about. Speaking of which, I much prefer to consume content as text than audio or video. I can scan, seek, pause, think, go back at will without fiddling with any controls.
Assuming we're in the unscientific realm, and that scientificity exists on a spectrum, I'd say the parent comment is more scientific, because at least it provides a description of a logical causal chain (collagen is a protein, we digest proteins indiscriminately, therefore ...).
Provision is a script I've grown over years of building small production systems on Debian and Ubuntu Server. It handles security basics, relies on system tools so you're locked to any cloud solution, installs common software, and is _mostly_ safe to re-run if you mess anything up. If you have suggestions or battle scars with similar projects, I'd love to hear it. Cheers!
But you did spend time learning unix, nginx, bash, CGI, SQLite, etc.
Sure I can agree these tools are worth learning, I'm just pointing out it's not the same thing. Netlify allows you to setup a static website with data collection, e-mail dispatch and whatnot all without code and much faster.
I've been hosting small and mid-sized web sites and applications for years without a hitch using my own script, written from reading it up online. It might not work exactly as is for you but perhaps can offer some reference: https://github.com/corenzan/provision
Should whether an application is running or not be a concern to you? Why would you ever want to actually quit an application?
Granted, I understand that in reality the app might be misbehaving, hogging resources and whatnot and that the underlying OS isn't perfect and we should absolutely have control over how our hardware is utilized, but from a UX PoV I think the notion of a running process isn't of interest to me unless I am developing software.
If I close a window, I just want to get it out of my way. I only leave things running as a workaround to the possible slow cold start the program may present.
> and doesn't allow the developer using it a choice in those matters.
I never took "opinionated" software as "no configuration". For me it's always been "custom defaults". It might provide some configuration, it might not, but what makes it opinionated is the fact that the defaults are not generally applicable, but rather tailored for a specific use case.
I honestly thought I was the only one, how cool is that? I too find `test` preferable over `[` since it leaves no doubt we're calling a command just like any other and the same assumptions still apply. I also avoid `[[` since it isn't posix and won't work in simpler shells.
Hi. I've been using Notion for all my paperwork and often times I wanted a simple stopwatch for timing my activity. Unfortunately Notion doesn't currently have anything like that but it embeds other URLs quite nicely, and that's when it hit me; I could make something and embed it on my documents.
Fits well also for full-screen display, large or small.
Picking a tool and setting up a knowledge base to have technical and design decisions, notes, and maintenance logs in a single place is usually straightforward. It's the consistently feeding the system with high-quality data part that we get stuck at.
Eventually, I have to input information when I lack the necessary mental bandwidth, I end up creating incomplete entries that lack essential details.
The initial idea is to develop a chat interface that allows us to share random pieces of information, either through text or voice, and have it scanned for actionable data that can be ingested. Then it should start probing the user, looking to expand and refine the new information, connecting it with existing data, and general maintenance of the repository. All with no active guidance.
For instance, I could launch it and say: “Today the team and I discussed how frequently the databases for project A and B should be synced. We established the user can tolerate up to 10 minutes of tardiness", at which point I would expect it to:
- Create a note of this discussion happening - Check and fix the existence of entries for projects A and B - Link the note to the projects - Create database entries for both projects - Look for previous entries that could be linked to the projects - Ask for further explanation of how we came to that conclusion